Rivista Italiana di Agrometeorologia
Transcript
Rivista Italiana di Agrometeorologia
ISSN 1824-8705 Rivista Italiana di Agrometeorologia Rivista Italiana di Agrometeorologia anno 9 - n. 1 – Ottobre 2004 Periodico quadrimestrale edito dall’AIAM Autorizzazione Tribunale di Firenze n. 5221 del 4/12/2002 Poste Italiane s.p.a. - Spedizione in Abbonamento Postale 70% - DCB UDINE. Italian Journal of Agrometeorology • Aiam News - Notizie dalla Associazione 3 • A Jackknife-derived visual approach for sample size determination - Un approccio 9 grafico derivato dal jackknife per la determinazione delle dimensioni del campione. Confalonieri R. • Metodologie e algoritmi per il controllo di qualità di dati orari e giornalieri acquisiti 14 da una rete agrometeorologica: applicazioni alle rete lucana SAL - Methods and algorithms for evaluating the data quality at hourly and daily time scales for an agrometeorological network: application to the regional net of Basilicata (Italy). Rana G., Rinaldi M., Introna M. • GSRad, un componente software per la stima della radiazione solare - GSRad, a 24 software component to estimate solar radiation. Donatelli M., Carlini L., Bellocchi G. • I fabbisogni idrici del pomodoro da industria in Capitanata - Water use of 31 processing tomato in the Capitanata region (Southern Italy). Rinaldi M., Rana G. • Remotely sensed vegetation indices: theory and applications for crop management - 36 Indici di vegetazione telerilevati: teoria ed applicazioni per la gestione agronomica delle colture. Basso B., Cammarano D., De Vita P. Rivista Italiana di Agrometeorologia Italian Journal of Agrometeorology Anno 9 - n. 1 – Ottobre 2004 Rivista Italiana di Agrometeorologia Periodico quadrimestrale edito dall’AIAM www.agrometeorologia.it ISSN 1824-8705 EDITORE: AIAM Associazione Italiana di Agrometeorologia Presidente: Luigi Mariani Consiglieri: Maurizio Borin, Andrea Cicogna, Antonino Drago, Vittorio Marletto, Simone Orlandini, Miriam Rosini, Emanuele Scalcione. Revisori dei conti: Federico Spanna, Maria Carmen Beltrano, Luigi Pasotti Sede legale - via Caproni 8, 50144 Firenze. Sede tecnica - via Modigliani 4, 20144 Milano Direttore responsabile: Marco Gani - e-mail: [email protected] Responsabile scientifico (Editor in Chief) : Maurizio Borin, Dipartimento Di Agronomia Ambientale e Produzioni Vegetali, Viale dell’Università 16, 35020 Legnaro (Pd), Italy Tel. +39 049 8272838, e-mail: [email protected] Segretaria di redazione: Roberto Confalonieri Dipartimento di Produzione Vegetale, Università degli Studi di Milano, Via Celoria 2, 20133 Milano, Italy e-mail: [email protected] Redazione a cura di: Andrea Cicogna e-mail: [email protected] Abbonamenti La rivista è spedita gratuitamente ai soci Aiam. La quota associativa all’AIAM per il 2005 è fissata in € 50 per i soci singoli ed in € 300 per gli enti. I versamenti possono essere effettuati sul CC postale n. 43686203 intestato ad Associazione Italiana di Agrometeorologia. In alternativa il costo dell’abbonamento alla sola rivista è di € 60 da versare sul medesimo CC postale. Obiettivi La Rivista Italiana di Agrometeorologia si propone di pubblicare contributi scientifici originali, in lingua italiana ed in lingua inglese, riguardanti l'agrometerologia, intesa come scienza che studia le interazioni dei fattori meteorologici ed idrologici con l'ecosistema agricoloforestale e con l'agricoltura intesa nel suo senso più ampio e le differenti aree tematiche specifiche di interesse della disciplina, quali ecofisiologia delle piante erbacee e arboree, fenologia delle piante coltivate, fitopatologia, entomologia, fisica del terreno e idrologia, micrometeorologia, modellistica di simulazione, telerilevamento, pianificazione territoriale, sistemi informativi geografici e tecniche di spazializzazione, strumentazione di misura di grandezze fisiche e biologiche, tecniche di validazione di dati, agroclimatologia, divulgazione in agricoltura e servizi di supporto per gli operatori agricoli. La Rivista si avvale di un Comitato scientifico, che è il garante della qualità delle pubblicazioni e che per tale scopo può avvalersi di referee esterni. Aims The Rivista Italiana di Agrometeorologia publishes English or Italian-written original papers about agrometeorology, that is the science which studies the interactions between meteorological, hydrological factors and the agro-forest ecosystem and with agriculture, including all the related themes: herbaceous and arboreal species ecophysiology, crop phenology, phytopathology, entomology, soil physics and hydrology, micrometeorology, crop modelling, remote-sensing, landscape planning, geographical information system and spatialization techniques, instrumentation for physical and biological measurements, data validation techniques, agroclimatology, diffusion of information and support services for farmers. Submitted articles are reviewed by two independent members of the Editorial Board or by other appropriate referees. Comitato dei referee (Editorial Board) M. Acutis, Milano A. Alvino, Campobasso F. Benincasa, Sassari M. Bindi, Firenze S. Bocchi, Milano M. Borin, Padova A. Brunetti, Roma S. Dietrich, Roma M. Donatelli, Bologna G. Genovese, Ispra 2 G. Maracchi, Firenze V. Marletto, Bologna M. Menenti, Ercolano F. Morari, Padova S. Orlandini, Firenze A. Pitacco, Padova F. Rossi, Bologna P. Rossi Pisa, Bologna D. Spano, Sassari Aiam News Notizie dall’Associazione E’ NATA RIAm: AUGURI ! Marco Gani – Direttore responsabile [email protected] Luigi Mariani – Presidente Aiam [email protected] Non capita tutti i giorni di fondare una rivista scientifica, intervenendo così nella genesi di quello che da Galileo in avanti si è imposto come uno degli ingranaggi fondamentali della trasmissione del sapere scientifico. Anche per questo si deve festeggiare la nascita della Rivista Italiana di Agrometeorologia (RIAm), ultimo segno della vitalità di un’associazione di amici che a sette anni dalla fondazione mostra di non aver esaurito la propria spinta propulsiva. RIAm raccoglie l’eredità di Aiam News, testata sociale attiva dal 1998 al 2003, ed è il risultato di un progetto in discussione da circa un biennio e sui cui obiettivi lasciamo giustamente la parola al presidente del Comitato Scientifico prof. Maurizio Borin. Vogliamo solo rammentare che la rivista dovrà essere nei prossimi anni uno degli strumenti chiave per affermare l’agrometeorologia come disciplina ponte fra discipline fisiche e biologiche in una visione complessiva di tipo agro – ecosistemico. Il giusto equilibrio fra diverse esigen- ze ed aspettative sarà fondamentale per il futuro del progetto RIAm; pensiamo in particolare all’equilibrio fra articoli di ricerca ed articoli a taglio divulgativo e più legati all’esperienza operativa dei servizi agrometeorologici. E poiché tutte le cose viaggiano con le gambe degli uomini, vogliamo lanciare un appello perché sia la rivista scientifica vera e propria (quella per intenderci sotto la supervisione del Comitato Scientifico), sia il notiziario non sottoposto a referaggio siano debitamente riforniti da parte dei colleghi di articoli ed informazioni provenienti dal mondo dei servizi agrometeorologici e dal mondo della ricerca. E’ necessario a questo punto sviluppare qualche considerazione di tipo economico. La produzione di 250 copie di una rivista scientifica non è uno scherzo per i considerevoli costi che comporta. Per tale ragione il Consiglio Direttivo ha deciso di aprire la rivista alla pubblicità e di aumentare le quote sociali portandole a 50 Euro/anno a partire dal 2005. Speriamo che queste scelte siano condivise dai soci e consentano alla RIAm di prosperare in un quadro economico di stabilità. Infine un ringraziamento a tutti coloro che hanno animato il dibattito che ha preceduto la genesi di RIAm, al comitato scientifico ed al suo presidente Maurizio Borin, agli autori degli articoli ed infine a chi garantisce l’indispensabile attività di segreteria della rivista (Andrea Cicogna e Roberto Confalonieri). RIAm: PRIMO NUMERO Maurizio Borin - Responsabile scientifico [email protected] Rivista Italiana di Agrometeorologia nasce con molte ambizioni. La prima, ovvia, direi istituzionale, è quella di offrire un’occasione di incontro e di dialogo fra coloro che, a diverso titolo, sono attivi in questa affascinante disciplina: accademici, insegnanti e ricercatori, tecnici che lavorano nei servizi, imprese private operanti nella produzione di tecnologie e nell’ erogazione di servizi, studi professionali e consulenti, associazioni di agricoltori e singoli imprenditori agricoli, organizzazioni interessate alla gestione ambientale, studenti, appassionati di vario genere e quanti altri ancora rivolgano i loro interessi alle interazioni fra le vicende atmosferiche e il mondo biologico. La seconda grande aspirazione è di realizzare una Rivista veramente multidisciplinare, che sappia cogliere appieno lo spirito dell’agrometeorologia. Anche questo è un obiettivo ambizio- Preistoria dell’AIAM. Questa foto fu scattata in occasione del primo incontro preliminare alla costituzione dell’associazione, tenutosi a Milano presso l’Ersal nel febbraio 1996, su invito di Luigi Mariani. In piedi, da sinistra, si riconoscono: Albino Libè, Andrea Pitacco, Paolo Lega, Sandro Gentilini , Marco Gani, Simone Orlandini, ?, Marco Bindi, Ezio Bongioni, Massimo De Marziis, Loredana Albano, Paolo Parati, Umberto Gualteroni, Marina Anelli, Michele Gioletta, Luigi Mariani, Graziano Lazzaroni e Vittorio Marletto. A sedere, da sinistra, Fernando Antenucci, ?, Andrea Cicogna, Gaetano Zipoli, Susanna Lessi, Lucio Botarelli, Laura Brazzoli. 3 so, in quanto, in Italia, attualmente, non tutti i settori scientifici che potrebbero contribuire al progredire della materia sono egualmente partecipi ed attivi. Diviene quindi strategico il coinvolgimento di esperti e di aree tematiche che finora sono rimaste un po’ ai margini dell’agrometeorologia, accanto, ovviamente al rafforzamento dei contributi provenienti dai settori tradizionali. Altro traguardo cui ambisce RIAm è di risultare appetibile sia per ricercatori che per tecnici. Spesso, infatti, i primi sono poco interessati a pubblicare in riviste di carattere divulgativo, mentre i secondi trovano difficoltà ad apprezzare i contributi di maggior contenuto scientifico. L’impostazione data alla nostra Rivista mira a superare queste barriere, offrendo una sezione referizzata a chi vuol vedere validato il proprio lavoro da un Comitato Scientifico, ed una di carattere tecnico che dia spazio ai contributi di carattere prevalentemente applicativo che rappresentano comunque un patrimonio imprescindibile della nostra materia. Ancora: RIAm ospita lavori sia in italiano che in inglese e, in tal modo, si propone anche di richiamare contributi da oltre confine. Si tratta di una scelta impegnativa, a mio avviso di grande rilevanza strategica. La posizione geografica e politica del nostro Paese offre grandi potenzialità affinché l’agrometeorologia italiana divenga un punto di Il Sito dell’AIAM : www.agrometeorologia.it 4 riferimento sia per i Paesi del Bacino Mediterraneo sia per quelli dell’Est Europeo. In entrambi i casi si tratta di mondi in grande fermento, che potrebbero offrire anche notevoli opportunità di lavoro ed interscambio. Non va trascurato un altro fondamentale elemento che vorremmo caratterizzasse la Rivista: la velocità di pubblicazione. La dinamicità con cui le novità scientifiche e tecniche vengono conseguite, nonché le giuste aspirazioni di coloro che propongono i loro contributi, ci spingono ad impegnarci in un lavoro di revisione serio ma rapido, in modo da ridurre il più possibile il tempo che intercorre fra l’invio e la pubblicazione del lavoro. Proprio in funzione di questo, RIAm si è data un Comitato Scientifico di esperti che hanno aderito entusiasticamente alla proposta e agli obiettivi: a loro va il mio sentito grazie. Ci sarebbero ancora molte altre ambizioni da ricordare; fra tutte, la piccola “follia” di uscire in tempi in cui storiche riviste tecnico-scientifiche del settore agricolo sono state chiuse o versano in condizioni di grande difficoltà e in cui vi è un crescente sviluppo delle testate pubblicate in internet che competono con le tradizionali su carta. Anzi, a tal proposito, ci proponiamo di sviluppare sinergie fra RIAm e il sito AIAM (www.agrometeorologia.it). Questo primo numero nasce come esempio di ciò che si vorrebbe realizza- re: una consistente rassegna bibliografica su un tema di notevole attualità e di grande interesse scientifico, quattro articoli referizzati, svariati contributi tecnici e segnalazioni di iniziative dell’Associazione. Un numero vivo, che ospita già due lavori in inglese, nello spirito di aprirsi verso tutte le direzioni. Mi auguro che la proposta susciti interesse e che i numeri che seguiranno siano ancora più ricchi ed accattivanti. Rivista Italiana di Agrometeorologia (Italian Journal of Agrometeorology) is an international journal with the aim of publishing original articles and reviews on the inter-relationship between meteorology and the fields of plant, animal and soil sciences, environmental biophysics, ecology, and biogeochemistry. This journal has been thought to address many objectives. The first, obvious and institutional, is to offer the chance of meeting and conversing among all who are actively involved in this fascinating subject, each for different reasons: academic scientists, teachers and researchers, technicians who work for extension services, private firms involved in the production of technologies and supply of services, professional studies and consultants, farmers associations and individual farmers, organizations interested in environment management, students, all who are fond of the subject and all who are interested in the interactions between the atmospheric events and the biological world. The second big aim is to create a truly multidisciplinary journal, able to fully catch the spirit of agrometeorology. This is also an ambitious objective, because in Italy at this time not all scientific sectors which could contribute to the advancement of the subject are equally present and active. The involvement of experts and research areas which have been left at the margins of agrometeorology is then crucial, together with the reinforcement of the contributions coming from the traditional areas. Another goal of RIAm (IJAm) is to be appealing to both researchers and technicians. Researchers are often not very interested in publishing on extension journals, while technicians may find it difficult to appreciate the most interesting scientific contributions. Our journal is structured to overcome these barriers, offering a refereed section to all who want to see their work validated by a Scientific Committee, and a technical section where mostly applicative and extensions contributions - that cannot be set aside in our subject - are welcome. Moreover, RIAm (IJAm) welcomes papers both written in Italian and in English, to draw attention from abroad. It is a demanding choice, of great strategic importance in my opinion. The geographic and political position of our country offers a great potential for the Italian agrometeorology to become a reference point both for the Mediterranean countries and of the Eastern Europe countries. Both are worlds characterized by a great ferment, which might offer interesting work and exchange opportunities. Another fundamental element that will characterize the journal should not be neglected: the publishing velocity. The dynamicity with which scientific news and techniques are achieved, and the right aspirations of all who submit their contributions force ourselves to commit to a serious but quick revision work, to reduce as much as possible the time between the submission and the publication of the paper. To achieve this objective RIAm (IJAm) has selected a Scientific Committee of experts who have enthusiastically accepted the invitation and the objectives: I am deeply thankful to all of them. There would still be many other goals to mention; among all the little “insanity” to appear at a time when historical technical-scientific journals of the agricultural area have been closed or are going through a critical period and there is a great development of journals published on the web, which compete with the traditional paper journals. In this regard, we intend to develop original synergies between RIAm and the website of the Associazione Italiana di Agrometeorologia (www.agrometeorologia.it). This first number is an example of what we would like to carry out: a substantial bibliographic review on a subject of actual and great scientific interest, four refereed papers, several technical contributions and the men- tion of all the activities of the Association. A live number, with two papers in English, in the spirit of being open to all directions. I wish this proposal may be of interest and that the next numbers will be as rich and appealing. CONVEGNO AIAM 2004 Emanuele Scalcione [email protected] La pressione antropica e le esigenze legate allo sviluppo economico sottopongono il territorio e le risorse naturali ad una pressione e che spesso supera il limite di sostenibilità dando luogo a grandi problemi ambientali che stanno assumendo il carattere della “globalità” e che rappresentano, nelle società attuali, gli elementi critici per lo sviluppo futuro. Tra questi grandi problemi assumono rilevanza straordinaria quelli del cambiamento climatico, della desertificazione e della biodiversità; si tratta di temi solo apparentemente separati, trattandosi, di fatto, di problemi che richiamano uno stesso argomento di fondo: lo sviluppo sostenibile delle popolazioni del pianeta. E’ in questo contesto che si è inserito l’appuntamento annuale dell’AIAM a Matera. Ovviamente, lo scenario dei Sassi e le bellissime giornate primaverili hanno fatto da splendida cornice alle due giornate di lavoro, risultate assai ricche di contenuto tecnico e scientifico e che hanno visto la partecipazione di oltre un centinaio di persone, per buona parte provenienti da fuori regione. Senza dubbio, il tema dell’incontro “Gli agroecosistemi nel cambiamento climatico” ha rappresentato un forte richiamo per la Comunità scientifica e per i Servizi regionali, in virtù del contributo che l’agrometeorologia può dare all’agricoltura su un tema di così forte attualità. Le giornate di lavoro hanno dato una conferma del crescente interesse che l’AIAM ha creato intorno a sé, rafforzando il progetto di dar vita, nel prossimo autunno, alla Rivista Italiana di Agrometeorologia. La stessa partecipazione di autorità regionali, fra gli altri del Presidente della Giunta Regionale ed il Sindaco di Matera oltre ad amministratori dell’ALSIA, ha mostrato l’interesse che la comunità della Basilicata nutre nei confronti dell’agrometeorologia e delle applicazioni di questa disciplina. A conclusione dei lavori è emerso che l’agrometeorologia può dare un contributo di assoluto rilievo allo studio dei fenomeni in atto con evidenti riflessi Un momento significativo del convegno: la signora Gabriella Conte ricorda la figura del marito Michele a cui quest’anno è stato dedicato il premio Aiam per tesi di Laurea in Agrometeorologia. A fianco della signora il Presidente della Giunta della regione Basilicata Filippo Bubbico, il Presidente dell’AIAM Luigi Mariani e il Direttore dell'ALSIA Michele Taranto 5 extra agricoli, visto che l’agricoltura sta assumendo sempre più un ruolo di gestione e di conservazione del territorio e delle risorse naturali. Infine, un particolare ringraziamento a tutti coloro che hanno partecipato al convegno di Matera e quanti hanno collaborato per la buona riuscita dell’evento. IL BOLLETTINO DEL CENTRO EUROPEO FA 100 Vittorio Merletto [email protected] Ogni trimestre, da 25 anni, il Centro Meteorologico Europeo di Reading, UK, pubblica il bollettino ECMWF Newsletter. Come si legge nel numero 100, pubblicato la scorsa primavera, il bollettino è nato nel 1980, quando il Centro entrò in piena operatività, cinque anni dopo la sua fondazione. Si tratta di un agile fascicolo a colori, ottenibile gratuitamente scrivendo un messaggio all’editor Peter White ([email protected]). Il numero 100 contiene tre articoli tecnico-divulgativi e numerose notizie. L’articolo più rilevante, a cura di Frederic Vitart, è intitolato “Previsioni mensili” e descrive il sistema sperimentale di previsione numerica a dieci-trenta giorni realizzato al Centro europeo per colmare la lacuna esistente tra il sistema di previsione a dieci giorni e quello delle previsioni stagionali, entrambi operativi presso il Centro. Le previsioni mensili vengono eseguite due volte al mese e girano per 32 giorni, con 51 membri, di cui uno di controllo. Si tratta quindi di previsioni stocastiche, eseguite con la tecnica dell’ensemble, in cui il modello accoppiato oceano-atmosfera, viene avviato perturbando opportunamente le condizioni iniziali. La parte atmosferica consiste nel modello dell’ ECMWF con risoluzione orizzontale di circa un grado e 40 livelli verticali, mentre la parte oceanica è costituita dal modello HOPE (Hamburg Ocean Primitive Equation) sviluppato in Germania. L’interfaccia tra oceano e atmosfera è gestita dal sistema OASIS (Ocean, Atmosphere, Sea-Ice, Soil) di origine francese. Come scrive suggestivamente l’autore: “Ogni ora i flussi atmosferici di quantità di moto, calore e acqua dolce vengono trasferiti 6 all’oceano e, in cambio, all’atmosfera viene passata la temperatura superficiale dell’oceano”. Questa frequenza di accoppiamento è ritenuta sufficiente per seguire lo sviluppo di alcuni sistemi a scala sinottica, come i cicloni tropicali. Mentre i dati atmosferici necessari ad inizializzare la previsione (le cosiddette “analisi”) sono prontamente disponibili presso il Centro, i dati oceanici arrivino con ben dodici giorni di ritardo, di conseguenza è stato necessario costruire un sistema per “prevedere” le condizioni iniziali oceaniche, integrando il modello del mare a partire dall’ultima analisi disponibile, “forzandolo” con dati dalle analisi atmosferiche disponibili. L’autore descrive anche la tecnica utilizzata per contenere la deriva del modello, che diventa rilevante a partire dalla seconda settimana di integrazione. Il sistema di previsione mensile funziona regolarmente dalla primavera del 2002 e le sue prestazioni sono interessanti, almeno fino alla seconda/terza settimana. Per esempio l’eccezionale anomalia termica dell’estate 2003 sull’ Europa centrale, in particolare i 10 gradi oltre la norma registrati sulla Francia tra il 3 e il 9 agosto, risultava prevista, anche se meno intensamente di quanto poi registrato, almeno da alcuni membri dell’ ensemble, già venti giorni prima dell’evento stesso. Gli altri articoli si occupano degli errori sistematici nel modello del Centro Europeo e della previsione di onde anomale. Auguri al bollettino ECMWF per i prossimi 100 numeri. EMMANUEL LE ROY LADURIE – HISTOIRE HUMAINE ET COMPAREE DU CLIMAT. CANICULES ET GLACIERS, XIIIE-XVIIIE SIECLE (Fayard, pagine 740, 25 euro) L.Mariani [email protected] Con questo suo libro, che vuole essere il primo tomo di una più ampia storia umana e comparata del clima, Emmanuel Le Roy Ladurie attira ancora una volta l’attenzione del vasto pubblico sullo stretto legame esistente fra evoluzione del clima e vita delle società umane. Professore al Collège de France, amministratore generale della Biblioteca Nazionale (1987-1994), membro dell’Académie des sciences morales et politiques, Emmanuel Le Roy Ladurie è uno fra i maggiori storici francesi viventi ed è stato fra i primi a parlare di “storia del clima” con il suo fondamentale testo del 1967 intitolato Histoire du climat depuis l'an mil, che ha avuto una vasta notorietà in Italia nella traduzione pubblicata da Einaudi con il titolo Tempo di festa, tempo di carestia. Storia del clima dall'anno Mille. “Je m’en fous comme de l’an quarante” (me ne frego come dell’anno quaranta) è un’espressione idiomatica francese che trae con ogni probabilità origine dall’inverno 1739-1740. Fu quello un inverno, racconta Leroy Ladurie, privo di neve e marcato da tre mesi di gelate ininterrotte. All’inverno seguirono poi una primavera e un’estate assai piovose che rovinarono il raccolto dei cereali dando anche luogo a funeste inondazioni in gran parte d’Europa. La stessa vendemmia fu disastrosa, il che è un fatto assai meno trascurabile che oggi, poiché allora il vino era un genere di prima necessità per un’umanità oppressa dal lavoro manuale. Una vera catastrofe dunque, che in Francia si tradusse in circa 200.000 morti e che bloccò la crescita demografica francese per almeno un decennio. Duecentomila morti sono tanti ma senza dubbio assai meno dei 600.000 morti della grande carestia del 1709 e del milione di morti provocato dalla grande carestia del 1693, merito delle misure di mitigazione basate sull’approvvigionamento delle zone più colpite con cereali provenienti da zone dell’est e dell’ovest della Francia, meno interessate dal maltempo. Anche il caldo fu responsabile in passato di molte vittime: ad esempio le due estati canicolari del 1718 e 1719 fecero in Francia un totale di 450.000 morti, uccisi dalla disidratazione e dalla dissenteria. E che dire poi del legame fra i roghi delle streghe e il cattivo tempo? In Europa il picco nell’uccisione di streghe si ha proprio nel periodo compreso fra 1540 e 1600, proprio nel culmine della piccola era glaciale. I roghi interessarono varie parti d’Europa fra cui Germania, Borgogna, Inghilterra e Svizzera. Cosa vi è infatti di più satanico del far cader dal cielo del ghiaccio in piena estate? E così i roghi per “crimini gelivi” diventavano rimedi contro grandine o nevicate fuori stagione. E qui potremmo dire che le nostre “madonne della neve” hanno donne della neve” hanno quanto meno avuto il merito di volgere in positivo fatti che erano tutt’altro che apprezzati da popolazioni sempre sull’orlo della catastrofe alimentare. E’ evidente, secondo Leroy Ladurie, che fame ed epidemie derivanti da cause climatiche sono una chiave di interpretazione per molte vicende storiche del passato. Occorre tuttavia procedere ad un equilibrato dosaggio delle cause. Se infatti nel XVI secolo le guerre di religione non hanno in alcun modo il clima fra le loro cause primarie, la Lega cattolica non esitò ad imputare ai politici i cattivi raccolti frutto di cause climatiche. Si può invece ritrovare una causa meteorologica primaria per la Fronda, fenomeno tutto francese che ebbe origine dai dissesti finanziari seguiti alla guerra dei trent’anni ma che trasse forza dai magri raccolti delle fredde primavere del triennio 1648-1650. Da non scordare infine il fatto curioso per cui il regno del Re Sole fu paradossalmente marcato da un minimo d’attività solare, il cosiddetto minimo di Mounder, che secondo diversi studiosi è all’origine della piccola era glaciale. Approfondimenti Intervista a Leroy Ladurie di Fabien Gruhier dal titolo Du climat et des hommes, pubblicata sul Nouvel Observateur dell’8-14 luglio 2004 (http://www.nouvelobs.com/articles/p2070/a245 630.html) ULISSE, NELLA RETE DELLA SCIENZA Una divulgazione scientifica corretta ed onesta è merce rara e più che mai necessaria per affermare le idee di progresso scientifico ed umano. Per tale ragione è utile segnalare che è da tempo attiva su Internet una testa elettronica con scopi di divulgazione scientifica e tecnologica; il suo nome è ULISSE, NELLA RETE DELLA SCIENZA ed è gestita dalla Scuola Internazionale Superiore di Studi Avanzati (SISSA) di Trieste. Ulisse pubblica un proprio notiziario settimanale con approfondimenti su temi di rilevanza scientifica ed inoltre risponde alle domande più disparate che provengono da un pubblico affezionato. Le risposte sono affidate a esperti nelle più svariate discipline, inseriti in questo contesto di rete e che operano a titolo volontario. L’esperto riceve la domanda e nel caso abbia tempo e cognizioni sufficienti per rispondere provvede in tal senso; in caso contrario passa la palla a qualcun altro. Fra gli esperti “caduti nella rete” e che hanno finora contribuito all’attività di Ulisse con risposte in ambito meteo climatico ed agrometeorologico ricordiamo fra gli altri Andrea Cicogna, Marco Gani, Luigi Mariani e Fulvio Stel. Ulisse è consultabile all’indirizzo http://ulisse.sissa.it/ Agrometeorologia, risorse naturali e sistemi di gestione del territorio 8° convegno annuale dell’AIAM primo annuncio L’edizione 2005 del Convegno Nazionale dell’Associazione Italiana di Agrometeorologia, dal titolo “Agrometeorologia, risorse naturali e sistemi di gestione del territorio” si svolgerà in Abruzzo; l’iniziativa rientra nel programma di Assistenza tecnica dell’ARSSA. La manifestazione si svolgerà nei giorni 3, 4 e 5 Maggio 2005, e sarà articolata nelle seguenti tre sessioni: · Riduzione dell’impatto ambientale · Sistemi informativi territoriali · Aree protette Il Convegno si svolgerà a Vasto nei primi due giorni (3-4 maggio), mentre l’ultima giornata (5 maggio), dedicata alle aree protette, avrà luogo a S. Eufemia a Maiella, presso la sala Convegni dell’Orto Botanico del Parco Nazionale della Maiella. L’organizzazione a livello locale sarà curata dai colleghi Ferdinando Antenucci e Bruno Di Lena. Per maggiori informazioni è possibile consultare il sito dell’aiam (www.agrometeorologia.it) ovvero contattare: Bruno Di Lena ([email protected]) o Andrea Cicogna ([email protected]) 7 R. Confalonieri - Rivista Italiana di Agrometeorologia 9-13 (1) 2004 Pubblicità sulla Rivista Italiana di Agrometeorologia In questa Rivista può trovare spazio pubblicità relativa al settore Agrometoerologico. Condizioni economiche: - pagina intera 4° di copertina 500 € - pagina intera – interna 200 € - modulo minimo: mezza pagina. - sconto per soci sostenitori dell'AIAM: 10%. Poichè la ragione sociale dell'Associazione impedisce attività commerciale i fondi saranno introitati a titolo di contributo e saranno finalizzati alla copertura delle spese di pubblicazione della rivista. Si precisa inoltre che il contributo non è deducibile dalla dichiarazione IRPEF in quanto AIAM non è una ONLUS. Le richieste devono essere inviate al presidente dell'Aiam dott. Luigi Mariani (all'indirizzo [email protected] o al numero telefonico 329/7027077) che è disponibile per ogni ulteriore dettaglio. Si precisa infine che le richieste che perverranno saranno soddisfatte in ordine di priorità di ricevimento. 8 R. Confalonieri - Rivista Italiana di Agrometeorologia 9-13 (1) 2004 A JACKKNIFE-DERIVED VISUAL APPROACH FOR SAMPLE SIZE DETERMINATION UN APPROCCIO GRAFICO DERIVATO DAL JACKKNIFE PER LA DETERMINAZIONE DELLE DIMENSIONI DEL CAMPIONE Roberto Confalonieri Department of Crop Science, Section of Agronomy, University of Milano, Via Celoria 2, 20133 Milano, Italy. Abstract The determination of the sample size is one of the most important topics in field research and preliminary sampling sessions should be carried out to determine the size of the sample before the collection of real data. A new method for sample size determination is proposed, based on a jackknife visual evolution. The method has been tested to determine (i) the sample size in a rice field in order to collect aboveground biomass data and (ii) the minimum number of time domain reflectometry (TDR) data acquisitions necessary to get a reliable estimation of soil water content in a maize field. The 18-plants sample size determined for rice aboveground biomass with the proposed method is coherent with the 20-plants samples traditionally harvested for field experiments with rice; while the 8 acquisitions indicated for soil water content take into account both the effort required for collecting data and the measurement accuracy. The method does not care if the sample units are normally distributed or not and takes into account both the variation of accuracy for increasing sample sizes and, indirectly, the effort required for getting the measurements. Keywords: Monte Carlo, rice, Oryza sativa L., resampling, statistical inference, number of plants, aboveground biomass, soil water content, TDR Riassunto La determinazione delle dimensioni del campione è uno degli aspetti più importanti nella ricerca di campo: sessioni di campionamento preliminari mirate alla sua valutazione dovrebbero sempre essere condotte prima della raccolta dei dati veri e propri. In questo lavoro viene proposto un nuovo metodo per la determinazione delle dimensioni del campione, basato su un evoluzione grafica del jackknife. Il metodo è stato testato (i) per la raccolta di dati di biomassa per la coltura del riso e (ii) per la determinazione del numero minimo di letture al TDR (time domain reflectometer) necessarie per una stima affidabile del contenuto idrico del terreno in un campo di mais. La dimensione del campione di 18 piante indicata dal metodo proposto è coerente con il campione di 20 piante tradizionalmente raccomandato per esperimenti di campo in risaia. Le 8 acquisizioni al TDR indicate per il contenuto idrico del terreno tengono in considerazione sia lo sforzo necessario per la raccolta del dato che l’affidabilità della misura. Il metodo funziona sia per campioni distribuiti normalmente che non e, indirettamente, tiene conto dello sforzo necessario per ottenere il dato sperimentale. Parole chiave: Monte Carlo, riso, Oryza sativa L., ricampionamento, inferenza statistica, numero di piante, biomassa, contenuto idrico del terreno, TDR Introduction The determination of the sample size is one of the most important topics for the collection of reliable measurements of variables describing plant growth (e.g. aboveground biomass (AGB), leaf area index) and, in general, for all the aspects of field research. In many cases, the classical inferential methods based on the t distribution for determining the sample size cannot be used: the characteristics of the population (µ and σ) are unknown. They varies according to many factors (e.g. homogeneity of the incorporation of residues, of Corresponding author. tel. +39 02 5031 6592 fax +39 02 5031 6575 e-mail address: [email protected] (R.Confalonieri). sowing, of germination, of emergence, etc.). For these reasons, it is impossible to have a presumptive knowledge of the sampling error that can be accepted for the calculation of the sample size: it could change from an experimental field (or situation) to another. For this reason, it is better to analyze the trend of the relative variability inside the same field when the sample size increases rather than assuming absolute criterions for the acceptance of a specific variability for the determination of the sample size. In this way, the effort required for processing a sample can be easily taken into account too. Moreover, a reliable evaluation of the fact that the population is normally distributed is often not possible because, above all for time-consuming or costly methods, only R. Confalonieri - Rivista Italiana di Agrometeorologia 9-13 (1) 2004 few observations can be determined in the preliminary study carried out for determining the sample size. It is possible to find in the Literature examples of alternative methods for sample size determination. Wolkowski et al. (1988), in an experiment aiming at comparing field plot techniques for corn grain and dry matter yield estimation, determined the sample size by harvesting all the plants belonging to a row and determining AGB on 1, 5, 10, 15, 20, 25, all plants of the row randomly extracted from the whole harvested plants and calculating the coefficient of variation [CV = standard deviation (average value)-1] of each sample. They found that the coefficient of variation minimizes for 10 and 15 – plants samples. The samples extracted from the same row can not be considered completely independent. Moreover, the Authors don’t specify if the plants belonging to a sample had been used to determine the AGB of other samples. It’s possible to find in the Literature studies for rice sample size (= number of plants) determination, although they were not directly related to AGB determination. For example, Tirol Padre et al. (1988), in an experiment about acetylene reduction activity in different rice varieties, noticed that the average CV (computed from the different varieties CV, obtained from many measurement replicates) decreased from 33% with 1 plants to 11% for 6 plants. The Authors discussed that, for some samples, the variances were correlated to the means (nothomoscedastic) and for others not (homoscedastic). For these reason, probably the CV is not the better parameter for sample size determination: it should be more correct the separated analysis of means and standard deviations. A different approach, based on the different components of the effort required for the whole process to determine an experimental value, was proposed by Yonezawa (1985) to determine the minimum number of plants required for the conservation of plant genetic resources. The Author concluded that (i) a sample size as small as 10 plants per site or population is reasonable to cover a large target area and (ii) it is more important to analyze a wide number of site or population. This conclusion could be extendable to an experimental field by considering, for example, the sites corresponding to the plots per treatment. In this way, it’s possible to conclude that, for the same available effort, it’s better to harvest few plants per plot but dispose of more plots for each treatment. From the 60s resampling methods are increasingly diffused in biological and environmental sciences. They conceptually derive from the “Monte Carlo” methods, for the first time applied to a physical process by Barker (1965) but they are based on the repeated use of the data from the same sample (the only one sample which has been collected). Efron and Tibshirani (1991) underline the enormous potentiality of these techniques, which widely use the calculation capability of computers, for the research on inferential statistics. The bootstrap (Efron, 1979; Efron and Tibshirani, 1993) and the jackknife (Quenouille, 1949; Tukey, 1958) are particularly important, both because they are increasingly used and because of the number of studies and developments they generated. 10 Therefore, the objectives of this study are (i) the development of an alternative method for sample size determination based on an evolution of the jackknife; (ii) its evaluation for determining the sample size for aboveground biomass in a rice field and for soil water content in a maize field. In the first case the sample size will be determined as number of plants and, in the second, as number of TDR (Time Domain Reflectometer) measurements. Materials and methods Experimental data Experimental data were collected in 2 experiments. The first was carried out in Besate (northern Italy, latitude 45° 18’ N, longitude 8° 58’ E) during 2003. Rice (Oryza sativa L; cv. Volano) was sown in April 28 and grown under flooded conditions. No water stresses were observed during the crop cycle and the management has allowed to prevent the presence of weed and pests. During this experiment, the studied variable was aboveground biomass. The samples were collected when the plants were young (3 leaves) and the dry weight of each plant was very low. For this reason, we have chosen to use a group of 3 plants as sampling unit instead of a single plant in order to avoid errors due to the measurement of very small quantities of biomass. The second experiment was carried out in Lodi Vecchio (northern Italy, latitude 45° 19’ N, longitude 9° 26’ E) during 2004. Maize (Zea mais L) was grown under optimal water and nutrients availability. The soil is a Ultic Haplustalfs fine silty, mixed, mesic (Soil Survey Staff, 1999), subacid, with medium organic matter content, sufficient available phosphorous and medium potassium content. The studied variable was soil water content. In this case, the sampling unit was a single TDR (Time Domain Reflectometer) measurement event. A visual jackknife evolution for determining the sample size The proposed method can be considered a modification of the jackknife, developed by Tukey (1958) at the end of the 50s on the basis of an idea of Quenouille (1949; 1956) of some years before and reviewed, with its recent developments, by Hinkley (1983). According to this technique, the original sample of N elements is divided into groups of k elements. If N is N! low, k may be equal to 1. virtual samples (N − k )!k! (combinations without repetitions) of (N-k) elements are N! generated by eliminating times k different val- (N − k )!k! ues from the original sample. Mean, standard deviation, etc. can be calculated for all the generated virtual samples. In the presented method, different values of k are used. N! The process of generation of the virtual sam(N − k )!k! ples is repeated (N-1) times with k assuming values from R. Confalonieri - Rivista Italiana di Agrometeorologia 9-13 (1) 2004 1 to (N-2), for a total of N −2 N! ∑ (N − k )!k! different virtual k =1 samples. Mean and standard deviation are computed for all the generated samples and plotted on two charts, with the values of (N-k) on the X-axis (Figures 1 and 2) and the means (or standard deviations) on the Y-axis, in order to get a visual representation of how the means and the standard deviations of the generated samples vary with increasing sample size. Analyzing the trends of the means and the standard deviations for increasing values of (N-k), the sample size is considered equal to the (N-k) value for which the variability among the means and the standard deviations computed for almost all the generated samples changes its decreasing rate, assuming a slow – asymptotic trend for increasing (N-k) values. This change of rate can be determined by the observation of how the variability among means and standard deviations evolves for increasing (N-k) values or by using an automatic algorithm. In the second case, the following procedure is adopted: 1. select a (N-k) value (N-k)’ higher than 2 and lower than N-2; 2. for i from 2 to (N-k)’, calculate mean and standard deviation of the 5% highest means (respectively Yi and si); 3. compute a linear regression weighted on the standard deviations assuming as independent variable the (N-k) values and as dependent one the Yi values and computing the R2 of the regression (R2_1); 4. repeat the steps [2] and [3] for the 5% lowest means, determining R2_2; 5. for i from (N-k)’ to N-1, repeat the steps from [2] to [4] to compute R2_3 and R2_4, imposing that the slope of the regression is 0; 6. calculate SR2 by adding R2_1, R2_2, R2_3 and R2_4; 7. repeat the steps from [2] to [6] for each of the (N-k)’ values higher than 2 and lower than N-2; 8. the automatically determined sample size based on the means (AM) is the (N-k)’ value corresponding to the lower SR2. The same procedure is used to determine the sample size basing on the standard deviations (ASD). The highest between AM and ASD is the automatic sample size. Two important features of the presented method for the sample size determination are: (i) the initial sample may be constituted both by independent and not-independent data and (ii) the data may be both normally or notnormally distributed. Moreover, it supplies a visual, easy-to-analyze representation of a particular sample size directly related to its variability and to the variability of bigger and smaller samples sizes. In this way, it is easier to consider the problem of sample size determination both from the statistical point of view and from the point of view of the available effort for carrying out the experimentation. Application of the proposed method for determining the sample size For the first experiment, AGB of 12 3-plants samples (36 plants; aggregated sample of 3 sub-samples) was measured. 250 dispositions without repetitions were generated through a randomization process using the 12 samples. Only 250 dispositions were randomized because, when the initial sample is not numerous, an increase of the resamplings number does not correspond to an effective improvement of the estimation (Manly, 1991). The obtained values were collected in a matrix (Matrix O; 12 rows; 250 columns) with the dispositions in the columns and the values of each disposition in the rows. A new matrix was created (Matrix M; 11 rows; 250 columns) and, for each of its columns, the mean value computed on the first 2 elements of correspondent column of the Matrix O was stored in the first row; the mean value computed on the first 3 elements in the second rows and so on. The same procedure used to create the Matrix M was followed to create a Matrix SD, containing the standard deviations of the values stored in the Matrix O. Therefore, the elements of Matrix M and Matrix SD are computed by using the following equations. n MatrixM nj = ∑ MatrixO ij i =1 n 2 n ⎛ ⎞ MatrixOij ⎟ ⎜ ∑ n ⎜ MatrixO − i =1 ⎟ ∑ ij ⎜ ⎟ n i =1 ⎜ ⎟ ⎝ ⎠ MatrixSDnj = n where j indicates the dispositions and n is equal to (N-k). The elements of Matrix M and Matrix SD were plotted separately with the 250 series corresponding to the columns of the two matrixes (Figure 1 and 2). The same procedure was followed for determining the sample size (number of TDR measurement events) for soil water content determination (second experiment). The only two differences are that the sampling unit was a single TDR data acquisition instead of an aggregated sample of 3 plants and that N was, in this case, 18 instead of 12 (Figure 3 and 4). Jack: a software for sample size determination using the visual jackknife A software was created for the application of the proposed method (Figure 5). It generates the virtual samples and the matrixes of the means and of the standard deviations. It produces the relative diagrams, draws the 4 regression lines used by the automatic method for sample size determination and compute the CV obtainable with the automatic method. It gives to the user the possibility of choosing different sample sizes allowing him to compare the CVs obtainable with the two methods (automatic or not). The software is free downloadable at the web site: http://users.unimi.it/agroecol/confalonieri.php. 11 R. Confalonieri - Rivista Italiana di Agrometeorologia 9-13 (1) 2004 12.80 25.00 12.20 23.80 11.60 22.60 11.00 21.40 10.40 20.20 9.80 19.00 2 3 4 5 6 7 8 9 10 11 2 3 4 5 6 7 8 Sample size 9 10 11 12 13 14 15 16 Fig. 1 – Aboveground biomass (rice) - Means of the generated populations when the jackknife is applied for different k values. (N-k) values on the X-axis, with k from (N-2) to 1. More details in the text. k is the number of observations not used by the jackknife; N is the total number of observations Fig. 1 - Biomassa aerea (riso) - Medie delle popolazioni generate applicando il jackknife per differenti valori di k. Sull’asse delle X i valori (N-k), con k da (N-2) a 1. Maggiori dettagli nel testo. k è il numero di osservazioni non utilizzate nel jackknife; N è il numero totale di osservazioni Fig. 3 – Soil water content - Means of the generated populations when the jackknife is applied for different k values. (N-k) values on the X-axis, with k from (N-2) to 1. More details in the text. k is the number of observations not used by the jackknife; N is the total number of observations. Fig. 3 - Contenuto idrico del terreno - Medie delle popolazioni generate applicando il jackknife per differenti valori di k. Sull’asse delle X i valori (N-k), con k da (N-2) a 1. Maggiori dettagli nel testo. k è il numero di osservazioni non utilizzate nel jackknife; N è il numero totale di osservazioni 2.60 4.50 2.08 3.60 1.56 2.70 1.04 1.80 0.52 0.90 0.00 0.00 2 3 4 5 6 7 8 9 10 11 Sample size (3-plants groups) Fig. 2 – Aboveground biomass (rice) - Standard deviations of the generated populations when the jackknife is applied for different k values. (N-k) values on the X-axis, with k from (N-2) to 1. More details in the text. k is the number of observations not used by the jackknife; N is the total number of observations Fig. 2 - Biomassa aerea (riso) - Deviazioni standard delle popolazioni generate applicando il jackknife per differenti valori di k. Sull’asse delle X i valori (N-k), con k da (N-2) a 1. Maggiori dettagli nel testo. k è il numero di osservazioni non utilizzate nel jackknife; N è il numero totale di osservazioni 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Sample size Fig. 4 – Soil water content - Standard deviations of the generated populations when the jackknife is applied for different k values. (N-k) values on the X-axis, with k from (N2) to 1. More details in the text. k is the number of observations not used by the jackknife; N is the total number of observations. Fig. 4 - Contenuto idrico del terreno - Deviazioni standard delle popolazioni generate applicando il jackknife per differenti valori di k. Sull’asse delle X i valori (N-k), con k da (N-2) a 1. Maggiori dettagli nel testo. k è il numero di osservazioni non utilizzate nel jackknife; N è il numero totale di osservazioni Fig. 5 – Jack: a software for the visual-jackknife application (http://users.unimi.it/agroecol/confalonieri.php). User’s interface Fig 5 - Jack: un software per l’applicazione del visual-jackknife (http://users.unimi.it/agroecol/confalonieri.php). Interfaccia utente 12 17 Sample size 17 R. Confalonieri - Rivista Italiana di Agrometeorologia 9-13 (1) 2004 Results and discussion Aboveground biomass Figure 1 shows that the differences between the means of the generated populations, obtained with the proposed method, decrease for sample size equal to (6×3) plants. These differences continue to slightly decrease with higher sample sizes but 6 groups of three plants can be considered a satisfactory compromise between the effort to get the measure and its reliability. In fact, although the differences between the average values decrease also for sample sizes higher than the indicated one, there is a clear change in the height of the region of the plot engaged by the 250 series. The same considerations can be discussed for Figure 2 (standard deviations of the generated populations). It’s possible to notice that the height of the region of the plot engaged by the 250 series is almost constant for sample sizes higher than 6 group of 3 plants, while higher difference in standard deviations can be observed considering less than 18 plants. In practice, a higher number of plants would increase the effort required by the measurement without a comparable increase in the accuracy. In this example, the sample size was determined by graphically analyzing the two diagrams. This “visual” procedure allows to determine sample sizes which take into account the effort required to get data and the experimenter’s judgement. Soil water content For the determination of the number of TDR measurement events required to get a reliable soil water content estimation, the automatic procedure was used (see the section “Materials and methods”). Figure 3 and 4 show, respectively, the means and the standard deviations of the generated populations. The grey straight lines represent the linear regressions on which the R2 are computed. The intersection between the oblique lines and the horizontal ones represents the automatically computed samples size, also indicated by the vertical series of dots. It is possible to notice that the automatically computed sample size is 7 in Figure 3 (means) and 8 in Figure 4 (standard deviations). In this kind of cases, the proposed algorithm advise to chose the highest. The used algorithm has been planned to reproduce the behavior of an experimenter in selecting a sample size and it is particularly useful when the area of the diagrams engaged by the generated series decreases, for increasing sample sizes, too regularly to allow the experimenter to confidently notice discontinuities. Conclusions When parametric modelling and theoretical analysis are difficult, the bootstrap (Efron, 1979; Efron and Tibshira- ni, 1993) and the jackknife (Quenouille, 1949; Tukey, 1958) are good alternatives for analyzing the characteristics of a population (Park and Willemain, 1999). The proposed method, based on a visual evolution of the jackknife has shown to be reliable for sample size determination, determining sample sizes of (i) 18 plants (6 aggregated samples of 3 plants) for rice aboveground biomass determination and (ii) 8 TDR measurement events for soil water content estimation in a maize field. The first value is coherent with the 20 plants per plot recommended to be harvested by Gomez (1972) in his manual about field experiments with rice and traditionally used in field experiments. Future studies will evaluate the applicability of the proposed method to other variables (e.g. leaf area index, soil nitrogen content, etc). References Barker, A.A., 1965. Monte Carlo calculations of the radial distribution functions for a proton-electron plasma. Australian Journal of Physics, 18, 119-133. Efron, B., 1979. Bootstrap methods: another look at the jackknife. Annals of statistics, 7, 1-26. Efron, B., Tibshirani, R., 1991. Statistical data analysis in the computer age. Science, 253, 390-395. Efron, B., Tibshirani, R., 1993. An introduction to the bootstrap. Chapman & Hall, New York. Gomez, K.A., 1972. Techniques for field experiments with rice: layout, sampling, sources of error. International Rice Research Institute, Los Baños, Philippines, pp. 46. Hinkley, D.V., 1983. Jackknife methods. Encyclopedia of Statistical Science, 4, 280-287. Manly, B.F.J., 1991. Randomization and Monte Carlo Methods in Biology. Chapman and Hall, London, New York, USA, pp. 281. Park, D., Willemain, T.R., 1999. The threshold bootstrap and threshold jackknife. Computational Statistics & Data Analysis, 31, 187-202. Quenouille, M.H., 1949. Approximate tests of correlation in time series. Journal of The Royal Statistical Society, Series B, 11, 68-84. Quenouille, M.H., 1956. Notes on bias in estimation. Biometrika, 43, 353-360. Soil Survey Staff, 1999. Soil Taxonomy, 2nd edition. Agric. Handbook n. 436, USDA-NRCS, 869 pp. Tirol Padre, A., Ladha, J.K., Punzalan, G.C., Watanabe, I., 1988. A plant sampling procedure for acetylene reduction assay to detect rice varietal differences in ability to stimulate N2 fixation. Soil Biology and Biochemistry, 20, 175-183. Tukey, J.W., 1958. Bias and confidence in not quite large samples (Abstract). Annals of Mathematical Statistics, 29, 614. Wolkowski, R.P., Reisdorf, T.A., Bundy, L.G., 1988. Field plot technique comparison for estimating corn grain and dry matter yield. Agronomy Journal, 80, 278-280. Yonezawa, K., 1985. A definition of the optimal allocation of effort in conservation of plants genetic resources with application to sample size determination for field collection. Euphytica, 34, 345-354. 13 G. Rana et al.- Rivista Italiana di Agrometeorologia 14-23 (1) 2004 METODOLOGIE E ALGORITMI PER IL CONTROLLO DI QUALITÀ DI DATI ORARI E GIORNALIERI ACQUISITI DA UNA RETE AGROMETEOROLOGICA: APPLICAZIONI ALLE RETE LUCANA SAL METHODS AND ALGORITHMS FOR EVALUATING THE DATA QUALITY AT HOURLY AND DAILY TIME SCALES FOR AN AGROMETEOROLOGICAL NETWORK: APPLICATION TO THE REGIONAL NET OF BASILICATA (ITALY) Gianfranco Rana*, Michele Rinaldi, Michele Introna Istituto Sperimentale Agronomico, Via C. Ulpiani, 5 70125 Bari Italy Riassunto In questo lavoro sono esaminate le tecniche che si dovrebbero utilizzare per effettuare un controllo di qualità di grandezze agrometeorologiche acquisite in campo da stazioni standard. Primo passo necessario è calibrare periodicamente i sensori, per calcolarne i moltiplicatori. Qui, inoltre, sono stati descritti i metodi e gli algoritmi che si utilizzano per i controlli fisici e formali che hanno come obiettivo quello di determinare se un dato è accettabile; poi si è descritto brevemente il metodo da utilizzare per costruire l’algoritmo necessario a ristabilire l'integrità dei dati quando uno di essi è stato dichiarato inaccettabile. I metodi descritti sono stati applicati ad un caso reale: una serie di dati acquisiti dalla rete agrometeorologica della regione Basilicata. I risultati mostrano che, se viene effettuata periodicamente una calibrazione corretta, i valori non risultano affetti da un errore significativo e risultano di qualità accettabile per le applicazioni pratiche. Parole chiave: controlli fisici, controlli statistici, rete agrometeorologica, calibrazione dei sensori, errore. Abstract In this paper the techniques to be used for the quality control of standard agrometeorological measurements are studied. First, it is necessary to calibrate periodically the sensors, in order to compute the conversion coefficients. Furthermore, the methods and algorithms used for the physical and statistical controls are detailed here, in order to determine if the data are correct; moreover, a short discussion on the subject of replacing incorrect data by a suitable value is carried out. The methods described have been applied to a series of data acquired by the local agrometeorological network of Basilicata (Southern Italy). The results showed that if a correct periodical calibration has been made periodically, the errors affecting the data are small, at least for practical applications. Keywords: physical controls, statistical controls, agrometeorological network, sensor calibration, error. Introduzione Partiamo dal presupposto che: è meglio non prendere in considerazione un cattivo dato che tenerne conto in ogni caso. Infatti, se si prende per buono un dato che buono non è, si commette senz'altro un errore molto maggiore di quello che si commetterebbe considerando un dato "medio", magari ricavato dalla serie storica (anche limitata nel tempo) della stazione. Per cui, prima di utilizzare un valore misurato da un sensore, è necessario controllarne l'integrità e la correttezza, in altre parole è necessario effettuare un controllo sulla sua qualità (WMO, 1983a). Se poi un dato climatico affetto da errore sistematico viene utilizzato in un modello matematico, ciò può portare ad errori, anche molto consistenti, sul risultato finale del modello. Per dare un’idea dell'ordine di grandezza *Corresponding author. tel. +39 080 5475011 fax +39 080 5475023 e-mail: [email protected] (G.Rana). 14 degli errori che si possono commettere facciamo un esempio: supponiamo di dover calcolare l'evapotraspirazione di riferimento (ET0) con il modello di PenmanMonteith (Allen et al., 1998); questo modello richiede misure di temperatura e umidità dell'aria, velocità del vento e radiazione solare globale. Se si suppone di commettere un errore sistematico di +2 °C sulla temperatura l'errore finale sull’ET0 nel periodo irriguo estivo (giugno, luglio e agosto) è di ben 40 mm, ovvero di 400 m3/ha di acqua. I dettagli relativi a questo esempio sono riportati in Tabella 1; nella stessa tabella è anche presentato l’effetto sull’ET0 di un errore sistematico di 10 punti percentuali nell’umidità relativa dell’aria. Se il dato dovesse risultare errato anche dopo i controlli descritti nel seguito, sarà necessario sostituirlo con un valore appropriato, perché la serie storica resti integra e/o perché possano essere effettuate tutte le operazioni previste con quel dato. G. Rana et al.- Rivista Italiana di Agrometeorologia 14-23 (1) 2004 Si presuppone ovviamente che il valore misurato provenga da un sensore opportunamente calibrato secondo i procedimenti indicati in WMO (1981; 1983b). Le modalità di calibrazione sono descritte in un paragrafo successivo. Il controllo di qualità dei dati agrometeorologici va fatto in due momenti distinti (SIAN, 1990): prima si effettuano dei controlli formali o fisici (controllo sulla accettabilità generale del dato), poi si eseguono dei controlli statistici, basati sia sull'analisi delle serie storiche sia sul controllo dei dati provenienti da stazioni limitrofe. Dal punto di vista operativo, si procede per i seguenti due passi: 1) Controlli fisici e formali per determinare se un dato è accettabile; 2) Adozione di un opportuno algoritmo per ristabilire l'integrità del dato quando esso è stato dichiarato inaccettabile. I controlli statistici, dettagliati in successivi paragrafi, devono far riferimento sia al clima della zona interessata, sia ai dati misurati nei giorni precedenti, sia alle misure effettuate dalle stazioni limitrofe. In sintesi, si effettuano i seguenti controlli (SIAN, 1990): a) controllo sulla variabilità interannuale (clima); b) controllo sulla persistenza temporale (dati rilevati nei giorni precedenti); c) controllo sulla coerenza spaziale (dati rilevati nelle stazioni limitrofe). Un discorso a parte merita il problema della standardizzazione dei dati agrometeorologici, ovvero il controllo della uniformità dei sensori utilizzati e delle condizioni ambientali in cui la misura è stata effettuata, ovvero la valutazione del sito in cui il dato è stato rilevato. La calibrazione dei sensori per confronto Per la calibrazione dei sensori e la verifica della loro funzionalità si seguono le indicazioni dell'Organizzazione Meteorologica Mondiale (OMM), in particolare, le procedure standard indicate nei documenti ufficiali WMO-no. 49 (1988 e successive modifiche e integrazioni), WMO-no. 134 (1981 e successive modifiche e integrazioni) e WMO-no. 8 (1983). La calibrazione di routine viene effettuata per confronto, ovvero mettendo in relazione i dati provenienti dalla stazione sotto controllo con quelli acquisiti da una stazione dotata di sensori di classe di precisione superiore (Schwerdtfeger, 1976). Nel caso della rete del Servizio Agrometeorologico Lucano (SAL) sono stati calibrati i sensori relativi alle seguenti grandezze: radiazione solare globale, temperatura dell'aria, umidità relativa, velocità del vento, bagnatura fogliare. Le misure sono state effettuate con le stesse modalità di acquisizione della rete SAL. Da una breve analisi di serie storiche di dati agrometeorologici relative ad alcune località della regione Basilicata (vedi Tab. 2), é risultato che il periodo di confronto ottimale tra stazione di calibrazione e stazione in osservazione è di circa una settimana. In alcuni casi, dopo l'immediata analisi dei dati, si è proceduto a prolungare la calibrazione per confronto fino ad una ulteriore settimana. La calibrazione per confronto fornisce i due coefficienti (s e i), rappresentanti la pendenza e l'intercetta della regressione lineare: M rif = s ⋅ M + i (1) dove Mrif rappresenta la misura effettuata dal sensore di riferimento e M la misura effettuata con il sensore della stazione. I due coefficienti vanno poi utilizzati per correggere i dati provenienti dalle stazioni. Tab. 1 - Errori totali commessi sull'evapotraspirazione di riferimento mensile calcolata con il modello di PenmanMonteith, supponendo di commettere errori sistematici sulla misura della temperatura dell'aria (T) e dell'umidità relativa dell'aria (UR). Tab. 1 - Total errors on the reference evapotranspiration computed by the Penman-Monteith model, supposing that constant errors have been made on the measurements of air temperature (T) and relative humidity (UR). Errore giugno 2 °C (T) 10 mm 10% (UR) 15 mm luglio 20 mm 25 mm agosto 10 mm 15 mm Totale 40 mm 55 mm Applicazioni alla rete SAL Nel periodo maggio 2000-ottobre 2001 sono state calibrate le 39 stazioni SAL riportate in Tabella II (tra parentesi la sigla di codifica della stazione). I sensori presenti nelle stazioni SAL sono di tipo standard, di differenti case costruttrici e modello, alcuni non largamente commercializzati. Vista, dunque, la loro grande varietà si preferisce fornire solo l’intervallo di precisione in cui cadono tutti i sensori analizzati, per ogni grandezza fisica considerata. - Radiazione solare globale, ±1% - ±3% del fondo scala. - Temperatura dell'aria, ±0.5 °C. - Umidità relativa, ±5 - ±10% del fondo scala. - Velocità del vento, ±5% del fondo scala, minimo 0.5 m s-1 soglia di avvio. - Bagnatura fogliare, precisione ±7 - ±10 % del fondo scala. Le caratteristiche metrologiche dei sensori utilizzati per la calibrazione sono qui di seguito riportate. - Radiazione solare globale, sensore Eppley PSP di classe I (Fritschen e Gay, 1979), linearità ±0.5% del valore di fondo scala, tempo di risposta 1 s, sensibilità 9 µV W-1 m2. - Temperatura dell'aria, sensore Rotronic MP 101, precisione ±0.3 °C, tempo di risposta 15 s. - Umidità relativa, sensore Rotronic MP101, precisione ±1% del fondo scala, stabilità <1% all'anno, tempo di risposta 10 s. 15 G. Rana et al.- Rivista Italiana di Agrometeorologia 14-23 (1) 2004 - Velocità del vento, sensore Met One 010C, precisione ±1% del fondo scala,soglia di avvio,0.27 m s-1. - Bagnatura fogliare, sensore Campbell Sci. Mod. 237, precisione ±5% del fondo scala. In Tabella 3 sono indicati i giudizi complessivi dati alle stazioni, seguendo i suggerimenti dell’OMM, il significato del giudizio sintetico è il seguente: A - sito rappresentativo, stazione in buona posizione (standard), sensori opportunamente disposti; B - sito rappresentativo, stazione in posizione accettabile ma migliorabile (quasi standard), posizione dei sensori da ottimizzare; C - sito non rappresentativo, stazione in posizione non standard, consigliabile lo spostamento. Per illustrare i risultati e presentare i principali problemi incontrati, i dati del confronto (tra sensori da calibrare e sensori di riferimento (rif)) sono illustrati per alcuni sensori nelle Figure 1-10. Qui di seguito è riportato il commento e le osservazioni alle figure. Tab. 2 - Elenco delle stazioni della rete SAL calibrate e loro ubicazione; in parentesi le sigle. Tab 2 - List of the calibrated stations of the SAL network and them site; in brackets the short name. Tab. 3 - Giudizio complessivo (vedi testo per i significati). Tab. 3 - Evaluation of each station (see text for the meaning of the score). Numero stazione 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 16 Denominazione e sigla Policoro - C.da Troyli (PO1) Tursi - C.da Marone (MO6) Pisticci - C.da Castelluccio (PI1) Pisticci Scalo - C.da Paolona (PI3) Policoro - Loc. Pantano Sottano (PO3) Grassano Scalo (GR1) Villa d'Agri - AASD Bosco Galdo (BG1) Metaponto - AASD Pantanello (PAN) S. Giorgio Lucano - P. delle Rose (SGL) Stigliano - C.da Torre (STI) Nova Siri - Az. La Collinetta (NS3) Montescaglioso-Az.Fortunato,C.da Fiumicello (MTS) Grottole - C.da Castellana (GRD) Grottole - C.da Serre (GRU) Campomagiore - C.da Montescristo (CMP) Brindisi di Montagna - Serra del Ponte (BRM) Laurenzana - Loc. Abetina (LAU) Ferrandina - C.da Follia (FER) San Mauro Forte - C.da S. Chiara (SMF) Montalbano - CC Del Fico (MO4) Guardia Perticara - C.da Lupara (GUP) Nemoli - Loc. Ventrona (NEM) Viggianello - Loc. Pedali (VIG) Rotonda - Loc. Piana Incoronata (ROT) Rotonda - Loc. Lori (ROS) Castelsaraceno - Loc. Fusci (CSR) Sarconi - C.da Trutolo (SAR) Stariano di Lucania - C.da S. Lucia (SAT) Acerenza - Villaggio Taor (ACE) Aliano - Az. Baderita della M. (ALI) Senise - Piano delle Maniche (SEN) Genzano - C.da Ripa d'Api (GE1) Santa Maria d'Irsi (SMI) Lavello - AASD Gaudiano (LAV) Matera - C-da Matinelle (MTP) Matera - C.da Due Gravine (MTQ) Craco Peschiera (CRP) Bernalda - C.da San Marco (ME3) Metaponto - Agrobios (AGR) Stazione MO6 PI1 PI3 PO3 GR1 BG1 PAN SGL ST1 NS3 PO1 MTS GRD GRU CMP BRM LAU FER SMF MO4 GUP NEM VIG ROT ROS CSR SAR SAT ACE ALI SEN GE1 SMI LAV MTP MTQ CRP ME3 AGR Giudizio A B A B A A A B A A C B A B A C C A A B A A B A A A A B C A A A A A C C A B A G. Rana et al.- Rivista Italiana di Agrometeorologia 14-23 (1) 2004 1 :1 Radiazione solare globale La Figura 1 mostra il confronto tra la radiazione globale oraria misurata dal sensore di riferimento e quella misurata dal sensore da calibrare per la stazione di Grassano Scalo (GR1), nel periodo 27 giugno - 4 luglio 2000. Visto che i due andamenti si sovrappongono, è evidente che i due sensori danno valori molto prossimi per tutto il periodo di osservazione e in tutte le ore della giornata. Il risultato della calibrazione è il seguente: riferimento sensore da calibrare 1 :1 Fig. 1 - Confronto tra la radiazione globale misurata dal sensore sotto osservazione e la radiazione di riferimento, per la stazione di Grassano Scalo (GR1). Fig.1 - Comparison between global radiation measured by the sensor under calibration and the reference radiation, in the station of Grassano Scalo (GR1). riferimento sensore da calibrare radiazione teorica R g , rif = 1.0074 ⋅ R g − 8.5853 (r2=0.9935) (2) Come si può vedere, in questo caso, il valore della pendenza è quasi identico all'unità, mentre il valore dell'intercetta (in W m-2) non è significativamente diverso da 0. Tale constatazione è confermata anche dai test statistici effettuati tramite il pacchetto STAT del SAS. Dai risultati di questa calibrazione si può affermare che il sensore di radiazione globale della stazione di Grassano Scalo è in ottime condizioni e non ha bisogno di nessuna correzione. In Figura 2 è riportato il risultato del confronto della radiazione globale, nella stazione di Policoro - C.da Troyli (P01) tra il 5 e il 12 maggio 2000. Per i due giorni più sereni (il quarto e il quinto), nelle ore centrali della giornata, il sensore in osservazione dà delle radiazioni molto elevate, ben oltre i 1000 W m-2, non riscontrabili dal sensore di riferimento. Per valutare tale comportamento, seguendo Allen (1996), abbiamo calcolato la radiazione teorica per il sito in oggetto e riportato questi valori in Figura 2 (linea punteggiata). Si può osservare che la radiazione solare misurata dal sensore di riferimento è sempre al di sotto della radiazione teorica, si può dunque concludere che è il sensore della stazione SAL a non funzionare correttamente per questi particolari giorni. Per il resto sembra procedere tutto regolarmente. E' evidente che per questo sensore non vi è alcuna possibilità di correzione del dato prima di aver compreso esattamente il suo comportamento e aver analizzato una serie di dati sufficientemente congrua (alcuni giorni di seguito per ciascuna stagione). Particolare attenzione va rivolta alla posizione del sensore all'interno del recinto della stazione in quanto la misura può essere notevolmente influenzata dalla presenza di pali o altre superfici riflettenti poste nelle vicinanze dell'apparecchio. Temperatura dell'aria In Figura 3 è illustrato il confronto tra la temperatura misurata con il sensore in osservazione e quella misurata con il sensore di riferimento, per la stazione di Pantanello, tra il 20 e il 27 giugno 2000. Si può chiaramente osservare che il sensore della stazione SAL funziona bene, infatti la regressione lineare dà il seguente risul-tato: Fig. 2- Confronto tra la radiazione globale misurata dal sensore sotto osservazione, la radiazione di riferimento e la radiazione teorica, per la stazione di Policoro - C.da Troyli (P01). Fig. 2- Comparison between global radiation of the sensor under calibration, reference radiation and theoretical radiation, for the station of Policoro – C.da Tryli (PO1). Trif = 1.0296 ⋅ T − 0.1273 (r2=0.9979) (3) con una pendenza leggermente superiore a 1 e un'intercetta appena superiore al decimo di grado. 17 G. Rana et al.- Rivista Italiana di Agrometeorologia 14-23 (1) 2004 1 :1 L’errore di sovrastima (ca. 3%) può essre considerato trascurabile agli effetti pratici. In Figura 4 è mostrato il confronto delle temperature per la stazione di Grassano Scalo (GR1), nel periodo 27 giugno - 4 luglio 2000. In questo caso il sensore della stazione sottostima notevolmente, infatti il risultato della regressione lineare dà: riferimento sensore da calibrare 2 (4) Trif = 1.0365 ⋅ T + 1.1304 (r =0.9957) con un'intercetta piuttosto elevata, pari a 1.1 °C circa. Poiché l'andamento nel periodo di osservazione è sostanzialmente corretto, sembrerebbe che sia presente uno scarto sistematico negativo, probabilmente dovuto all'errore di ripetibilità del sensore (errore sistematico relativo all’intercambiabilità del sensore), ed eliminabile via programma. 1 :1 Umidità dell'aria In Figura 5 è illustrato il confronto tra l’umidità relativa misurata con il sensore in osservazione e quella misurata con il sensore di riferimento per la stazione di Pantanello, tra il 20 e il 27 giugno 2000. 1 :1 Fig. 3 - Confronto tra temperatura misurata dal sensore sotto osservazione e la temperatura di riferimento, per la stazione di Metaponto -AASD Pantanello (PAN). Fig. 3- Comparison between temperature of the sensor under calibration and reference temperature, for the station of Metaponto –AASD Pantanello (PAN). riferimento sensore da calibrare riferimento sensore da calibrare Fig. 5 - Confronto tra umidità relativa dal sensore sotto osservazione e l'umidità relativa di riferimento, per la stazione di Metaponto - AASD Pantanello (Pan). Fig. 5 - Comparison between the relative humidity measured by the sensor under calibration and the reference relative humidity, for the station of Metaponto –AASD Pantanello(Pan). Fig. 4 - Confronto tra temperatura misurata dal sensore sotto osservazione e la temperatura di riferimento, per la stazione di Grassano Scalo (GR1). Fig. 4 - Comparison between temperature measured by the sensor under calibration and the reference temperature, for the station of Grassano Scalo, (GR1). 18 Anche in questo caso, si può chiaramente osservare che il sensore della stazione SAL funziona bene, infatti la statistica dà il seguente risultato: (r2=0.9984) (5) RH rif = 1.0183 ⋅ RH − 2.1485 con una pendenza leggermente superiore 1 e un'intercetta molto bassa, intorno ai 2 punti percentuali. 1 :1 1 :1 G. Rana et al.- Rivista Italiana di Agrometeorologia 14-23 (1) 2004 riferimento sensore da calibrare riferimento sensore da calibrare In Figura 6 è mostrato il confronto delle temperature per la stazione di Tursi - C.da Marone (MO6), tra il 30 maggio e il 3 giugno 2000. In questo caso il sensore della stazione sottostima notevolmente, infatti il risultato della regressione lineare dà: RH rif = 0.974 ⋅ RH + 19.019 Fig. 7 - Confronto tra velocità del vento del sensore sotto osservazione e velocità del vento di riferimento, per la stazione di Metaponto - AASD Pantanello (Pan). Fig. 7 - Comparison between the wind speed measured by the sensor under calibration and reference wind speed, for the station of Metaponto –AASD Pantanello (Pan). 1 :1 Fig. 6 - Confronto tra umidità relativa dal sensore sotto osservazione e umidità relativa di riferimento, per la stazione di Tursi - C.da Marone (MO6). Fig. 6 - Comparison between relative humidity measured by the sensor under calibration and the reference humidity, for the station of Tursi – C.da Marone (MO6). (r2=0.9732) (6) con un'intercetta piuttosto elevata pari a 20 punti percentuali. Inoltre, i dati del giorno 2 giugno, dalle 11:00 alle 23:00 sono completamente errati. L'impressione, in questo caso, è che il sensore debba essere sostituito e che i dati in archivio siano comunque recuperabili applicando la (6). riferimento sensore da calibrare Velocità del vento In Figura 7 è illustrato il confronto tra velocità del vento misurata con il sensore in osservazione e quella misurata con il sensore di riferimento per la stazione di Pantanello, tra il 20 e il 27 giugno 2000. Ancora una volta si può osservare che il sensore della stazione SAL ha delle ottime prestazioni, e a conferma di ciò la statistica dà il seguente risultato: Vrif = 0.9719 ⋅V − 0.0156 (r2=0.9974) (7) con una pendenza leggermente inferiore 1 e un'intercetta trascurabile. Fig. 8 - Confronto tra velocità del vento del sensore sotto osservazione e velocità del vento di riferimento, per la stazione di Policoro - C.da Troyli (PO1). Fig. 8 - Comparison between wind speed measured by the sensor under calibration and reference wind speed, for the station of Policoro – C.da Tryli (PO1). 19 G. Rana et al.- Rivista Italiana di Agrometeorologia 14-23 (1) 2004 In Figura 8 è mostrato il confronto delle velocità del vento per la stazione di Policoro - C.da Troyli (PO1), nel periodo 10 maggio - 16 maggio 2000. In questo caso il sensore della stazione sottostima, infatti il risultato della regressione lineare dà: 2 V rif = 0.9114 ⋅ V + 0.5265 (r =0.9015) 10% a un valore sostanzialmente identico a quello misurato dal sensore di riferimento (Figura 10). (8) con un'intercetta piuttosto elevata, pari a 0.5 m s-1. Molti anemometri degli anemometri SAL analizzati hanno presentato questo problema, cioè una intercetta piuttosto alta, segno che i sensori devono essere lubrificati più spesso. Bagnatura fogliare Il confronto tra il sensore della stazione e quello di riferimento ha mostrato che questa grandezza è misurata in maniera sostanzialmente corretta. C'è comunque da sottolineare che la misura dipende molto dalla corretta posizione del sensore (che va curata anche al centimetro) e dal modello di apparecchio utilizzato, per cui in questo caso si è trattato di un test di funzionalità piuttosto che di una calibrazione vera e propria. co rre zio n e Fig. 10 - Effetto della calibrazione sulla misura di umidità relativa, per la stazione di Policoro - C.da Troyli (PO1). Fig. 10 - Effect of the calibration, on the measurement of relative humidity for the station of Poliporo - C.da Troyli (PO1) Calcolo dell’intervallo di variabilità dei dati agrometeorologici Prenderemo in considerazione le seguenti variabili agrometeorologiche: la radiazione solare (globale), la temperatura dell'aria, l'umidità dell'aria, la velocità del vento, la direzione del vento. Per ognuna di esse proporremo un intervallo di variabilità, basandoci su considerazioni fisiche e sul metodo di misura adottato per rilevare la variabile. Si prenderanno in considerazione sia i dati rilevati su scala temporale oraria che giornaliera, specificando di volta in volta il tipo di controllo. sensore da calibrare riferimento co rre zio n e Fig. 9 - Confronto tra umidità relativa del sensore sotto osservazione e umidità di riferimento, per la stazione di Policoro - C.da Troyli (PO1), prima e dopo la correzione dei moltiplicatori. Fig. 9 - Comparison between relative humidity measured by the sensor under calibration and reference humidity, for the station of Policoro – C.da Troyli (PO1), before and after the correction of the sensor coefficients. Efficacia della calibrazione per confronto Per valutare l'efficacia e la validità del metodo di calibrazione da noi adottato, ricorriamo al confronto effettuato sull'igrometro della stazione di Policoro - Troyli (PO1). Dal confronto effettuato tra il 7 e il 10 maggio 2000 (Figura 9), risulta che: RH rif = 0.95 ⋅ RH sensore − 3.11 (9) con un r2=0.84. Applicando i coefficienti della (9) ai dati acquisiti nei giorni 11 - 13 maggio 2000 otteniamo l'umidità relativa riportata nella coda del grafico in Figura 9. Come si vede si passa da una sottostima di circa il 20 La radiazione solare (globale) Il flusso di radiazione solare (Rs) misurato da un solarimetro in una data località per un certo intervallo di tempo, deve essere inferiore o uguale a quello misurato in una giornata limpida e serena per lo steso intervallo di tempo (Rs0) (10) R s ≤ Rs 0 Una stima accurata della radiazione solare che raggiunge la superficie terrestre Rs0, può essere fatta utilizzando la radiazione solare extraterrestre Ra e la trasmittanza atmosferica, rappresentabile tramite l'indice di trasparenza (KT, adimensionale) Rs0 = K T ⋅ Ra (11) L'indice di trasparenza può essere stimato (Allen et al., 1994; Monteith e Unsworth, 1990) dalla relazione: K T = 0.75 + 2 −5 z (12) con z (m) altezza sul livello del mare del sito in oggetto. Questa relazione è stata ricavata per un angolo solare medio di 45° sull'orizzonte. La relazione (12) è stata scelta dopo un'accurata sperimentazione e per la sua semplicità, vi è però la possibilità di migliorare le stime di KT nel caso di siti particolarmente inquinati o in cui G. Rana et al.- Rivista Italiana di Agrometeorologia 14-23 (1) 2004 l'umidità dell'aria è, normalmente, molto più elevata della media dell'area in esame (si veda Allen, 1996). Nella (11) rimane da calcolare la radiazione extraterrestre Ra. Controllo sulla radiazione solare a scala giornaliera In questo caso Ra, per una data località, è funzione della latitudine e del giorno dell'anno ed è data dalla relazione (Brutsaert, 1982): Ra = 24 ⋅ 3600 GSC d r (ω s senϕsenδ + cos ϕ cos δsenω s ) π 10 6 (13) con Gsc costante solare (1367 W m-2); dr è la distanza relativa terra-sole, ϕ è la latitudine in radianti, δ è la declinazione solare e ωs è l'angolo solare. Qui di seguito sono illustrate, separatamente, le relazioni da utilizzare per determinare queste ultime grandezze. dr è data dalla relazione: ⎛ 2π ⎞ d r = 1 + 0.033 ⋅ cos⎜ J⎟ ⎝ 365 ⎠ (14) con J giorno dell'anno (1- 365). La conversione della latitudine ϕ da a gradi (°), b primi (') e c secondi (") a radianti si effettua secondo la seguente relazione: π b c (15) ) ϕ= ⋅ (a + + 180 60 3600 La declinazione solare δ è data da: ⎡ (284 + J ) ⎤ (16) δ = 0.4093 ⋅ sen ⎢ 2π 365 ⎥⎦ ⎣ L'angolo solare ωs in radianti può essere calcolato come: ω s = arccos[− tan (ϕ ) tan (δ )] (17) Controllo sulla radiazione solare a scala oraria Per intervalli di tempo di un'ora, o più brevi, la radiazione solare extraterrestre media nel periodo considerato (Rah, in Wm-2) è data dalla relazione (Allen, 1996) Rah = 12 G sc d r [(ω 2 − ω1 )senϕsenδ + cos ϕ cos δ (senω 2 − senω1 )] ∆tπ (18) con ω1 e ω2 rispettivamente l'angolo solare all'inizio e alla fine del periodo considerato in radianti: π∆ t (19) 24 π∆ t (20) ω2 = ω + 24 con ω angolo solare all'istante centrale del periodo considerato ∆t (lunghezza del periodo in ore) in radianti, dato dalla relazione: π (21) ω = {[t + 0.06667(L z − Lm ) + S c ] − 12} 12 ω1 = ω − con: t, ora standard all'istante centrale del periodo considerato in ore (per esempio, per l'intervallo tra le 9:00 e le 10:00 è t=9.5; per l'intervallo tra le 14:00 e le 15:00 è t=14.5); Lz è la longitudine della località geografica in cui per convenzione viene misurata l'ora standard, in gradi Ovest da Greenwich ( per esempio, per Greenwich è Lz=0°, nel nostro caso l'ora standard è misurata nel fuso orario di Parigi, per cui è Lz=345°); Lm è la longitudine del sito di misura, in gradi Ovest da Greenwich. Sc è una grandezza per correggere l'ora solare stagionale in ore: (22) S c = 0.1645 sin( 2b) − 0.1255 cos(b) − 0.025 sin(b) con b: b= 2π ( J − 81) 364 (23) Nota bene che è possibile che la misura istantanea della radiazione solare Rs possa superare il valore "teorico" Rs0 in giornate parzialmente nuvolose. Questo fenomeno è causato dalla riflessione di raggi solari diretti da nubi vicine; ciò si verifica comunemente in montagna se il solarimetro è situato in una valle. In ogni caso, tali valori non possono superare la radiazione teorica per più del 2030%, per cui ne va valutata la reale esistenza e portata tramite un confronto con i dati rilevati in istanti precedenti e/o da stazioni limitrofe. Temperatura dell'aria L’intervallo di variazione della temperatura dell'aria varia, ovviamente, da stazione a stazione. Esso va, dunque, ricavato dall'analisi delle serie storiche; dall'analisi dei dati lucani in nostro possesso risulta che un valore di temperatura può essere ritenuto teoricamente "accettabile" se varia tra -10°C e +50°C. Nel caso un valore misurato cada al di là di questi limiti è necessario effettuare i controlli formali descritti nei paragrafi successivi (variabilità interannuale, persistenza temporale e coerenza spaziale). Umidità dell'aria Poiché la pressione di vapore è difficile da misurare si ricorre spesso a sensori elettronici per la misura diretta dell'umidità relativa in percentuale (UR). Tali sensori presentano, normalmente, difetti insiti nel processo costruttivo (isteresi, non linearità, coefficiente di calibrazione non accurato). Altri errori possono essere legati all'esposizione all’aperto del sensore, dunque a polvere, insetti, sostanze inquinanti, usura. Se il dato di umidità relativa dell'aria è di fondamentale importanza, per il servizio che la rete agrometeorologica deve fornire, l'unico mezzo sicuro per il controllo di qualità del dato è la duplicazione del sensore. Ovvero bisogna porre due sensori identici a pochi centimetri di distanza l'uno dall'altro (WMO, 1983b; Allen, 1996), e il dato dell'uno sarà confermato o confutato dalla misura effettuata con l'altro. Purtroppo questa è una tecnica costosa e quasi sempre impossibile da attuare nelle stazioni standard, per cui vediamo come effettuare altri controlli automatici su questa 21 G. Rana et al.- Rivista Italiana di Agrometeorologia 14-23 (1) 2004 grandezza. Un primo controllo da effettuare sull'UR riguarda la tendenza: normalmente, durante la giornata, l'umidità ha un andamento opposto a quello della temperatura, ovvero quando T aumenta UR diminuisce e viceversa. Un altro controllo va fatto sulla temperatura di rugiada Td (°C), variabile legata alla pressione parziale di vapore dell'aria (e). Partiamo dalla considerazione che e è esprimibile tramite l'umidità relativa (UR) e la pressione di vapor saturo alla temperatura, es(T) (kPa), come: e= UR e (T ) 100 s (24) La relazione tra e e la temperatura di rugiada Td (temperatura alla quale l’aria diventerebbe satura se venisse raffreddata senza che intervenga una variazione di pressione) è la seguente (Tetens, 1930): Td = 116.91 + 237 ⋅ ln (e ) 16.78 − ln (e ) (25) Mentre l'UR varia considerabilmente durante l'arco delle 24 ore e, come si è detto, in maniera opposta alla temperatura, Td ed e sono relativamente stabili nella giornata, se non vi è avvezione (trasporto di vapore e/o calore dalle zone limitrofe alla stazione), presentando un debole andamento a "campana", nelle giornate serene, conforme all'andamento della temperatura. A meno che non vi sia un fronte umido (da valutare con altri mezzi, per esempio la pressione atmosferica) le variazioni massime di Td da un'ora all'altra non devono superare il 20%, se ciò avviene questo può essere sintomo di malfunzionamento dell'igrometro. Un ulteriore controllo sulla Td va fatto anche in relazione alla temperatura minima dell'aria (Tmin), infatti, per la maggior parte dei giorni la differenza tra Td e Tmin non supera i 2°C. Solo eccezionalmente, in giornate particolarmente secche e in ambienti aridi e semi-aridi in presenza di avvezione, tale differenza può raggiungere i 3°C (Allen et al., 1998), valore che dunque può ritenersi un limite da non superare. In presenza di rugiada mattutina Td rimane molto vicina a Tmin e in questo caso l'UR è vicina al 100%. Tali considerazioni sono valide anche in caso di pioggia. Se un sensore elettronico di umidità relativa supera il valore massimo (100%) di più del 5% (ovvero sia >105%) vi è necessità di ricalibrazione e di riaggiustamento dei dati raccolti. Velocità del vento L’intervallo di funzionamento di un anemometro può essere posto tra 0.5 e 50 ms-1 (WMO, 1983b). Anche in questo caso, per stazioni di particolare importanza e se il dato di velocità del vento (v) è fondamentale, l'ideale sarebbe duplicare il sensore. Particolare importanza va data alla soglia dello strumento, ovvero a quel valore minimo di v ( normalmente presente nei programmi di gestione della stazione) che vince l'attrito del rotore sul corpo dell'anemometro. E' un valore dato dal costruttore, ma vanno verificati conti- 22 nuamente i periodi di calma (velocità del vento inferiore a 0.5 ms-1) nell'arco dell'intera giornata (che normalmente sono abbastanza rari), in quanto la velocità di soglia di un anemometro è fortemente dipendente dall'usura e dall'esposizione. In pratica, se la calma è misurata per più di un'ora durante il giorno, il sensore va tenuto sotto controllo. Durante le ore notturne è possibile che i periodi di calma siano più lunghi di un'ora. Un altro controllo necessario per la velocità del vento in una rete agrometeorologica consiste nell'effettuare il cosiddetto "bilancio di massa" per il quale è necessario mettere in relazione i valori cumulati giornalieri delle v misurate in due stazioni limitrofe. La pendenza negli andamenti di questi due cumulati non dovrebbe cambiare sensibilmente nel tempo (Allen, 1996). Direzione del vento L’intervallo di funzionamento va ovviamente tra 0° e 360°. Se il territorio non è omogeneo la direzione del vento (dir) può variare sensibilmente da stazione a stazione. Un controllo può essere fatto tra dati provenienti da stazioni limitrofe, se si è in possesso di una adeguata serie storica di misure di dir. Normalmente, l'orografia del territorio rende quasi sempre costante la differenza tra le dir misurate in stazioni limitrofe. Controlli statistici Dopo aver effettuato i controlli di tipo fisico sopra descritti, sia su scala oraria che, nel caso della radiazione, su scala giornaliera vanno eseguiti i controlli formali che convalidano definitivamente il dato proveniente dalla stazione (WMO, 1983a; 1983c). Si farà riferimento ai soli dati giornalieri. Controllo sulla variabilità interannuale Per ogni stazione, a partire da una serie storica dei dati, viene definita la "climatologia" di ogni giorno dell'anno (SIAN, 1990), come la media temporale, su tutti gli anni a disposizione, dei valori osservati nell'intervallo di 31 giorni centrato nel giorno in esame. Per esempio, la climatologia delle temperature medie di una data stazione, per il giorno 16 marzo, è la media delle osservazioni rilevate dal 1 al 31 marzo di tutti gli anni a disposizione. Dal punto di vista formale, si tratta, per la stazione xi, di prendere la media mobile della variabile z, su n anni (a disposizione nella serie storica), di 31 giorni centrati sul giorno τ: 1 n τ +15 (26) z z ,τ = ∑ ∑ z sat m a =1 t =τ −15 con zsat valore misurato dalla stazione s nell’anno a (con a=1, …,m) e giorno dell’anno t (con t=τ-15, …,τ+15). Si definisce "anomalia" della variabile z misurata nella stazione xi nel giorno ta, la differenza tra la climatologia e il valore misurato z(xi,τa)=zsat: ysat = z s ,τ − z sat (27) Per ogni stazione e per ogni giorno-tipo viene calcolata la deviazione standard delle anomalie, ovvero delle dif- G. Rana et al.- Rivista Italiana di Agrometeorologia 14-23 (1) 2004 ferenze tra il valore misurato e la climatologia di stazione: 1 n τ +15 2 (28) ∑ ∑ y sat m a =1 t =τ −15 Il valore misurato z(xi,ta) viene considerato sospetto se fuoriesce da un intervallo di 3 deviazioni standard, σ(xi,t) dalla climatologia di stazione. In altre parole il valore misurato è valido se è verificata la seguente relazione: σ s ,τ = y sat ≤ 3σ s ,τ (29) Controllo sulla persistenza temporale L'analisi dell'autocorrelazione temporale del segnale mostra che il 95% della varianza viene perso dopo circa 15 giorni, e il coefficiente autoregressivo (ovvero la percentuale di varianza temporale spiegata dalla covarianza tra il giorno t e il giorno t+1 in ciascuna stazione xi) è di circa il 70% d'inverno e leggermente inferiore d'estate: ciò indica una buona correlazione tra misure rilevate in giorni successivi. Per cui, generalmente, brusche variazioni nei valori osservati vengono considerate sospette, anche se nella realtà questa evenienza può verificarsi. Viene dunque calcolata la deviazione standard, σ(k,k-1), degli incrementi tra un giorno e il giorno precedente (che viene definita "tendenza") per tutte le sequenze di dati validi a disposizione. Si considerano sospetti quei valori la cui differenza in valore assoluto δ tra il valore della variabile al giorno k e quello nel giorno precedente k-1 supera di 3 volte la deviazione standard delle tendenze: (30) δ = z sat , k − z sat ,k −1 ≤ 3σ (k , k − 1) Controllo sulla coerenza spaziale Il primo e il secondo controllo statistico danno indicazioni sulla coerenza climatologica e temporale (tra anni successivi e giorni successivi) in una singola stazione. Tuttavia, un valore ritenuto sospetto dopo questi due controlli, può essere causato da una re-ale anomalia meteorologica e pertanto ogni dato segna-lato come sospetto deve essere confrontato con i dati ri-levati nello stesso giorno dalle stazioni limitrofe (test della coerenza spaziale), in un intorno definito dal inter-vallo di correlazione spaziale e con un intervallo fidu-ciale definito dalla deviazione standard delle differenze di misura tra stazioni. Anche in questo caso la devia-zione standard delle differenze tra la stazione xi e le sta-zioni limitrofe xj: d ij = z sati − z satj (31) viene calcolata con una media mobile su 31 giorni e mediata sulle j stazioni limitrofe. Se un dato viene confermato da almeno una stazione limitrofa, ovvero se la differenza con il dato rilevato nella stazione limitrofa è compresa nell'intervallo fiduciale definito in precedenza, allora esso viene considerato corretto, altrimenti viene ritenuto definitivamente errato. Questo procedimento può essere applicato tranquillamente alla verifica di tutte le variabili agroclimatiche precedentemente elencate, mentre va applicato con precauzione nel caso della pioggia, in quanto possono real- mente verificarsi forti differenze tra le precipitazioni effettivamente cadute in stazioni limitrofe. Sono comunque da considerarsi sospetti i dati di precipitazione nulla quando nell’intorno piove. Risultato del controllo e sostituzione dati mancanti Se, dopo i controlli descritti in precedenza, il valore in esame dovesse risultare errato, si procede all’accensione di un segnale di dato non corretto e all’aggiunta di un valore sostitutivo opportunamente calcolato in un altro campo. Dovendo sostituire dei valori ritenuti errati, si utilizzerà un metodo basato sull'interpolazione del dato mancante a partire dai dati presi dalle stazioni limitrofe. Si dovrà dunque tener conto del grado di correlazione del valore misurato in una stazione con quelli acquisiti nelle stazioni vicine. Vi è dunque le necessità di uno studio ad hoc sulla rete agrometeorologica lucana. Ringraziamenti Gli autori ringraziano Nicola Cardinale, Domenico Fiorino e Giovanni Lazzari di Metapontum Agrobios per la stretta collaborazione nella raccolta dei dati e gestione della rete SAL e Pasquale Introna per suo prezioso contributo nella calibrazione dei sensori in campo. Bibliografia Allen, R.G., Smith, M., Pereira, L.S., Perrier, A., 1994. An update for the calculation of reference evapotranspiration. ICID Bull., 4(2), 35-94. Allen, R.G., Pereira, L.S., Raes, D., Smith, M., 1998. Crop evapotranspiration. Guidelines for computing crop water requirements. FAO Irr. Drain Paper no. 56, Rome, 299 pp. Allen, R.G., 1996. Assessing integrity of weather data for reference evapotranspiration estimation. J. Irr. Drain. Eng., 122(2), 97-106. Fritschen, L.J., Gay, L.W., 1979. Environmental instrumentation. Springer-Verlag, New York, 215 pp. Brutsaert, W., 1982. Evaporation into atmosphere. D. Reidel Pub. Co., Dordrecht, Holland. Monteith, J.L., Unsworth, M.H., 1990. Principles of environmental physics, 2nd ed., Edward Arnold, London, UK. Schwerdtfeger, P., 1976. Physical principles of micrometeorological measurement. Elsevier Sci. Pub. Co, Amsterdam, 116 pp. SIAN (Sistema Informativo Agricolo Nazionale), 1990. Analisi climatologica e progettazione della rete agrometeorologica nazionale. Ministero dell'Agricoltura e delle Foreste, Direzione Generale della Produzione Agricola), aprile 1990. Tetens, O. 1930. Uber einige meteorologische. Begriffe. Z. Geophys. 6 : 297-309. WMO, 1981. Guide to agricultural meteorological practices, no. 134. Geneva, Switzerland. WMO, 1983a. Document no. 8. Guide to meteorological instruments and methods of observation. Secretariat of World Meteorlogical Prectices, Geneva, Switzerland. WMO, 1983b. Document no. 49. Technical regulation Volume I, General meteorological standards and recommended practices. Secretariat of World Meteorlogical Organisation, Geneva, Switzerland. WMO, 1983c. Document no. 100, Guide to climatological practices. Secretariat of World Meteorlogical Prectices, Geneva, Switzerland. 23 M. Donatelli et al. - Rivista Italiana di Agrometeorologia 24-30 (1) 2004 GSRAD, UN COMPONENTE SOFTWARE PER LA STIMA DELLA RADIAZIONE SOLARE GSRAD, A SOFTWARE COMPONENT TO ESTIMATE SOLAR RADIATION Marcello Donatelli*, Laura Carlini, Gianni Bellocchi Istituto Sperimentale per le Colture Industriali (ISCI), via di Corticella 133, 40128 Bologna Riassunto GSRad è un componente software utilizzabile in applicazioni che richiedano la stima della radiazione solare in ambiente Windows, indipendentemente dal linguaggio di programmazione usato purché compatibile con la piattaforma .NET o il modello COM. Il componente contiene metodi per il calcolo dei valori giornalieri della radiazione potenziale e per la stima, su base giornaliera e oraria, della radiazione globale al suolo. Per le stime al suolo sono disponibili modelli a base fisica il cui input è rappresentato dalla temperatura dell’aria o dalle ore di insolazione. In aggiunta, è disponibile un metodo per la generazione della radiazione utilizzando parametri statistici. Il componente implementa inoltre un metodo per il calcolo dei valori di inclinazione e orientamento delle superfici captanti non orizzontali a partire dai valori di altimetria circostanti la superficie in esame. Parole chiave: radiazione solare, slope, aspect, .NET, componenti software Abstract GSRad is a cross-language, COM and .NET compatible, software component to be used into Windows-based applications for estimating global solar radiation. The component implements methods to calculate values of both extra-terrestrial and ground-level solar radiation. Daily and hourly data of solar radiation over horizontal or sloped surfaces are estimated according to equations relating atmospheric radiation to either air temperature or sunshine duration. A statistical based approach is also offered as an option to generate global solar radiation at ground level Finally, a method is implemented to compute slope and aspect parameters of non-horizontal surfaces using a digital elevation model as a basis. Keywords solar radiation, slope, aspect, .NET, software components Introduzione Lo sviluppo di componenti software per la realizzazione di applicazioni nel settore della modellistica agrometeorologica costituisce una delle principali attività scientifiche in corso presso la sezione di Tecniche Colturali dell’ISCI di Bologna (Fila et al., 2002), per corrispondere ai moderni paradigmi di sviluppo software. Anche nell’ambito della modellazione agro-ambientale le architetture software di applicazioni stanno infatti evolvendo verso una implementazione che utilizza componenti, cioè verso entità discrete di software progettate per essere sostituibili senza dover modificare l’a-pplicazione che le utilizza, e riutilizzabili in altri software (Szyperski, 2002). In questo contesto è stato affrontato lo sviluppo di un componente per la stima della radiazione solare giornaliera a partire da variabili meteorologiche facilmente reperibili. La radiazione solare, infatti, costituendo la sorgente energetica per la fotosintesi e l’evapotraspirazione, rappresenta un input basilare per i modelli di simulazione della crescita delle colture. A *Corresponding author: tel.. +39 051 6316843 e-mail: [email protected] (M.Donatelli). 24 fronte di tale importanza, risulta ancora limitato il numero delle stazioni che rilevano questa variabile climatica, in rapporto a quelle che registrano temperatura aerea e piovosità (Thornton e Running, 1999; Rivington et al., 2002). Inoltre, le stazioni abilitate alla registrazione della radiazione solare frequentemente manifestano irregolarità nel funzionamento. Tali condizioni costituiscono una grave limitazione per molti studi agronomici e ambientali e impongono la necessità di stime accurate di radiazione solare laddove le misure siano assenti o di scarsa qualità. L’utilizzo di immagini da satellite per la stima della radiazione al suolo rimane ancora una possibilità limitata, dati i vincoli tecnici e le risorse economiche richieste (e.g., Iehlé et al., 1997). Oltre agli approcci usati per generare la radiazione solare su base statistica (e.g., Richardson, 1981; 1982), la ricerca ha prodotto e valutato numerosi modelli per la stima della radiazione solare che utilizzano relazioni fisiche tra questa e altre variabili meteorologiche (e.g. Bristow and Campbell, 1984; Donatelli and Bellocchi, 2001; Weiss et al., 2001; Ball et al., 2004). In particolare, i modelli che utilizzano dati di temperatura aerea o il numero di ore di luce assicurano stime di radiazione sufficientemente accurate (e.g., Ri- M. Donatelli et al. - Rivista Italiana di Agrometeorologia 24-30 (1) 2004 vington et al., 2002). Sebbene l’implementazione di tali modelli in software non commerciali (e.g., RadEst, Donatelli et al., 2003) abbia permesso di supportare studi di modellazione a livello di sistemi colturali (e.g. Mavromatis et al., 2002; Bellocchi et al. 2003; Ferrari et al. 2003; Lyon et al., 2003), lo sviluppo di un componente software specifico (GSRad, Global Solar Radiation) intende mettere a disposizione degli utenti una libreria di funzioni per la stima della radiazione solare, riutilizzabili entro nuove applicazioni. Scopo dell’articolo è la presentazione delle funzionalità e della struttura del componente in rapporto alle potenzialità d’uso in software dedicato. Stima della radiazione solare Il componente consente, per ciascuna latitudine e per ogni giorno dell’anno, di calcolare la radiazione solare potenziale (radiazione solare al di fuori dell’atmosfera terrestre) e di eseguire stime della radiazione alla superficie del suolo. Funzionalità sono inoltre previste per il calcolo dei parametri di orientamento (γ) e inclinazione (α) della superficie captante a partire da modelli digitali di elevazione (digital elevation model o digital terrain model). Radiazione solare potenziale Il calcolo della radiazione potenziale giornaliera incidente su superfici orizzontali, RadPi (MJ m-2 d-1), è derivato da routine basate sui concetti generali della geometria solare (costante solare, declinazione solare, durata del giorno) (Spitters et al., 1986; Pickering et al., 1994). Una generalizzazione per superfici non orizzontali, RadPi(α,γ), è stata realizzata introducendo la dipendenza dall’orientamento e dall’inclinazione della superficie captante (Stine e Harrigan, 1985): RadPi (α , γ ) = RadPi ⋅ cos(α ) + cRadPi ⋅ sin (α ) modello Bristow-Campbell (Bristow e Campbell, 1984) ⎡ ⎛ − b ⋅ ∆Ti2 tt i = τ ⋅ ⎢1 − exp⎜⎜ ⎢⎣ ⎝ ∆Tm - ⎞⎤ ⎟⎥ ⎟ ⎠ ⎥⎦ (4) modello Campbell-Donatelli (Donatelli e Campbell, 1998) (5) tt i = τ ⋅ [1 − exp(−b ⋅ f(Tavg) ⋅ ∆Ti2 ⋅ f(Tmin))] - Donatelli-Bellocchi (Donatelli e Bellocchi, 2001). ⎛ − b ⋅ ∆Ti2 ⎡ ⎛ ⎛ π π ⎛ ⎞ ⎞⎤ ⎡ ⎞ ⋅ f(c2) ⎟ ⎟⎟ ⎥ ⋅ ⎢1 − exp⎜ ⋅ c2 ⎟ + cos ⎜ ir ⋅ tt i = τ ⋅ ⎢1 + c1 ⋅ ⎜⎜ sin⎜ i r ⋅ ⎜ ∆Tw ⎝ 180 ⎠ ⎠ ⎦ ⎣⎢ ⎠ ⎝ ⎝ 180 ⎣ ⎝ ⎞⎤ ⎟⎥ ⎟⎥ ⎠⎦ (6) La radiazione solare può essere generata in base all’approccio di Richardson (1981), per cui i valori giornalieri di temperatura aerea massima e minima e di radiazione solare sono determinati come parte di un processo stocastico multivariato. La dipendenza tra le tre variabili è descritta in Richardson (1982). La stima oraria dei valori di radiazione, Radi(α,γ)-hr, è fatta ridistribuendo il dato giornaliero in base alla variazione oraria dell’angolo zenitale espresso in radianti (Ψhr): (7) cos(ψ ) Rad i (α , γ ) − hr = Rad i (α , γ ) ⋅ hr (ss ) hr ∑( )cos(ψ i hr ) (1) Radiazione solare al suolo Metodi alternativi sono disponibili per la stima al suolo della radiazione solare, Radi(α,γ), (MJ m-2 d-1), basati sull’eliofania o sull’escursione termica giornaliera (dettagli in Tab. 1 e Tab. 2). La stima in funzione delle ore di luce deriva da una implementazione (Rivington et al., 2002) del modello di Johnson-Woodward (Johnson et al., 1995; Woodward et al., 2001): RadPi ⋅ RadDi ⋅ ( f blue ⋅ (1 − ci ) + f cloud ⋅ ci ) RadPi (α ,γ ) (2) Stime al suolo in funzione della temperatura aerea sono derivate dal prodotto della radiazione potenziale per un coefficiente di trasmissività dell’atmosfera (tti): Radi (α,γ ) = RadPi (α,γ ) tti - hr sri Per i dettagli vedere Tab. 1 e Tab. 2. Radi (α , γ ) = SSDi ⋅ RadSi + H i ⋅ giornalieri utilizzando tre modelli alternativi (dettagli in Tab. 1 e Tab. 2): (3) Il coefficiente di trasmissività è stimato dai valori termici I termini sri e ssi rappresentano l’ora dell’alba e del tramonto rispettivamente. Il coseno dell’angolo zenitale è ricavato dalle formule della geometria solare da cui è derivata la radiazione potenziale. Calcolo dei parametri di inclinazione e orientamento La quantità di radiazione incidente su una superficie non orizzontale è influenzata dalle caratteristiche del pendio. L’angolo di incidenza è definito dalla pendenza, mentre l’orientamento della superficie e la latitudine determinano l’insolazione. Il componente implementa routine per il calcolo dei parametri di inclinazione e orientamento della superficie captante, derivate dall’approccio ESRI (http://www.esri.com) e basate sull’utilizzo di dati di elevazione (zi,j) da un grid 3x3 di celle equidistanti (Zi,j): ⎛ z 1,1 ⎜ Z i, j = ⎜ z 2,1 ⎜ ⎝ z 3,1 z 1,2 z 2,2 z 3,2 z 1,3 ⎞ ⎟ z 2,3 ⎟ z 3,3 ⎟⎠ (8) Tale approccio permette il calcolo dei parametri α e γ per un punto tenendo conto dei dati degli otto punti più vicini (Tab. 3). Tutte le metodologie di calcolo sono descritte in dettaglio nel file di help del componente. 25 M. Donatelli et al. - Rivista Italiana di Agrometeorologia 24-30 (1) 2004 Calcolo dei parametri di inclinazione e orientamento di una superficie captante non orizzontale da un modello digitale di elevazione. Calcolo della radiazione solare potenziale. Calcolo della radiazione solare al suolo. Calcolo degli input di temperatura aerea per i modelli della trasmissività. Fig. 1 – Diagramma UML dei casi d’uso. Fig. 1 – UML use cases diagram. Architettura del componente Il componente è stato sviluppato nei linguaggi di programmazione Visual Basic 6 e C#, seguendo il paradigma della programmazione orientata ad oggetti (OOP: Object-Oriented Programming), facendo riferimento alle specifiche della programmazione orientata a componenti (Löwy, 2003), e utilizzando la notazione standard UML (Unified Modeling Language) per sviluppare e documentarne l’architettura. Inoltre, l’approccio design by contract è stato adottato per spiegare l’insieme delle richieste e delle aspettative del componente Diagrammi UML Lo sforzo di progettazione è illustrato mediante i diagrammi UML dei casi d’uso (Use Cases) (Fig. 1) e delle attività (Activity) (Fig. 2). Il diagramma Use Cases mostra l’utente del componente, ovvero l’applicazione, e l’insieme delle connessioni con le funzionalità che il componente dovrà realizzare all’interno dell’applicazione. Il diagramma Activity mostra il flusso di attività svolte dal componente per il calcolo della radiazione globale al suolo. Nel diagramma si ritrovano gli elementi decisionali e condizionali del componente. Ad esempio, il calcolo dei parametri di orientamento e inclinazione di una superficie captante generica costituisce una attività opzionale. La radiazione al suolo può essere stimata con il modello JW che fa uso dell’eliofania (SSD: sunshine duration), con i modelli della trasmissività BC, CD, DB, o con il metodo di Richardson. Tutti i modelli a base fisica richiedono il calcolo della radiazione potenziale. È possibile utilizzare i tre metodi pubblici del componente per la stima degli input, dipendenti dalla temperatura aerea, dei modelli BC, CD e DB. Le classi accessibili dalle applicazioni che usano il componente sono distinte in due tipologie fondamentali: tipi dati (RadData e RadDataTemp) e classi che contengono i metodi pubblici (GSRad e info). GSRad è l’interfaccia del componente e contiene quindi i metodi pubblici per il calcolo dei parametri di orientamento e inclinazione di superfici non orizzontali, 26 SlopeAspect(RadData), della radiazione potenziale, PotentialRadiation(RadData), e della radiazione globale al suolo, GroundRadiation (RadData), usando uno dei modelli previsti (quattro a base fisica e uno a base statistica). La chiamata dei metodi avviene passando un riferimento a un oggetto, che è un’istanza della classe RadData. L’oggetto RadData contiene sia gli input che gli output scritti dal metodo una volta eseguito. L’interfaccia implementa inoltre tre metodi pubblici per il calcolo dei valori di input dei modelli in cui sono richiesti i dati di temperatura aerea: media mensile, MonthlyMeans (RadTempData); media mobile su un periodo di sette giorni, MobileWeekMeans (RadTempData); escursione termica giornaliera, DailyDeltaT (RadTempData). Questi metodi possono essere ignorati nel caso in cui si disponga di questi valori. Esempi di uso in C# e Visual Basic 6 sono commentati nel file di help e resi disponibili come esempi di applicazioni (applicazioni in C# ed Excel). Design by contract Il componente segue l’approccio design by contract (Meyer, 1997), per cui l’integrità degli oggetti viene regolarmente esercitata tramite la verifica di pre-condizioni e post-condizioni. Queste garanzie pubbliche rendono trasparente l’uso degli oggetti, chiarendo i limiti di applicazione. Nei metodi sono specificate le condizioni da soddisfare prima dell’esecuzione del metodo, garantite le quali sono specificate anche quelle che seguono l’esecuzione. Il mancato rispetto delle pre-condizioni rende imprevedibile il comportamento del componente. Pertanto, l’applicazione che utilizza il componente deve esercitare una verifica affinché i dati in input necessari all’esecuzione di ciascun metodo rispettino le precondizioni. Le pre-condizioni e le post-condizioni sono riportate nel manuale d’uso del componente. Conclusioni I componenti riutilizzabili, oltre a favorire la tra-sparenza del know-how inerente agli applicativi sviluppati, accrescono sensibilmente le capacità di lavoro dei modellisti, favorendone la finalizzazione al problem solving e limitando i notevoli sforzi tradizionalmente riservati alla fase di programmazione. Il componente per il calcolo dei valori potenziali e al suolo della radiazione solare è uno strumento utile a supportare collaborazioni tra diversi ricercatori impegnati nella creazione di modelli per l’agronomia e l’agrometeorologia entro il paradigma object-oriented. In particolare, GSRad è parte di una vasta azione scientifica (progetto SIPEAA, sito http://www.sipeaa.it) indirizzata verso la realizzazione di una architettura a componenti e lo sviluppo di applicazioni in cui le funzioni, l’ambiente di esecuzione e la rappresentazione dei dati sono mantenuti separati. Le tecnologie COM e .NET su cui è basato offre agli sviluppatori ampie possibilità di utilizzo del componente in applicazioni scritte con diversi linguaggi (e.g., Delphi, C++, VB.NET, C#). Una versione scritta in linguaggio Java, utilizzabile anche su sistemi operativi Linux è in fase di sviluppo. Questo componente sarà utilizzato nella nuova versione del software RadEst, e come sub- M. Donatelli et al. - Rivista Italiana di Agrometeorologia 24-30 (1) 2004 componente dei componenti CLIMA e RadMap, tutti in sviluppo presso ISCI. Il package per l’installazione, che comprende il manuale d’uso ed esempi di applicazione del componente all’interno dell’editor VBA di MS Excel e in C#, è distribuito gratuitamente per scopi non commerciali tramite il sito http://www.sipeaa.it/tools. Fig. 2 – Diagramma UML delle attività. Fig. 2 – UML activity diagram. Ringraziamenti Ricerca svolta nell’ambito del Progetto SIPEAA (Strumenti Informatici per la Pianificazione Eco-compatibile delle Aziende Agrarie) finanziato dal Ministero delle Politiche Agricole e Forestali, pubblicazione n. 26. Bibliografia Ball, R.A., Purcell, L.C., Carey, S.K., 2004. Evaluation of solar radiation prediction models in North America. Agron. J., 96, 391-397. Bellocchi, G., Donatelli, M., Fila, G., 2003. Calculating reference evapotranspiration and crop biomass using estimated radiation inputs. Ital J. Agron. , 7, 95-102. Bristow, K.L., Campbell, G.S., 1984. On the relationship between incoming solar radiation and daily maximum and minimum temperature. Agr. Forest Meteorol., 31, 159-166. Donatelli, M., Bellocchi, G., 2001. Estimates of daily global solar radiation: new developments in the software RadEst3.00. Proc. 2nd Int. Symp. Modelling Cropping Syst., 16-18 July, Florence, Italy, 213-214. Donatelli, M., Bellocchi, G., Fontana, F., 2003. RadEst3.00: software to estimate daily radiation data from commonly available meteorological variables. Eur. J. Agron., 18, 369372. Donatelli, M., Campbell, G.S., 1998. A simple model to estimate global solar radiation. Proc. 5th Eur. Soc. for Agron. Congr., 28 June-2 July, Nitra, Slovak Republic, 2, 133-134. Ferrari, F., Trevisan, M., Capri, E., 2003. Predicting and measuring environmental concentration of pesticides in air after soil application. J. Environ. Qual., 32, 1623-1633. Fila, G., Donatelli, M., Bellocchi, G., 2002. Developing dynamic link libraries for agrometeorology and agricultural modelling. Proc. 7th Eur. Soc. for Agron. Congr., 15-18 July, Cordoba, Spain, 285-286. Iehlé, A., Lefèvre, M., Bauer, O., Martinoli, M., Wald, L., 1997. Meteosat: a valuable tool for agro-meteorology. Study report, Space Applications Institute, Joint Research Centre of the European Commission, Ispra, Italy. Johnson, I.R., Riha, S.J., and D.S. Wilks. 1995. Modelling daily net canopy photosynthesis and its adaptation to irradiance and atmospheric CO2 concentration. Agric. Syst., 50, 1-35. Lyon, D.J., Hammer, G.L., McLean, G.B., Blumenthal, J.M., 2003. Simulation supplements field studies to determine notill dryland corn population recommendations for semiarid Western Nebraska. Agron. J., 95, 884-891. Löwy, J., 2003. Programming .NET components. O’Reilly & Associates, California, CA, USA. Mavromatis, T., Jagtap, S.S., Jones, J.W., 2002. El NiñoSouthern Oscillation effects on peanut yield and nitrogen leaching. Climate Res., 22, 129-140. Meyer, B., 1997 Object-oriented software construction (2nd ed.). Prentice Hall, Upper Saddle River, NJ, USA. Pickering, N.B., Hansen, J.W., Jones, J.V., Wells, C.M., Chen, V.K., Godwin, D.C., 1994. WeatherMan: A utility for managing and generating daily weather data. Agron. J., 86, 332-337. Richardson, C.W., 1981. Stochastic simulation of daily precipitation, temperature, and solar radiation. Water Resour. Res., 17, 182-190. Richardson, C.W., 1982. Dependence structure of daily temperature and solar radiation. Trans. of ASAE, 25, 735-739. Rivington, M., Matthews, K.B., Buchan, K., 2002. A comparison of methods for providing solar radiation data to crop models and decision support systems. Proc. 1st biennial meeting of the Int. Environ. Modell. and Software Soc., 2427 June, Lugano, Switzerland, 3, 193-198. Spitters, C.J.T., Toussaint, H.A.J.M., Goudriaan, J., 1986. Separating the diffuse and direct component of global radiation and its implications for modeling canopy photosynthesis. Part I. Components of incoming radiation. Agric. For. Meteorol., 38, 217-229. Stine, W.B., Harrigan, R.W., 1985. Solar energy systems design. John Wiley and Sons, Inc., New York, USA. Szyperski, C., 2002. Component software (2nd ed.), AddisonWesley, UK, 589 pp. Thornton, P.E., Running, S.W., 1999. An improved algorithm for estimating incident daily solar radiation from measurements of temperature, humidity, and precipitation. Agr. Forest Meteorol., 93, 211-228. Weiss, A., Hays, C.J., Hu, Q., Easterling, W.E., 2001. Incorporating bias error in calculating solar irradiance: implications for crop yield simulations. Agron. J., 83, 1321-1326. Woodward, S.J.R., Barker, D.J., and R.F. Zyskowski. 2001. A practical model for predicting soil water deficit in New Zealand pastures. New Zeal. J. Agr. Res., 44, 91-109. 27 M. Donatelli et al. - Rivista Italiana di Agrometeorologia 24-30 (1) 2004 Tab. 1 - Variabili di input e output del componente. Per ciascun input, un numero identificativo (ID output) elenca gli output calcolati usando quello specifico input. Tab. 1 - Input and output variables of GSRad. For each input, the outputs calculated using that specific input are listed as ID output. Variabile Input i, giorno giuliano λ', latitudine τ, trasmissività del cielo sereno α', inclinazione della superficie captante γ', orientamento in senso orario rispetto al sud della superficie captante Tmaxi, massima temperatura aerea giornaliera Tmini, minima temperatura aerea giornaliera ∆Ti, intervallo giornaliero delle temperature aeree§§ ∆Tm, media mensile dei ∆Ti’ (media fissa)§§§ ∆Tw,i, media settimanale dei ∆Ti’ (media mobile)§§§§ Tavgi, media delle temperature aeree giornaliere Tnc, fattore delle temperature notturne estive ir, opzione di reverse (i or (361-i) c1, primo parametro di stagionalità c2, secondo parametro di stagionalità b, parametro radiazione-temperatura SSDi, eliofania giornaliera FF, fattore di ripartizione della intensità luminosa Zi,j, matrice delle altitudini da un modello digitale di elevazione§§§§§ s, dimensione delle celle della matrice Zi,j A, matrice di funzioni di correlazione 3x3 tra i residui di radiazione e temperatura aerea massima e minima B, matrice di funzioni di correlazione 3x3, ritardata di un giorno, tra i residui di radiazione e temperatura aerea massima e minima Unità Modello§ Output Output calcolato (ID) ° ° BC, CD, DB, JW BC, CD, DB, JW BC, CD, DB, JW BC, CD, DB, JW 1, 2, 5, 6, 7 1, 2 2 1, 2 ° BC, CD, DB, JW 1, 2 °C °C °C °C °C °C h - BC, CD, DB BC, CD, DB BC, CD, DB BC DB CD CD DB DB DB BC, CD, DB JW JW 2, 5, 6, 7 2, 5, 6, 7 2 2 2 2 2 2 2 2 2 2 2 m BC, CD, DB, JW 1, 2, 3, 4 m - BC, CD, DB, JW R 1, 2, 3, 4 2 - R 2 ID output Output RadPi(α,γ), radiazione potenziale giornaliera Radi(α,γ), radiazione solare giornaliera al suolo α', inclinazione della superficie captante (da un modello digitale di elevazione) γ', orientamento della superficie captante (da un modello digitale di elevazione) ∆Ti, intervallo giornaliero delle temperature aeree ∆Tm, media mensile dei ∆Ti’ (media fissa) ∆Tw,i, media settimanale dei ∆Ti’ (media mobile) MJ m-2 d-1 MJ m-2 d-1 1 2 ° 3 ° 4 °C °C °C 5 6 7 § BC, modello Bristow-Campbell della trasmissività; CD, modello Campbell-Donatelli della trasmissività; DB, modello DonatelliBellocchi della trasmissività; JW, modello Johnson-Woodward della radiazione solare al suolo; R, modello Richardson della generazione stocastica di radiazione solare al suolo. §§ ∆Ti=Tmaxi-0.5ּ(Tmini+Tmini+1) §§§ ∆T = m i f,m ∑ i =i0,m §§§§ ∆T = w,i ∆Ti' , dove m è un mese dell’anno, 0 è il giorno iniziale del mese, f è il giorno finale del mese, ∆Ti’=Tmaxi-Tmini i f,m - i0,m + 1 i+3 ∑ j =i − 3 §§§§§ Z i, j 28 ⎛ z 1,1 ⎜ = ⎜ z 2,1 ⎜ ⎝ z 3,1 ∆Ti' (j) , dove j è un indice del periodo di 7 giorni 7 z 1,2 z 2,2 z 3,2 z 1,3 ⎞ ⎟ z 2,3 ⎟ , dove zi,j sono i valori di altitudine sopra il livello del mare (m) ⎟ z 3,3 ⎠ M. Donatelli et al. - Rivista Italiana di Agrometeorologia 24-30 (1) 2004 Tab. 2 - Componenti per il calcolo della radiazione solare giornaliera, Radi(α,γ) (MJ m-2 d-1), per una superficie captante generica. Gli input richiesti sono riportati in Tab. 1. Tab. 2 – Components of the solar radiation equations for a generic surface. See Tab.1 for the inputs required. Variabile RadPi, radiazione potenziale giornaliera incidente su una superficie orizzontale C, costante solare di, valore inverso della distanza relativa terra-sole ssi, prodotto giornaliero seno declinazioneseno latitudine Unità 2 MJ m d Formulazione -1 MJ m-2 h-1 ( 24 ⎛ RadPi = C ⋅ d i ⋅ ⎜ ss i ⋅ H i + cc i ⋅ ⋅ 1 − x 22 π ⎝ d i = 1 + 0.00334 ⋅ cos(0.01721 ⋅ i − 0.0552) - ssi = sin(δ i ) ⋅ sin(λ ) δ i = 0.4093 ⋅ λ, latitudine rad λ=λ’ּ(π/180) h x2, secondo fattore della durata del giorno - x1, primo fattore della durata del giorno - fblue, intensità relativa della radiazione per cielo sereno fcloud, intensità relativa della radiazione per cielo coperto ci, frazione giornaliera della copertura nuvolosa f(Tavg), funzione della media giornaliera delle temperature aeree f(Tmin), funzione della temperatura minima giornaliera f(c2), funzione del secondo parametro di stagionalità MJ m2 d-1 rad rad MJ m2 d-1 rad MJ m2 d-1 H i = 12 + arcsin( x2) ⋅ 24 π x2 = min(max(x1, − 1), 1) ss x1 = i cci cci = cos(δ i ) ⋅ cos(λ ) cRadPi = C ⋅ 24 ⋅ d i ⋅ (asci ⋅ sin(γ − π ) + acci ⋅ cos(γ − π )) π α=α’ּ(π/180) γ=γ’ּ(π/180) ⎛π ⎞ asci = − cos(δ i ) ⋅ sin ⎜ ⋅ H i ⎟ ⎝ 24 ⎠ π acci = sin (δ i ) ⋅ cos(λ ) ⋅ ⋅ H i 24 2 RadS i = C ⋅ ⋅ sin(hi ) ⋅ τ 1/sin (hi ) π hi = arcsin(sin(δi ) ⋅ sin(λ ) + cos(δi ) ⋅ cos( λ )) ( ( ) 1 RadDi = C ⋅ ⋅ sin(hi ) ⋅ 1 + τ 1/sin (hi ) π ) 1 − τ 1 / sin (hi ) - f blue = - f cloud = FF ⋅ f blue - ci = 1 − - f(Tavg): 0.017ּexp[exp(-0.053ּ Tavgi)] - f(Tmin)=exp(Tmini/Tnc) - ⎟ ⎠ 180 ⎛ 2π ⎞ ⋅ sin ⎜ ⋅ (284 + i )⎟ π ⎝ 365 ⎠ rad Hi, durata del giorno 0.5 ⎞ 4.921 - δi, declinazione solare giornaliera cci, prodotto giornaliero coseno declinazione-coseno latitudine cRadPi, componente della radiazione potenziale dovuta all’orientamento della superficie captante α inclinazione della superficie captante γ orientamento della superficie captante asci, componente giornaliera del seno dell’orientamento della superficie captante acci, componente giornaliera del coseno dell’orientamento della superficie captante RadSi, componente diretta della radiazione solare giornaliera hi, altezza del sole sull’orizzonte a mezzogiorno RadDi, componente diffusa della radiazione solare giornaliera ) 1 + τ 1 / sin (hi ) SSD i Hi f(c2)=1-1.90ּ c3+3.83ּc32§ § Il parametro c3 è pari alla parte decimale di c2 29 M. Donatelli et al. - Rivista Italiana di Agrometeorologia 24-30 (1) 2004 Tab. 3 - Componenti per il calcolo dei parametri di inclinazione (α) e orientamento (γ) della superficie captante. Gli input richiesti sono riportati in Tab. 1. Tab. 3 – Components of the slope (α) and aspect (β) equations. See Tab. 1 for the inputs required. Variabile α, inclinazione della superficie captante rispetto a una superficie orizzontale fx, primo fattore dell’inclinazione fy, secondo fattore dell’inclinazione γN, orientamento in senso orario rispetto al nord della superficie captante Unità rad Formulazione α = arctan fx 2 + fy 2 - ⎧ z 2,1 − z 2,3 if z 1,1 = z 1,3 = z 3,1 = z 3,3 = 0 ⎪⎪ fx ⎨ 2 ⋅ s z + 2 ⋅ z 2,1 + z 3,1 − z 1,3 − 2 ⋅ z 2,3 − z 3,3 ⎪ 1,1 8⋅s ⎩⎪ - ⎧ z 3,2 − z 1,2 if z 1,1 = z 1,3 = z 3,1 = z 3,3 = 0 ⎪⎪ 2⋅s fy ⎨ z + 2 ⋅ z 2,3 + z 3,3 - z 1,1 - 2 ⋅ z 1,2 − z 1,3 ⎪ 3,1 ⎪⎩ 8⋅s rad if α ≠ 0 π + c a if fx < 0 2 3 ⋅ π + c a if fx > 0 2 γN = π if fy > 0 and fx = 0 0 if fy < 0 and fx = 0 if α = 0 0 γ, orientamento in senso orario rispetto al sud della superficie captante rad ca, coefficiente dei fattori di inclinazione - 30 γ − π if γ= N γ N + π if γN ≥ π γN < π ⎛ fy ⎞ c a = − arctan⎜⎜ ⎟⎟ ⎝ fx ⎠ M. Rinaldi e G. Rana - Rivista Italiana di Agrometeorologia 31-35 (1) 2004 I FABBISOGNI IDRICI DEL POMODORO DA INDUSTRIA IN CAPITANATA WATER USE OF PROCESSING TOMATO IN THE CAPITANATA REGION (SOUTHERN ITALY) Michele Rinaldi* e Gianfranco Rana Istituto Sperimentale Agronomico - via C. Ulpiani, 5 70125 Bari Tel +39 080 5475016 Fax: +39 080 5475023 Riassunto Il pomodoro da industria (Lycopersicon esculentum Mill.) è un’importante coltura irrigua del Sud Italia e della Capitanata in particolare. Gli elevati volumi stagionali di adacquamento impongono una razionalizzazione della pratica irrigua, che passa anche attraverso una verifica ed una più attenta stima dei coefficienti colturali (Kc = ETm/ETo). Nella presente ricerca, tale verifica è stata attuata misurando l’evapotraspirazione massima (ETm) con il metodo dell’”Eddy covariance”, utilizzando, tra l’altro, un anemometro sonico per la misura del flusso di calore sensibile. Le prove sono state condotte su pomodoro nel 2000 (varietà a bacca lunga, da pelato) e nel 2001 (varietà a bacca tonda, da concentrato), su ampi appezzamenti omogenei nella provincia di Foggia. L’evapotraspirazione di riferimento (ETo) è stata stimata con la formula di Penman-Monteith e, in alternativa, con quella di Hargreaves che richiede solo le temperature massima e minima, previa una sua calibrazione locale. I Kc ottenuti, confrontati con quelli suggeriti dalla FAO, evidenziano valori maggiori nella fase intermedia (+0.13), specie nel pomodoro a bacca lunga. La calibrazione locale della formula di Hargreaves ha consentito di migliorare molto la stima dell’ETo. I fabbisogni idrici così determinati (500 e 550 mm, rispettivamente nelle due annate) sono stati in accordo con l’acqua effettivamente erogata dagli agricoltori, con efficienze del metodo irriguo (a microportata) variabili tra l’83 e l’85%. Parole chiave: Coefficienti colturali, evapotraspirazione, Hargreaves, anemometro sonico, Eddy covariance. Abstract Processing tomato (Lycopersicon esculentum Mill.) is an important irrigated crop in Southern Italy and, mainly, in the Capitanata plain. High values of seasonal irrigation water force the farmer to a more rational water management, that can be obtained by means of a more precise estimation and evaluation of crop coefficients (Kc = ETm/ETo). In this research Kc estimation was carried out by measuring maximum evapotranspiration (ETm) with ”Eddy covariance” method using, among the others, a sonic anemometer to measure sensible heat flux. The field experiments were carried out with “square shape” tomato (for peeled) in 2000 and “deep globe shape” tomato (for paste) in 2001, placing the instruments in the center of large homogenous fields (> 10 ha) in the Foggia plain (Capitanata). Reference evapotranspiration (ETo) was estimated by means of the Penman-Monteith formula and, in addition, with Hargreaves equation requiring maximum and minimum daily temperatures only and a local calibration. Experimentally measured and FAO suggested crop coefficients were compared: peeled tomato, in the intermediate phase, showed Kc greater of 0.13 than that reported by FAO. Local calibration of Hargreaves equation allowed a significant improvement of ETo estimation. Seasonal water use was estimated (500 and 550 mm, in 2000 and in 2001, respectively) and resulted in accordance to the actual farmer irrigation supply, taking into account an irrigation efficiency of drip irrigation of about 83-85%. Keywords: crop coefficient, evapotranspiration, Hargreaves, sonic anemometer, Eddy covariance. Introduzione La Capitanata (provincia di Foggia) è una delle aree più vocate all'agricoltura di qualità e alla coltivazione del pomodoro in particolare, con circa 28.000 ha investiti mediamente ogni anno nel periodo 1998-2003. I fabbisogni idrici di questa coltura altamente produttiva sono elevati e, considerando gli scarsi apporti di pioggia, essi devono essere soddisfatti quasi completamente con l'irrigazione: 500-600 mm di acqua sono forniti, in media, *Corresponding author. tel. +39 080 5475046 fax. +39 080 5475023 e-mail: [email protected] (M.Rinaldi). per l'irrigazione di questa coltura con punte di 800 mm in funzione dell'annata e della varietà. Attualmente, pertanto, risulta di particolare interesse pratico lo studio e la determinazione accurata e puntuale dei consumi idrici del pomodoro in aree meridionali a scarsa disponibilità idrica, per una gestione, il più razionale possibile, dell'acqua per irrigazione. In quest’articolo sono presentati i risultati sperimentali di una prova svolta in Capitanata dai ricercatori dell'Istituto Sperimentale Agronomico di Bari, con l'obiettivo di determinare i consumi idrici stagionali del pomodoro da industria. M. Rinaldi e G. Rana - Rivista Italiana di Agrometeorologia 31-35 (1) 2004 Un altro obiettivo dell'esperimento è quello di studiare il metodo più idoneo per la stima dell'evapotraspirazione reale della coltura (ETc) da utilizzare nella gestione dell'acqua d'irrigazione nell'area in oggetto, sia a livello aziendale che comprensoriale. Per scopi pratici legati alla gestione efficiente dell’irrigazione, la determinazione dei fabbisogni irrigui di una coltura necessita di misure meteorologiche e avviene in due fasi successive: innanzitutto si determina il potere evaporativo dell'ambiente calcolando l'ET di riferimento (ETo); tale valore si moltiplica quindi per un coefficiente colturale empirico (Kc) che tiene conto delle caratteristiche della pianta, ottenendo così l'evapotraspirazione effettiva della coltura (Doorenbos e Pruitt, 1977), che coincide con l’evapotraspirazione massima della coltura (ETm) nel caso di condizioni idriche non limitanti. Questi coefficienti sono stati determinati oltre 20 anni fa (con tecniche di coltivazione e varietà anche diverse dalle attuali) e spesso in zone differenti da quelle in cui vengono applicati; da qui ne discende la necessità di una loro verifica, anche in considerazione del fatto che possono portare a situazioni di stress idrico, se sono stati sottostimati, e a sprechi di acqua se, invece, sono stati sovrastimati. Con queste informazioni, sarà possibile razionalizzare la pratica irrigua del pomodoro da industria, le cui produzioni, sia dal punto di vista qualitativo che quantitativo, possono essere limitate sia da deficit che eccessi irrigui. Materiali e Metodi Le prove sperimentali Sono state eseguite due prove: una nel 2000 su pomodoro a bacca allungata (da pelato) e una nel 2001 su pomodoro a bacca rotonda (da concentrato). Nel primo anno la prova si è svolta presso l'azienda agricola dei f.lli Zerillo, in agro di Manfredonia (FG) località Cutino, il cui terreno è di medio impasto tendente al limoso. Il pomodoro da industria a bacca allungata (cv Pull) è stato trapiantato tra il 14 e il 22 aprile 2000, a file binate distanti tra loro 50 cm e 180 cm tra le bine, con un investimento di circa 2.5 piante m-2. Il metodo irriguo a microportata prevedeva ali gocciolanti lungo la bina, con gocciolatori autocompensanti da 4 L ora-1; l'acqua prelevata da un serbatoio attiguo era distribuita ad ogni settore con cadenza bisettimanale, con quantitativi variabili da 20 a 30 mm, secondo i consumi calcolati con l’equazione di Penman-Monteith. A questi regimi idrici la coltura era nello stato idrico ottimale per cui l’ETc può essere considerata l’evapotraspirazione massima. La raccolta meccanica del pomodoro è iniziata il 20 luglio e la produzione finale si è attestata sulle 100 t ha-1 di bacche commercializzabili. Nel 2001 la prova si è svolta presso l'azienda agricola del Dr. Vincenzo Fratta, in agro di Foggia, località Cunicella. Il terreno aveva le stesse caratteristiche medie del terreno precedentemente descritto. Il pomodoro da industria a bacca rotonda (ibrido PS 1296) è stato trapiantato tra il 2 e il 9 maggio 2001. Il sesto d'impianto e le modalità d'irrigazione erano identiche a quelle della prova precedente e riflettono, in media, il comportamento degli agricoltori della Capitanata nei riguardi del pomodoro da in- 32 dustria. In questo caso la raccolta meccanica del pomodoro è iniziata il 27 luglio ed ha reso circa 110 t ha-1 di bacche. I volumi irrigui sono stati misurati con un contatore posto all’ingresso del campo. I metodi di misura dei consumi idrici L'ETc della coltura è stata misurata con un metodo micrometeorologico d'avanguardia, il cosiddetto metodo “Eddy covariance” (Kaimal et al., 1972; Tanner et al., 1985), come ETm = ETc = Rn − G − H λ (1) dove Rn (W m-2) è la radiazione netta misurata direttamente con un net-radiometro tipo Fritschen (REBS Q*6, USA), G (W m-2) è il flusso di calore nel suolo misurato direttamente con tre piastre di flusso (Campbell Scientific, USA) poste a 3 cm nel terreno, H (W m-2) è il flusso di calore sensibile misurato direttamente con un anemometro sonico (USA-1 Metek, Germany). ET è in mm con λ calore latente di evaporazione dell'acqua in J kg-1. I sensori sono stati installati al centro del campo, di dimensione superiore ai 10 ettari in entrambi gli anni, il giorno dopo la fine della fase di trapianto della coltura. I dati che presentiamo sono, dunque, relativi all'intera stagione colturale. Il metodo della “Eddy covariance” necessita, come si vede, di sensori complessi e difficili da gestire, per cui è un metodo di misura dell'evapotraspirazione reale ancora ristretto all'ambito della ricerca. Per applicazioni pratiche, il metodo più idoneo per stimare l'evapotraspirazione reale è quello consigliato dalla FAO (Allen et al., 1998), in cui l'ETm si calcola come: ETm = Kc ⋅ ET0 (2) con Kc coefficiente colturale ed ETo evapotraspirazione di riferimento giornaliera, in mm. Il metodo ideale per la stima di quest'ultima grandezza è il modello di PenmanMonteith (Allen et al., 1998): ET0 = 0.408 ⋅ (Rn − G ) + γ ⋅ 900 ⋅ u ⋅ (e s − e a ) T + 273 ∆ + γ ⋅ (1 + 0.34 ⋅ u ) (3) dove ∆ è la pendenza della curva pressione vapore (kPa °C-1), γ è la costante psicrometrica (kPa °C-1), T è la temperatura dell'aria (°C), u è la velocità del vento a 2 m di altezza dal suolo (m s-1), es è la pressione di vapore saturo dell'aria (kPa), ea è la pressione di vapore reale dell'aria (kPa). Nel nostro caso i parametri in input al modello sono stati prelevati nel 2000 dalla stazione agrometeorologica di Tavernola, a 3 km dal sito sperimentale (rete agrometeorologica del Consorzio di Bonifica della Capitanata) e nel 2001 dalla stazione agrometeorologica curata della Syngenta S.p.A. e sita nell'azienda stessa. M. Rinaldi e G. Rana - Rivista Italiana di Agrometeorologia 31-35 (1) 2004 Alternative alla Penman-Monteith Poiché il modello di Penman-Monteith (3) ha bisogno di numerosi parametri in input, dal punto di vista pratico riveste particolare importanza il poter avere a disposizione una formula alternativa a questa, più semplice ed operativa. Seguendo sempre i suggerimenti di Allen et al. (1998) qui si propone di analizzare le performance del modello di Hargreaves e Samani (1985). Tale modello necessita della sola temperatura minima e massima giornaliera: ET0 = a ⋅ (Tmedia + b ) ⋅ (Tmax − Tmin ) Ra c (5) con Ra radiazione solare massima teorica per il giorno in questione, funzione della sola latitudine del luogo, tabulata o ricavabile dall'equazione (21) sul quaderno 56 della FAO (Allen et al., 1998). Nella formulazione originale i coefficienti “a”, “b” e “c” valgono rispettivamente 0.0023, 17.8 e 0.5, ma è consigliata una ricalibrazione per tener conto delle caratteristiche pedoclimatiche locali. Noi abbiamo ricalcolato questi coefficienti. radiazione disponibile evapotraspirazione reale mm/giorno 8 6 4 2 0 9 apr 29 apr 19 m ag 8 giu 28 giu 18 lug 7 ago Fig. 1 - Evapotraspirazione reale e radiazione disponibile giornaliere durante il ciclo colturale del pomodoro a bacca lunga nel 2000. Fig. 1 - Daily crop evapotranspiration and available radiation during “square shape” tomato crop cycle in 2000. Kc sperimentale medio Kc teorico (FAO 56) Kc sperimentale 1.5 Kc=(ETm/ET0) Il coefficiente colturale Dalla descrizione precedente sui metodi di stima dell'ETc risulta evidente il ruolo chiave del coefficiente colturale Kc. Infatti, mentre l'evapotraspirazione di riferimento, dipendendo esclusivamente da parametri climatici, rappresenta la domanda evapotraspirativa dell'atmosfera, il Kc è un coefficiente empirico che deve tenere conto sia delle caratteristiche colturali (architettura della pianta, comportamento idrico) ma che dipende anche dal pedoclima locale (Rana e Katerji, 2000). Il Kc si trova tabulato in letteratura (p. es. Allen et al., 1998) e può anche essere corretto per il clima locale (Kc “single”). In questo lavoro abbiamo voluto confrontare il Kc “single” dato dalla letteratura per il pomodoro coltivato in ambiente mediterraneo e il Kc realmente misurato in campo come: ETm (4) Kc = ET0 1.0 0.5 9 m ag 29 m ag 1 8 giu 8 lug 28 lu g 17 a go 7 set Fig. 2 - Valori del coefficiente colturale (Kc) del pomodoro a bacca lunga nel 2000. Fig. 2 - Daily crop coefficients (Kc) of “square shape” tomato crop in 2000. Risultati Kc sperimentale medio Kc teorico (FAO 56) Kc sperimentale 1.5 Kc=ETm/ET0 In figura 1 sono mostrati gli andamenti giornalieri di evapotraspirazione reale del pomodoro per l’anno 2000 (pomodoro a bacca allungata). Si può osservare che durante il primo periodo, quello di crescita attiva delle piante (da 25 a 30 giorni dal trapianto) l'ETc rappresenta il 70% circa della radiazione disponibile in quanto la coltura copre il suolo solo per il 50% circa (LAI inferiore a 1). A fioritura avvenuta (da 30 a 60 giorni dal trapianto), in presenza di un LAI compreso tra 1 e 2, l'ETc rappresenta circa il 90% della radiazione disponibile. Nella fase finale l'ETc è circa il 50 % della radiazione, segno che la pianta inizia a chiudere gli stomi e ad invecchiare. Analogo comportamento ha mostrato nell’anno successivo il pomodoro a bacca rotonda. 1.0 0.5 29 ap r 19 m ag 8 giu 2 8 giu 1 8 lug 7 ag o Fig. 3 - Valori del coefficiente colturale (Kc) del pomodoro a bacca tonda nel 2001. Fig. 3 - Daily crop coefficients (Kc) of “deep globe shape” tomato crop in 2001. 33 M. Rinaldi e G. Rana - Rivista Italiana di Agrometeorologia 31-35 (1) 2004 Nelle Figure successive, sono riportati i valori di Kc per il pomodoro a bacca allungata (Fig. 2) e per il pomodoro a bacca rotonda (Fig. 3). Sono rinvenibili (i) alcuni vaTab. 1 - Valori dell’evapotraspirazione (mm) e dei volumi stagionali irrigui del pomodoro nei due anni di prova. lori giornalieri del Kc sperimentale ottenuto con la relazione (3), (ii) i valori medi del coefficiente calcolati con delle interpolazioni lineari dei dati e (iii) i valori del Kc teorico così come suggeriti del quaderno 56 della FAO nella modalità “single crop coefficient”. Come si vede dalle figure 2 e 3, i valori di Kc trovati sperimentalmente sono differenti da quelli teorici tabulati: nella fase intermedia sono superiori di 0.13 per il pomodoro a bacca allungata e di 0.03 per il pomodoro a bacca rotonda. Quest’osservazione evidenzia come il pomodoro a bacca allungata consumi mediamente il 1015% più del pomodoro a bacca rotonda. Se si utilizza il metodo di Hargreaves per la stima dell'evapotraspirazione di riferimento si constata una sovrastima (del 150%) se vengono utilizzati i coeffcienti “a”, “b” e “c” standard, sia per il 2000 che per il 2001. Una ricalibrazione della formula di Hargreaves fornisce i seguenti risultati: Agro di Manfredonia (anno 2000): ET0 = 0.002 ⋅ (Tmedia + 14) ⋅ (Tmax − Tmn ) 0.45 Ra (6) Agro di Foggia (anno 2001): ET0 = 0.0018 ⋅ (Tmedia + 8) ⋅ (Tmax − Tmn ) 0.31 Ra (7) I risultati complessivi della prova sono riportati in Tabella 1: nella prima colonna (acqua data dall'agricoltore) è riportata la quantità d'acqua fornita realmente alla coltura; nella seconda colonna (ET misurata) è riportato il valore di evapotraspirazione reale, misurata con il metodo "Eddy covariance"; nella terza colonna viene calcolata l'efficienza irrigua (rapporto tra evapotraspirazione reale della coltura e acqua fornita) e nella quarta vi è il valore teorico stimato dalla relazione ETm=Kc*ETo, con Kc tabulato nel quaderno 56 della FAO (metodo “single crop coefficient”) ed ETo stimato con la PenmanMonteith. Tab. 1 - Valori dell’evapotraspirazione (mm) e dei volumi stagionali irrigui del pomodoro nei due anni di prova. Tab. 1 - Evapotranspiration values (mm) and seasonal irrigation depths of tomato in the two years of experiment. Anno 2000 2001 34 Acqua data dall' agricoltore 600 650 ET misurata (Eddy covariance) 500 551 Efficienza irrigua (%) 83.3 84.8 ETm ETm con (FAO-56) ETo da Single crop Hargreaves coefficient calibrata method localmente 442 475 500 550 E’ interessante notare come il metodo ETm “single crop coefficient” sottostima di circa 58 mm i consumi idrici del pomodoro. Dai dati riportati si rileva che l'efficienza dell'irrigazione è di circa l'83% per il 2000 e 85% per il 2001, un valore vicino al massimo teorico (90%) possibile con il sistema d'irrigazione a microportata di erogazione (Jensen et al., 1990). Se si calibra il modello di Hargreaves, dando un valore locale ad hoc ai coefficienti “a”, “b” e “c” (eq. 6 e 7) si ottiene un miglioramento notevole delle prestazioni, con un' ET reale molto più vicina a quella misurata (Tabella 1). Conclusioni Queste due campagne di misura ci hanno permesso di applicare i risultati della ricerca sviluppata negli ultimi decenni nel campo della stima dei consumi idrici di una coltura. Abbiamo dimostrato che se si adotta un coefficiente colturale appropriato, avendo a disposizione i dati meteorologici rilevati in siti vicini a quello in oggetto, si riesce ad ottenere un valore di consumo idrico del pomodoro molto vicino a quello reale. Tale dato di evapotraspirazione potenziale è indispensabile al calcolo del fabbisogno idrico finalizzato ad una gestione razionale dell'acqua d'irrigazione. In realtà il Kc adottato qui per ottenere i migliori risultati è quello calibrato in campo, a partire da dati sperimentali piuttosto difficili da ottenere di routine. Vi è però un metodo, descritto esaurientemente da Allen et al. (1998) e da Allen (2000), per ricavare un Kc secondo una modalità chiamata "dual crop coefficient", in cui si calcola separatamente l'acqua evaporata dal suolo e l’acqua traspirata e che migliora notevolmente la stima dell'ET reale, senza dover ricorrere ad una calibrazione locale del coefficiente colturale. Nel nostro caso l'adottare tale Kc porta ad una stima dell'ET reale di un valore di appena 10 mm lontano da quello realmente misurato, sia per il pomodoro a bacca allungata (2000) che per quello a bacca rotonda (2001). Questa strada è dunque molto promettente e va senz'altro perseguita e analizzata in ricerche applicate del prossimo futuro. L’importanza di avere i dati giornalieri di temperatura minima e massima, scaturisce dalla possibilità di adottare formule per la stima dell’ETo che si avvicinano con buona approssimazione a quelle di Penman-Monteith, universalmente riconosciuta come la più affidabile, ma che richiede altri parametri non sempre disponibili (radiazione solare, umidità dell’aria, velocità del vento). Con la formula di Hargreaves, utilizzando opportuni coefficienti (previa calibrazione) come calcolati in que-sto lavoro (formule 6 e 7), è possibile avere un’adeguata stima dell’ETo e di conseguenza dell’ETc. Va posto l’accento come in questo studio le quantità di acqua fornite con l'irrigazione si sono avvicinate molto ai fabbisogni irrigui calcolati con le diverse formule: questo dipende, a nostro avviso, da almeno due ragioni. Innanzitutto, l'esperienza tecnica degli imprenditori della Capitanata nella gestione dell'irrigazione a microportata del pomodoro, coltura da oltre 20 anni entrata a pieno titolo negli ordinamenti irrigui della zona e sistema irriguo or- M. Rinaldi e G. Rana - Rivista Italiana di Agrometeorologia 31-35 (1) 2004 mai largamente diffuso anche per la possibilità di effettuare la fertirrigazione. In secondo luogo, il costo dell'acqua, sia nelle aree servite dal Consorzio per la Bonifica della Capitanata di Foggia (la tariffazione non è solo funzione della superficie servita, come in altre realtà, ma prevalentemente in base ai consumi reali e con tariffe crescenti con i consumi), che per quelle in cui l'acqua deve essere sollevata da falde freatiche e spinta in pressione nell'impianto irriguo: questo costo, che oscilla tra i 9 e i 24 centesimi di € per m3, porta ad un utilizzo più razionale di questa risorsa, sempre più preziosa, specie negli ambienti meridionali. Ringraziamenti Questo lavoro è stato interamente finanziato dal POM B19 (Coordinatore Prof. Luigi Postiglione). Si ringraziano i f.lli Zerillo e il Dr. Fratta per aver ospitato la prova. I rilievi sperimentali di campo sono stati cura dal Sig. Michele Introna dell'Istituto Sperimentale Agronomico di Bari. Si ringraziano, inoltre il Consorzio di Bonifica della Capitanata e la Syngenta S.p.A. per aver fornito i dati agrometeorologici per il calcolo dell'evapotraspirazione di riferimento. Bibliografia Allen, R.G., 2000. Using the FAO-56 dual crop coefficient method over an irrigated region as part of an evapotranspiration intercomparison study. J. Hydrol., 229, 27-41. Allen, R.G., Pereira, L.S., Raes, D., Smith, M., 1998. Crop evapotranspiration. Guidelines for computing crop water requirements. Irrigation and Drainage Paper 56, Food and Agric. Organization of the United Nations, Rome, Italy, 300 pp. Doorenbos, J., and W. O. Pruitt. 1977. Crop water requirements. Irrigation and Drainage Paper 24, Food and Agric. Organization of the United Nations, Rome, Italy, 144 pp. Hargreaves, G. H., Samani, Z. A., 1985. Reference crop evapotranspiration from temperature. Applied Engign. In Agric., 1 (2), 96-99. Jensen, M.E., Burman, R.D. and Allen, R.G. (eds), 1990. Evapotranspiration and Irrigation Water Requirements. Manuals and Reports on Engineering. Practice No. 70, ASCE, New York, NY, 332 pp. Kaimal, J.C., Wyngaard, J.C., Izumi, Y., Coté. O.R., 1972. Spectral characteristics of surface layer turbulence. Q. J. Roy. Metereol. Soc., 98, 563-589. Rana, G., Katerji, N., 2000. Measurement and estimation of actual evapotranspiration in he field under mediterranean climate. A review. Eur. J. Agron., 13, 125-153. Tanner, B.D., Tanner, M.S., Dugas, W.A., Campbell, E.C., Blad, B.L., 1985. Evaluation of an operational eddy correlation system for evapotranspiration measurements. In: Advance in Evapotranspiration, December 16-17, ASAE, Chicago, IL, 87-99. 35 B. Basso et al. - Rivista Italiana di Agrometeorologia 36-53 (1) 2004 REMOTELY SENSED VEGETATION INDICES: THEORY AND APPLICATIONS FOR CROP MANAGEMENT INDICI DI VEGETAZIONE TELERILEVATI: TEORIA ED APPLICAZIONI PER LA GESTIONE AGRONOMICA DELLE COLTURE Bruno Basso* 1, Davide Cammarano1, Pasquale De Vita2 1 Department of Cropping Systems, Forestry and Environmental Sciences. University of Basilicata, Via Ateneo Lucano, 10. Potenza 85100. Italy 2 ENEA C.R. Trisaia. S.S. 106 Jonica Km 419,5 75026 Rotondella (MT) Abstract Remote sensing of soil and crop can be an attractive alternative to the traditional methods of field scouting because of the capability of covering large areas rapidly and repeatedly providing spatial and temporal information necessary for a sustainable soil and crop management. The potential of remote sensing in agriculture is very high because it is able to infer about soil and vegetation amount as a non-destructive mean. Numerous spectral vegetation indices (VIs) have been developed to characterize vegetation canopies. Plant canopy reflectance factors and derived multispectral VIs are receiving increased attention in agricultural research as robust surrogates for traditional agronomic parameters. Spectral reflectance and thermal emittance properties of soils and crops have been used extensively to predict ecological variables, such as percent vegetation cover, plant biomass, green leaf area index and other biophysical characteristics. VIs are strongly modulated by interactions of solar radiation with photosynthetically active plant tissues and thus also are indicative of dynamic biophysical properties related to productivity and surface energy balance. Recent advances on the resolution and availability of remote sensing imagery, coupled with a decrease in its associated costs, have allowed the collection of timely information on soil and crop variability by examining spatial and temporal patterns of vegetation indices. Precision agriculture applications rely on some form of VIs to quantify spatial variability within a field. The objective of this paper is to describe the biophysical principles of vegetation indices and to present a review of remote sensing applications for crop management. The paper first describes the techniques and capabilities of remote sensing then presents a series of novel and practical applications of different types of vegetation indices in agricultural research. The future challenges and opportunities section highlights the benefits and limitations of vegetation indices and remote sensing application in agriculture as well as the integration with decision support system and management-based crop simulation models. Keywords: remote sensing, vegetation indices, crop management, spatial variability Riassunto Il telerilevamento viene utilizzato in agricoltura come mezzo non distruttivo per la stima delle condizioni delle colture nello spazio e nel tempo. Dallo studio della riflettanza spettrale della vegetazione sono state definite delle relazioni quantitative tra la fenologia della coltura ed i dati telerilevati elaborate in indici di vegetazione (IV). Numerosi studi condotti in questi ultimi anni hanno evidenziato la possibilità di utilizzare il telerilevamento per la stima di parametri agronomici tradizionali come ad esempio l’indicie di area fogliare (LAI), la percentuale di copertura vegetale, la biomassa ed altri parametri biofisici. Gli IV risultano particolarmente sensibili a tre fattori esterni: l’effetto del sole; il background del suolo e gli effetti atmosferici. Il recente progresso scientifico e tecnologico associato ad una riduzione dei costi, consente di ottenere informazioni tempestive sulle colture mediante un’analisi spazio-temporale degli indici di vegetazione. L’uso degli indici di vegetazione rivesta un’importanza notevole nel contesto di applicazioni di agricoltura di precisione per la determinazione della variabilità spaziale delle produzioni. Il presente lavoro descrive i principi biofisici degli indici di vegetazione e riporta una sostanziale rassegna bibliografica sulle applicazioni del telerilevamento per gestione ottimale del suolo e delle colture. L’articolo, prima analizza le tecniche e le potenzialità del telerilevamento, poi prosegue con la descrizione di una serie di approcci innovativi di indici di vegetazioni in agricoltura. La sezione delle prospettive future evidenzia i benefici ed i limiti delle applicazioni degli indici di vegetazione e descrive l’integrazione del telerilavamento con i modelli di simulazione ed i sistemi di supporto alle decisioni. Parole chiave: telerilevamento, indici di vegetazione, gestione agronomica, variabilità spaziale *Corresponding author tel. + 39 0971205386 fax. +39 0971205378 e-mail: [email protected] (B. Basso). B. Basso et al. - Rivista Italiana di Agrometeorologia 36-53 (1) 2004 Introduction Agricultural practices determine the level of food production and, to a great extent, the state of the global environment. Agriculture production strategies have changed significantly over the last few years because of economic decisions to reduce inputs and maximize profits and by environmental guidelines to achieve a better a safer and more efficient use of agricultural chemicals. Current technologies available to farmers allow them to select the most profitable management strategy spanning from timing of planting (anticipated or delayed planting based on El Nino or La Nina years to cultivar selections (GMOs), from adopting conservation practices for selling carbon credits to industry to a variable rate applications through precision agriculture. Agricultural production systems are inherently variable due to spatial variation in soil properties, topography, and climate are spatially variable. To achieve the ultimate goal of sustainable cropping systems, variability must be considered both in space and time because the factors influencing crop yield have different spatial and temporal behavior. Advances in technologies such as Global Positioning Systems (GPS), Geographic Information Systems (GIS) and remote sensing have created the possibility to assess the spatial variability present in the field and manage it with appropriate site-specific practices. Remote imagery, an old technology that has recently become widely available through small commercial vendors and advances in satellite capabilities, also confirms large differences in canopy development patterns that lead to yield variability. Thus, producers and researchers alike are inundated with evidence of yield variability. However, evidence of producers developing innovative management strategies that capitalize on variability has been limited. The objective of this paper is to describe the biophysical principles of vegetation indices and to present a review of remote sensing applications for crop management. The paper first describes the techniques and capabilities of remote sensing then presents a series of novel and practical applications of different types of vegetation indices in agricultural research. The discussion highlights the benefits and limitation of vegetation indices and remote sensing application in agriculture as well as the integration with decision support system and simulation models. Remote Sensing Techniques and Capabilities Remote sensing is the science and art of obtaining information about an object through the analysis of data acquired by a device that is not in contact with the object (Lillesand and Keifer, 1994). Remotely sensed data can be of many forms, including variations in force distribution, acoustic wave distribution or electromagnetic energy distributions and can be obtained from a variety of platforms, including satellite, airplanes, remotely pilot vehicles, handheld radiometers or even bucket trucks. They may be gathered by different devices, including sensors, film camera, digital cameras, video recorders. Our eyes acquire data on variations in electromagnetic radiations. Instruments capable of measuring electromagnetic radiation are called sensors. Sensors can be differntiated in : Passive sensors: without their own source of radiation. They are sensitive only to radiation from a natural origin. Active sensors: which have a built in source of radiation. Examples are Radar (Radio dection and ranging) and Lidar ( Light detection and ranging). This can be analogue (photography) or digital (multispectral scanning, thermography, radar). The elements of a digital image are called resolution cells (during the data acquisition) or pixels (after the image creation). The implementation of remote sensing data by the user requires some knowledge about the technical capabilities of the various sensor systems. The technical capabilities of the sensor systems can be listed in three resolutions: • Spatial resolution: concerns the size of the resolution cell on the ground in the direction of the flight and across. The size of the pixel determines the smallest detectable terrain feature. • Spectral resolution: concerns the number, location in the electromagnetic spectrum and bandwidth of the specific wavelength bands or spectral bands. This resolution differs from sensor to sensor and largely determines the potential use of the sensor system. • Temporal resolution: concerns the time lapse between two successive images of the same area. This primarily determned by the platform used, and secondly by the atmospheric conditions. Fig. 1 - Spectral signature of soil (wet and dry) and wheat crop Fig. 1 – Firma spettrale di un suolo (umido e asciutto) e di una coltura di frumento Biophysical principles of remote sensing in agriculture The potential of remote sensing in agriculture is very high because multispectral reflectance and temperatures of the crop canopies are related to two important physiological processes: photosynthesis and evapotranspiration. Much research has been carried out with the goal of inferring vegetation amount from remote sensing. Chlorophyll pigment absorbs mainly in the Blue and Red part of the electromagnetic spectrum and reflects the green (Chappelle et al., 1992). Near-infrared (NIR) radiation is reflected from the structure of the spongy 37 B. Basso et al. - Rivista Italiana di Agrometeorologia 36-53 (1) 2004 Fig. 3 - Primary absorption bands and factors responsible for changes in the spectral reflectance curve. Fig. 3 – Bande primarie di assorbimento e fattori responsabili dei cambi della curva di riflettanza spettrale mesophyll tissue and cavities within the leaf. Therefore the percentage of radiation reflected from the leaf will be higher in the NIR than in the Green (Gausman et al., 1969, Gausman, et al., 1971). This spectral behaviour is useful to assess plant vigour and to separate canopy from bare soil (Fig.1, Fig.2 pag 48). Furthermore the discrimination of vegetation classes is possible using NIR reflectance due to the different NIR reflectance among plant species (Fig.3). The spectral behaviour of the leaf changes during senescence and in plants subjected to stress (e.g disease, pest, N shortage) by reflecting more Red light and absorbing more NIR. Opposite behaviour is shown in healthy plants with high values of reflectance in the NIR region and low values in Red portion (Gausman et al, 1977a; Gausman, 1981; Pinter et al., 2003). Soil reflects low in the blue, and its reflectance properties increase monotonically in the visible and NIR regions of the spectrum (Price, 1990; Rondeaux et al. 1996) . Spectral properties of the soil, however, depends by soil constituents such as soil organic matter, iron oxides and soil water, and soil roughness such as particle and aggregate size (Rondeaux et al. 1996). High soil water and high organic matter contents show lower reflectance while soils with low water content and smooth surface 38 tend to be brighter (Daughtry, 2001). In the presence of iron oxides soil reflectance is higher in the red portion of the spectrum. Crop residues on soil surface also causes variation in reflectance compared to bare soil and partial canopy cover (Daughtry et al., 1996; Nagler et al., 2000; Barnes et al., 2003). For a given type of soil variability, the soil reflectance (ρ) at one wavelength is often functionally related to the reflectance in another wavelength (Jasinki and Eagleson, 1989; Rondeaux et al. 1996). So the relationship between two wavelength (λ1, λ2 ) can be expressed as follow: ρ(λ2) = a ρ(λ1) + b (1) The slope a and intercept b are dependent on the both wavelength and the type of variability. The relationship shown above yield a line called soil line, defined as an hypothetical line in spectral space that describes the variation in the spectrum of bare soil in the image. Other lines that are important in the developed VIs are Vegetation isoline and index isoline. Vegetation isoline are formed by a set of reflectance points representing the same optical and structural properties of the canopy, so that have constant leaf reflectance, leaf transmittance, B. Basso et al. - Rivista Italiana di Agrometeorologia 36-53 (1) 2004 single green leaves measured under controlled light conditions (Pinter et al., 2003). Although the leaf reflectance signature may be the same during the season, the dynamic proportion of canopy/soil affects the final values of canopy spectral reflectance. 50 45 40 Reflectan ce (%) 35 30 25 20 Estimation of Vegetation Properties 15 Use of remote sensing in agricultural management has been indirect for many years, beginning with mapping the soil resource on aerial photographs as early as 1929 (Bauer, 1975 ). Soils delineated according to landscape patterns shown on aerial photographs had properties that caused differences in productivity and agricultural inputs requirements. Aerial photographs filtered at critical wavelengths showed tone patterns related to stress in crop plants (Colwell 1974; Bauer, 1975). When the satellite technology became available, analysis of remotely sensed data was, and still is, the objective of many multidisciplinary research programs around the world. Examples of an airborne false color composite imagery is shown in Fig. 4 (pag.48) . 10 5 0 400 550 700 850 1000 1150 1300 1450 1600 Wavelength (nm ) Barley full cover N o W eed s Durum Wheat Lo w Weeds D urum w heat High Weeds Durum Wheat M id- Weed s D urum Wheat N o W eeds Fig. 5 – Spectral reflectance of barley, durum wheat and different level of pappy infestation in Puglia, Italy. Fig.5 – Riflettanza spettrale di orzo, frumento duro e diversi livelli di infestazione di papavero LAI and percentage of green cover, but different brightness conditions. An index isoline, shows any point on the line with the same index value (Yoshioka et al., 2000). Huete and Jackson (1988) have studied the variation of VIs against variation in canopy background brightness. In fact if the noise effect is low the assessment of VIs will permit to improve estimation of vegetation parameters. Therefore Qi et al., (1994) have used vegetation isoline to reduce noise effect. Yoshioka et al. (2000) noticed that the use of vegetation isolines is essential because VIs should yield constant values of all reflectance points on the same vegetation isoline, constant vegetation isoline with different backgrounds. Three important properties are useful in the use of vegetation isoline to design VIs: • The intersection between soil line and vegetation isoline (as a function of LAI) • The isoline slope • The intercept (NIR-intercept) Baret and Guyot (1991), Qi et al.(1994) and Huete (1990) pointed out that the slope of a vegetation isoline tends to increase exponentially with LAI; the intercept shows the inverse behaviour of the slope and the intersection between the two lines occur generally in the third quadrant and tend to reach the first quadrant as the LAI increase. Not all soils are alike. Different soils have different reflectance spectra. As discussed above, all of the vegetation indices assume that there is a soil line, where there is a single slope in Red-NIR space. Normally the assumption is that some VIs are obtained by the convergence of the two lines at the origin or that isoline and soil line are parallel. Soils show different Red-NIR slopes in a single image, thus the assumption proposed above is not exactly right and changes in soil conditions will give incorrect information in vegetation index. Soil noise is highly significant when vegetation cover is low. The spectral signature of crop canopies in the field are more complex and often quite different from those of Vegetation monitoring is usually accomplished by simple regression approach, modeling approach using remote sensing data and by computing vegetation indices (VIs). Simple Regression Approach A simple regression approach can be applied because the reflectance in the red spectral region decreases while that in the near-infrared (NIR) region increases when the vegetation density (LAI) increases. The relation between LAI and spectral reflectance can be obtained by simply regress ground measured LAI and the surface reflectance. This approach has the advantage to be simple but has several limitations. The first limitation is that to get this relationship ground truth LAI measurements are necessary at the same site and time as the spectral reflectances are collected. The second limitation is that the relationship between LAI and spectral reflectance is crop type dependent. This approach is also vulnerable to noise from soil background, atmospheric effect, and especially bidirectional properties of the vegetation. Soil influences on incomplete canopy spectra are due to the soil background signal on the optical properties of the overlying canopy (Jackson et al., 1980; Huete, 1988). Differences in Red and NIR flux transfers through a canopy result in a complex soil-vegetation interaction, which makes it difficult to subtract for soil background influences (Kimes et al., 1985; Sellers, 1985; Choudhury, 1987). The influences are dependent on the reflectance properties of the soil. Soil background effect is considered significant at intermediate canopy covers (Huete et al., 1985). A vegetated canopy will scatter and transmit NIR flux toward the soil as well as in between individual plants. The soil reflects part of the scattered and transmitted flux back to the sensor. The upper leaves absorb the red light, and the irradiance at the soil surface is only the one received directly from the sun and sky through canopy gaps. 39 B. Basso et al. - Rivista Italiana di Agrometeorologia 36-53 (1) 2004 Modeling Approach This approach includes radiative transfer and empirical models. The empirical models have the advantage of being simple but when inverted they infer little information on the vegetation. The radiative transfer model approach characterize light interaction with vegetation canopies and predicts the bidirectional reflectance distribution function (BRDF) as function of the observation geometry (Verhoef, 1984; Deering et al., 1990; Pinty et al., 1992; Strahler, 1994). Verhoef (1984) developed the SAIL (Scattering by Arbitrarily Inclined Leaves) model as function of solar position and sensor’s viewing geometry. The SAIL model assumes that the leaves are randomly oriented and uniformly distributed in a single layer. The model requires reflectance of the soil underneath, LAI, leaf reflectance and leaf transmittance as inputs. Inverting the model with remote sensing data, LAI can be estimated. There are several other models more complicated due to parameters that are difficult or impossible to measure in the field. The first two approaches to estimate vegetation with remote sensing have the advantage to be simple but the technology transfer behind these approaches is limited due to the ground truth measurements necessary, and for the sensitivity to birectional effect. The advantage of the modeling approach is that most radiative transfer models are based on the birectional properties of the natural land surface, therefore, by inverting them surface physical properties can be more objectively inferred (Qi et al., 1995). The other advantage of the models is that the optical properties of leaves are characterized by parameters (leaf reflectance, transmittance, absorptance, LAI etc.) required by the model thus BRDF models can predict the bidirectional reflectance in different viewing directions and illumination conditions. This modeling approach has great potential for application with multidirectional measurements. One major limitation, though, is the limited availability of multidirectional measurements. They cannot be obtained from a single sensor and this consequently limits the direct application of the model for vegetation assessment. Qi et al., 1995 applied a model-to model approach to overcome this limitation. The approached consisted in using a series of models inverted to obtain the parameters required for the simulation of bidirectional reflectances. Remote sensing data were collected using a Modular Multi-band Radiometer (MMR). They found satisfactory results, but the accuracy of predicting LAI with this approach is dependent on the accuracy of the models and on the atmospheric corrections. Vegetation indices Several vegetation indices have been developed by linear combination or ratios of red, green and near-infrared spectral bands. Vegetation indices are more sensitive than individual bands to vegetation parameters (Baret and Guyot, 1991; Qi et al., 1993). Plant canopy reflectance factors and derived multispectral VIs are receiving increased attention in agricultural research as robust surrogates for traditional agronomic parameters (e.g. leaf area index (LAI), fraction of green cover, fraction of absorbed photosynthetically active radiation (fAPAR) etc. Often viewed simply as measures of plant biomass or 40 green leaf area index, VIs are strongly modulated by interactions of solar radiation with photosynthetically active plant tissues and thus also are indicative of dynamic biophysical properties related to productivity and surface energy balance. Vegetation indices (VIs) have been designed to find a functional relationship between crop characteristics and remote spatial observation (Wiegand et al., 1990). VIs tend to reach a saturation level asymptotically for values of LAI between 3 to 6, based on the type of index used and type of plant (Carlson, et al., 1997; Aparicio et al., 2000). Another application is the use of VIs as a mapping device. In this case VIs are use in image classification, to separate vegetated from non-vegetated areas, to distinguish between different types and densities of vegetation, to monitor seasonal variations in vegetative vigor, abundance and distribution (Campbell, 1996; Barnes et al., 2003). VIs are influenced by external and internal factors (Yoshioka et al. 2000; Huete, 1989; Huete and Jackson, 1988; Baret and Guyot, 1998). External factor such as sensor calibration, sun and view angle and atmospheric condition, internal factors, instead, are variation in canopy and leaf optical properties and canopy background. To understand how VIs are designed, it is essential to know some concepts related to influence of soil and the use of the soil line and vegetation isoline. At this point is useful to introduce the different kind of VIs that have developed over the years. Some of the indices have developed considering that all vegetation isoline converge at a single point. These indices are called “ratio-based” and measure the slope of the line between the point of convergence and the soil line. The indices are: Normalized Difference Vegetative Index (NDVI), Soil Adjusted Vegetative Index (SAVI) and Ratio Vegetative Index (RVI). When the vegetation isoline are considered parallel to soil line, and the distance is measured perpendicular to the soil line, the indices are called “perpendicular” vegetation indices. These indices are: Perpendicular Vegetative Index (PVI), Weighted Difference Vegetative Index (WDVI), Three Dimensional Greenness Index (GVI3) and Difference Vegetative Index (DVI). Daughtry et al. (2000) classified VIs into two categories: Intrinsic indices, that include ratios of two or more bands in the visible and NIR wavelengths (NIR/Red; NIR/Green; NDVI; Green Normalized Difference Vegetative Index). These indices are sensitive to background reflectance properties and are often difficult to interpret at low LAI (Daughtry et al.,2000; Rondeaux et al., 1996); Soil-line vegetation indices, use the information of soil line in NIR-Red reflectance to reduce the effect of the soil on canopy reflectance (SAVI; Optimized Soil Adjusted Vegetative Index (OSAVI); Transformed Soil Adjusted Vegetative Index (TSAVI)). Baret and Guyot (1991) have classified VIs into two categories: Indices characterized by “slope”: RVI; NDVI; SAVI; TSAVI. Indices characterized by “distance”: PVI; WDVI; GVI. B. Basso et al. - Rivista Italiana di Agrometeorologia 36-53 (1) 2004 Theory of Vegetation Indices Where a and b are the coefficient derived from the soil line: NIRsoil= a REDsoil + b. RVI is the Ratio Vegetation Index (Jordan, 1969; Pearson and Miller, 1972). A common practice in remote sensing is the use of band ratios to eliminate various albedo effects. In this case the vegetation isoline converge at origin. Soil line has slope of 1 and passes through origin, it range from 0 to infinity. And it is calculated as follow: WDVI is the Weighted Different Vegetation Index (Clevers, 1988) and like PVI is sensitive to atmospheric variation (Qi et al., 1994). Vegetation isolines are parallel to soil line. Soil line has arbritary slope and passes through origin, vegetation index range is inifinite. RVI = ρNIR/ρred Where a is the slope of the soil line. The Green Normalized Vegetative Index (GNDVI) is a modification of the NDVI where the Red portion is substituted by the reflectance in the Green band (Gitelson et al., 1996). DVI is the Difference Vegetation Index, (Richardson and Everitt (1992), but appears as VI in Lillesand and Kiefer (1994). Vegetation isolines are parallel to soil line. Soil line has arbitrary slope, passes through origin, and index range is infinite. (4) PVI is the Perpendicular Vegetation Index (Crippen, 1990), and it is sensitive to atmospheric variation. In this case vegetation isolines are parallel to soil line. Soil line has arbitrary slope, passes through origin and the index range from -1 to 1. PVI = 1/√a2+1 (ρNIR - aρred -b) Huete (1988) proposed a Soil Adjusted Vegetation Index (SAVI) to account for the optical soil properties on the plant canopy reflectance. SAVI involves a constant L to the NDVI equation. The index range is from -1 to +1. SAVI = ρNIR - ρred / (ρNIR+ρred+L) (1+L) (7) (3) VIs assume that external noise (soil background, atmosphere, sun and view angle effect) is normalized, but this assumptions is not always true. The relative percentage of sunlit, shaded soil and plants components is highly dependent upon the view angle. Qi et al. (1995) studied the effect of multidirectional spectral measurements on the biophysical parameter estimation using a modeling approach. When the bidirectional effect is transformed from reflectance domain into vegetation index domain, it could be reduced (Jackson et al., 1990; Huete et al., 1992) or increased (Kimes et al., 1985; Qi et al., 1994b), depending on the vegetation types and solar zenith angles. Qi (1995) suggested that when bidirectional effect is a major concern (NDVI/NDVIo > 1) it is better to use NIR rather than NDVI, and that bidirectional effect on vegetation indices must be quantified before a quantitative VI-LAI relationship can be used. DVI = ρNIR-ρred (6) (2) NDVI is the Normalized Difference Vegetation Index (Kriegler, 1969; Rouse et al., 1973) and it is the common vegetation index referring to. This index vary between 1 and 1. In this case vegetation isoline are considered to be convergent at origin and soil line slope is 1 and passed through origin. It is calculated as: NDVI = ρNIR-ρred / ρNIR-ρred WDVI = ρNIR - aρred (5) The constant L is introduced in order to minimize soilbrightness influences and to produce vegetation isolines independent of the soil background (Baret and Guyot, 1991). This factor vary from 0 to infinity and the range depends on the canopy density. For L=0 SAVI is equal to NDVI, for L tends to infinity, SAVI is equal to PVI. However for intermediate density L was found equal to 0.5. Huete (1988) suggested that there maybe two or three optimal adjustment factor (L) depending on the vegetation density (L=1 for low vegetation; L=0.5 for intermediate vegetation densities; L=0.25 for higher density). TSAVI is the Transformed Adjusted Vegetation Index (Baret et al., 1989), and it is a measure of the angle between the soil line and the vegetation isoline. The soil line has arbitrary slope and intercept. The interception between soil line and vegetation isoline occur somewhere in the third quadrant. Baret and Guyot (1991) have proposed an improving of the initial equation as follow: TSAVI = a(ρNIR - aρred –b) / [aρNIR + ρred – a b + χ (1 + a2)] (8) Where a and b are soil line parameters (slope and intercept of the soil line) and χ has been adjusted so as minimize background effect, and its value is 0.08. TSAVI values ranging from 0 for bare soil and is close to 0.70 for very dense canopies as reported from Baret and Guyot (1991). At 40% green cover, the noise level of the NDVI is 4 times the WDVI and almost 10 times the SAVI, corresponding to a vegetation estimation error of +/- 23% for the NDVI, +/- 7% cover for the WDVI, and +/- 2.5% for the SAVI. Therefore the SAVI is a more representative vegetation indicator than the other Vis, but an optimization of the L factor will further increase his value (Qi et al., 1994). 41 B. Basso et al. - Rivista Italiana di Agrometeorologia 36-53 (1) 2004 Qi et al. (1994) developed a Modified Soil Vegetation Index (MSAVI). This index provide a variable correction factor L. Geometrically vegetation isolines don’t converge to a fixed point as SAVI, and soil line has not fixed slope and passes through origin. Correction factor is based on calculation of NDVI and WDVI as shown by equations 9 and 10: MSAVI = ρNIR - ρred / (ρNIR+ρred+L) (1+L) (9) where L is calculated as follow: L = 1 – 2 a * NDVI * WDVI (10) This term is computed to explain the variation of L among different types of soils, moreover L varies with canopy cover, and it’s range varies from 0 for very sparse canopy to 1 for very dense canopy. To further minimize the soil effect Qi et al. (1994) use an L function with boundary condition of 0 and 1 (Ln= 1-MSAVIn-1) and an MSAVI equal to: Stress related Vegetation Index (STVI) (Gardener, 1983): STVI = ρMIR * ρred / ρNIR (15) Cubed ratio index (CRVI) (Thenkabail et al.,1994): CRVI = (ρNIR / ρMIR)3 (16) The VIs that account for soil effect, do not consider atmospheric conditions, sensor viewing angle, solar illumination conditions. Kaufman and Tanré (1992) developed the Atmospherically Resistant Vegetation Index (ARVI) and the Soil and Atmospherically Resistant Vegetation Index (SARVI and SARVI2) where the reflectances are corrected for molecular scattering and ozone absorption. Liu and Huete (1995) incorporated a soil adjustment and atmospheric resistance concepts into a Modified Normalized Vegetation Index (MNDVI). SARVI2 as well as ARVI, SARVI are able to remove smoke effect and cirrus clouds from images (Huete et al., 1996). MSAVIn=[(ρNIR - ρred)/ρNIR + ρred+1– MSAVI n-1]*(2-MSAVI n-1) (11) The final solution for MSAVI is: MSAVI = 2 ρNIR + 1 – [(2 ρNIR + 1)2 –8(ρNIR - ρred) ]0.5 / 2 (12) OSAVI is the Optimized Soil Adjusted Vegetation Index. This index has the same formulation of the SAVI family indices, but the value L or X as refered by Rondeaux et al. (1996) is the optimum value that minimizing the standard deviations over the full range of cover. OSAVI = ρNIR - ρred / (ρNIR+ρred+0.16) (1+0.16) (13) GESAVI is the Generalized Soil Adjusted Vegetation Index. This index is based on an angular distance between the soil line and the vegetation isolines. GESAVI is not normalized and vary from 0 to 1 (from bare soil to dense canopies). Vegetation isolines are neither parallel nor convergent at the origin.Vegetation isolines intercept the soil line at any point depending on the vegetation amount. GESAVI = ρNIR - ρred b - a / ρred + Z (14) Z is the soil adjustment coefficient, and its based on the assumption that vegetation isolines intercept soil line at any point in the third quadrant. Z decrease when vegetation cover increase. However, practically, Z consider vegetation isolines convergent in a point. At least this hypotesis may be limited for dense canopies (Gilabert et al., 2002). To normalize soil effects Z value is found at 0.35. Indices that include the Mid-InfraRed Band (MIR) are: 42 Remote sensing application for evapotranspiration estimation All objects on the Earth’s surface emit radiation in the thermal-infrared (TIR) part of the spectrum (~ 8 to 14 µm). This emitted energy has proven useful in assessing crop water stress because the temperature of most plant leaves are mediated by soil water availability and its effect on crop evaporation (Jackson, 1982; Hatfield et al., 1983; Moran et al., 1989b; Pinter et al., 2003). In recent years, there has been much progress in the remote sensing of some of the parameters that can contribute to the estimation of evapotranspiration (ET). These include surface temperature, surface soil moisture, vegetative cover and incoming solar radiation. The surface temperature can be estimated from measurements at the thermal infrared wavelengths of the emitted radiant flux, that is the 10.5 and 12.5 µm. The microwave emission and reflection or backscatter from soil, primarily for wavelengths between 5 and 21 cm, are dependent on the dielectric properties of the soil, which are strong functions of the soil moisture content. Thus, measurements of these microwave properties can be used to obtain estimates of the surface soil moisture. Crop stress, due to water deficiency, crop diseases, is often shown with a decrease in the transpiration rate of the crop. Several studies have been carried on estimating ET with remote sensing data (Reginato, 1985; Jackson et al., 1987; Moran, et al., 1992, 1994, 1995; Maas, 1992,1993a, 1993b; Carlson et al., 1995, Hunsaker et al., 2003). A combination of remote sensing data and soil-plant-atmosphere models is commonly seen in the literature for ET estimation. The location of the “red edge” obtained with hyperspectral measurements shows potential for early detection of water stress (Shibayama et al., 1993). “Stress-Degree-Day” (SDD; Idso et al., 1977b), “Crop Water Stress Index” (CWSI; Idso et al., 1981; Jackson et al., 1981), “Non-water-stressed baseli- B. Basso et al. - Rivista Italiana di Agrometeorologia 36-53 (1) 2004 ne (Idso et al., 1982), “Thermal Kinetic Window” (TKW; Mahan and Upchurch, 1988), and “Water Deficit Index” (WDI, Moran et al., 1994) are indices that measure plant stress induced by water stress. These indices have been used in research on more than 40 different crop species (Gardner et al., 1992a; Gardner et al., 1992b). Most studies have shown that the thermal infrared is more sensitive to water stress than is reflectance in visible or NIR. However, the reflective portion of the spectrum and VIs also respond to plant water stress status when the canopy changes architecture through the leaf rolling or wilting (Moran et al., 1989a) or alters the senescence rate (Pinter et al.,1981). Thermal plan water stress indices provide valuable information and adequate lead time to schedule irrigations. Thermal indices can overestimate water stress when canopy cover is full and the sensors view a combination of cool plant and warm soil temperatures. The WDI a combination of VI and TIR (Moran et al., 1994; Clarke, 1997 and Clarke et al, 2001) seems to have overcome this problem since it accounts for the amount of plant cover through the VI part of the index A cost benefit study by Moran (1994) shows that irrigation scheduling with thermal infrared sensors on aircraft is both practical and affordable if growers join together to purchase the images. Hatfield (1984c) found that spatial variation of surface temperature in wheat changed with the degree of water availability. One alternative tool for a spatially variable irrigation can be to mount infrared sensors on irrigation booms to provide the capability to vary irrigation amounts as the unit travels across the field. VIs can be then used as surrogates for crop coefficients (Kcb). Crop coefficients are usually obtained from curves or tables and they lack flexibility to account for spatial and temporal crop water needs caused by uneven plant population, unusual weather patterns, non uniform water application, nutrient stress or pest pressures (Bausch and Neale, 1987; Choudhry et al., 1994; Pinter et al., 2003). Soil Salinity Remote sensing can also be used to map areas of soils that have been contaminated by salt. The principles behind this applications is that salt in the soil produces an unusually high surface reflectance. Salted areas can also be identified by detecting areas with reduces biomass or changes in spectral properties of plant growing in affected areas (Barnes et al., 2003). Leone et al. (2001) evaluated the impact of soil salinity induced through irrigation with saline water on plant characteristics and assessed the relationships between these characteristics and spectral indices. They showed that soil salinity had a clear impact on plant characteristics and significant relationships between chlorophyll content, biomass, NDVI and red edge peak. Studies have also shown an increase in canopy temperature of plants exposed to excessive salts in irrigation water (Howell et al., 1984a; Wang et al, 2002b), suggesting the possibility of previsual detection of stress which can manage with the appropriate measure of leaching or irrigation with good quality water. Remote Sensing in Precision Agriculture Direct Application The past research efforts on remote sensing have provided a rich background of potential application to sitespecific management of agricultural crops. In spite of the extensive scientific knowledge, there few examples of direct application of remote sensing techniques to precision agriculture in the literature. The reasons are mainly due to the difficulty and expense of acquisition of satellite images or aerial photography in timely fashion. With the progress in GPS and sensor technology direct application of remote sensed data is increasing. Now an image can be displayed on the computer screen with real-time position superimposed on it. This allows for navigation in the field to predetermined points of interest on the photograph. Blackmer et al.(1995) proposed a system for N application to corn based on photometric sensors mounted on the applicator machine. They showed that corn canopy reflectance changed with N rate within hybrids, and the yield was correlated with the reflected light. Aerial photographs were used to show areas across the field that did not have sufficient N. The machine reads canopy colors directly and applies the appropriate N rate based on the canopy color of the control (well fertilized) plots (Blackmer and Schepers, 1996; Schepers et al., 1996). Sensor technology has seen many innovations, but it is currently behind other technologies and their availability has been cited as the most critical factor preventing the wider implementation of precision agriculture. Management zones can be extracted using VIs maps and with the use of a geographic information system (GIS) can be viewed over the a remotely sensed image. The computer monitor displayed the image along with the current position as the applicator machine moved on the field. When interfaced with variable rate sprayer equipment, real time canopy sensors could supply sitespecific application requirements improving nutrient use efficiency and minimizing contamination of groundwater (Schepers and Francis, 1998). Indirect Applications The most common indirect use of remote sensing images is as a base map on which other information is layered in a GIS. Other indirect applications include use of remotely measured soil and plant parameters to improve soil sampling strategies, remote sensed vegetation parameters in crop simulation models, and use in understanding causes and location of crop stress such as weeds, insect, and diseases. Satellite based images have limited use in precision agriculture due the cost related to the image acquisition and to the restricted spectral resolution, coarse spatial resolution and the inadequate temporal resolution of the images. Moran et al. (1997) in their excellent review on opportunities and limitations for image-based remote sensing in 43 B. Basso et al. - Rivista Italiana di Agrometeorologia 36-53 (1) 2004 precision agriculture, classify the information required for site specific management in information on seasonally stable conditions, information on seasonally variable conditions, and information to find the causes for yield spatial variability and to develop a management strategy. The first class of information includes condition that do not vary during the season (soil properties) and only need to be determined at the beginning of the season. Seasonally variable conditions, instead, are those that are dynamic within the season (soil moisture, weeds or insect infestation, crop diseases) and thus need to be monitored throughout the entire season for proper management. The third category is comprehensive of the previous two to determine the causes of the variability. Remote sensing can be useful in all three types of information required for a successful precision agriculture implementation. Muller and James (1994) suggested a set of multitemporal images to overcome the uncertainity in mapping soil texture due to differences in soil moisture and soil roughness. Moran et al. (1997) also suggested that multispectral images of bare soil could be used to map soil types across a field. Crop growth and intercepted radiation Remote sensing techniques have also been applied to monitor seasonally variable soil and crop conditions. Knowledge of crop phenology is important for management strategies. Information on the stage of the crop could be detected with seasonal shifts in the “red edge” (Railyan and Korobov, 1993), bidirectional reflectance measurements (Zipoli and Grifoni, 1994), and temporal analysis of NDVI (Boissard et al., 1993). Moreover Wiegand et al. (1991) consider them as a measure of vegetation density, LAI, biomass, photosynthetically active biomass, green leaf density, photosynthesis rate, amount of photosynthetically active tissue and photosynthetic size of canopies. Aparicio et al. (2000) using three VIs (NDVI; Simple Ratio; Photochemical Reflectance Index) to estimate changes in biomass, green area and yield in durum wheat. They results suggest that under adequate growing conditions, NDVI may be useful in the later crop stage, as grain filling, where LAI values are around 2. SR, under rainfed condition, correlated better with crop growth (total biomass or photosynthetic area) and grain yield than NDVI. This fact is supported by the nature of relationship between these two indices and LAI. SR and LAI show a linear relationship, compared to the exponential relationship between LAI and NDVI. However the utility of both indices, as suggested by the authors, for predicting green area and grain yield is limited to environments or crop stages in which the LAI values are < 3. They found that in rainfed conditions, the VIs measured at any stage were positively correlated (P < 0.05) with LAI and yield. Under irrigation, correlations were only significant during the second half of the grain filling. The integration of either NDVI, SR, or PRI from heading to maturity explained 52, 59 and 39% of the variability in yield within twenty-five genotypes in rainfed conditions and 39, 28 and 26% under irrigation. 44 Shanahan et al. (2001) use three different kinds of VIs (NDVI, TSAVI, GNDVI) to asses canopy variation and its resultant impact on corn (Zea mays L.) grain yield. Their results suggest that GNDVI values acquired during grain filling were highly correlated with grain yield, correlations were 0.7 in 1997 and 0.92 in 1998. Moreover they found that normalizing GNDVI and grain yield variability, within treatments of four hybrids and five N rates, improved the correlations in the two year of experiment (1997 and 1998). Correlation, however, increases with a net rate in 1997 from 0.7 to 0.82 rather than in 1998 (0.92 to 0.95). Therefore, the authors suggest that the use of GNDVI, especially acquiring measurements during grain filling is useful to produce relative yield maps that show the spatial variability in field, offering an alternative to use of combine yield monitor. Raun et al. (2001) determined the capability of the prediction potential grain yield of winter wheat (Triticum aestivum L.) using in-season spectral measurements collected between January and March. NDVI was computed in January and March and the estimated yield was computed using the sum of the two postdormancy NDVI measurements divided by the Cumulative Growing Degree Days from the first to the second reading. Significant relationships were observed between grain yield and estimated yield, with R2 = 0.50 and P > 0.0001 across two years experiment and different (nine) locations. In some sites the estimation of potential grain yield, made in March and measured grain yield made in mid-July differed due to some factors that affected yield. The capability of VIs to estimate physiological parameters, as fAPAR, is studied on other crops, faba bean (Vicia faba L.) and semileafless pea (Pisum sativum L.) that grows under different water condition, as an experiment followed by Ridao et al. (1998) where crops see above grew both under irrigated and rainfed conditions. They have computed several indices (RVI, NDVI, SAVI2, TSAVI, RDVI, PVI) and linear, exponential and power relationship between VI and fAPAR were constructed to assess fAPAR from VIs measurements. During the preLAImax phase, in both species, all VIs correlated highly with fAPAR, however R2 at this stage did not differ significantly between indices that consider soil line (SAVI2 and TSAVI) and those that did not consider it (NDVI, RVI, RDVI). In post-LAImax phase the same behaviour was observed. All VIs are affected by the hour of measurement at solar angles grater than 45°. Authors conclude that simple indices as RVI and NDVI, can be used to accurately assess canopy development in both crops, allowing good and fast estimation of fAPAR and LAI. Nutrient Management Appropriate management of nutrients is one of the main challenges of agriculture productions and at the same environmental impact. Remote sensing is able to provide valuable diagnostic methods that allow for the detection B. Basso et al. - Rivista Italiana di Agrometeorologia 36-53 (1) 2004 of nutrient deficiency and remedy it with the proper application. Several studies have been carried out with the objective of using remote sensing and vegetation indices to determine crop nutrient requirements (Schepers et al., 1992; Blackmer et al., 1993; Blackmer et al., 1994; Blackmer et al., 1996a; Blackmer et al., 1994b; Blackmer and Schepers 1996; Daughtry et al. (2000). Results from these studied concluded that remote sensing imagery can be a better and quicker method compared to traditional method for managing nitrogen efficiently. Bausch and Duke (1996) developed a N reflectance index (NRI) from green and NIR reflectance of an irrigated corn crop. The NRI was highly correlated to with an N sufficiency index calculated from SPAD chlorophyll meter data. Because the index is based on plant canopy as opposed to the individual leaf measurements obtained with SPAD readings, it has great potential for larger scale applications and direct input into a variable rate application of fertilizer. Ma et al. (1996) studied the possibility to evaluate if canopy reflectance and greenness can measure changes in Maize yield response to N fertility. They have derived NDVI at three growing stage: preanthesis, anthesis and postanthesis. NDVI is well correlated with leaf area and greenness. At preanthesis NDVI showed high correlation with field greenness. At anthesis correlation coefficient of NDVI with the interaction between leaf area and chlorophyll content was not significant with yield. Ma et al. (1996) summarized that reflectance measurements took prior to anthesis predict grain yield and may provide inseason indications of N deficiency. Gitelson et al. (1996) pointed out that in some conditions, as variation in leaf chlorophyll concentration, GNDVI is more sensitive than NDVI. In particular is the green band, used in the computing GNDVI that is more sensitive than the red band used in NDVI. This changes occurs when some biophysical parameters as LAI or leaf chlorophyll concentration reach moderate to high values. Fertility levels, water stress and temperature can affect the rate of senescence during maturation of crops. In particular Adamsen et al. (1999) used three different methods to measure greenness during senescence on spring wheat (Triticum aestivum L.): digital camera, SPAD, hand-held radiometer. They derived G/R (green to red) from digital camera, NDVI from an hand-held radiometer and SPAD readings was obtained from randomly selected flag leaves. All three methods showed the similar temporal behaviour. Relationship between G/R and NDVI showed significant coefficient of determination and their relationship were described by a third order polynomial equation (R2 = 0.96; P < 0.001). Relation is linear until G/R > 1, when canopy approach to maturity (G/R < 1) NDVI is still sensitive to the continued decline in senescence than did G/R. This fact suggest that the use of the visible band is limited in such conditions. However authors found that G/R is more sensitive than SPAD measurements. Daughtry et al. (2000) have studied the wavelengths sensitive to leaf chlorophyll concentration in Maize (Zea mays L.). VIs as NIR/Red, NDVI, SAVI and OSAVI, have shown LAI as the main variable, accounting for > 98% of the variation. Chlorophyll, LAI, and their interaction accounted for > 93% of the variation in indices that compute the green band. Background effect accounted for less than 1% of the variation of each index , except for GNDVI, which was 2.5%. Serrano et al. (2000) studied the relationship between VIs and canopy variables (aboveground biomass, LAI canopy chlorophyll A content and the fraction of intercepted photosynthetic active radiation (fIPAR) for a wheat crop growing under different N supplies. The VIs-LAI relationships varied among N treatments. The authors also showed that VI were robust indicators of fIPAR independently of N treatments and phenology. Li et al. (2001) studied spectral and agronomic responses to irrigation and N fertilization on cotton (Gossypium hirsutum L.) to determine simple and cross correlation among cotton reflectance, plant growth, N uptake, lint yield, site elevation, soil water and texuture. NIR reflectance was positively correlated with plant growth, N uptake. Red and middle-infrared reflectance increased with site elevation. Li et al. (2001) found that soil in depression areas contains more sand on the surface than on upslope areas. This behaviour modified reflectance patterns. As a result, a dependence on sand content was shown by NDVI with higher values in the depression areas and lower values in areas where the soil had more clay. In addition cotton NIR reflectance, NDVI, soil water, N uptake and lint yield were significantly affected by irrigation (P < 0.0012). The N treatment had no effect on spectral parameters, and interaction between irrigation and N fertilizer was significant on NIR reflectance ( P < 0.0027). Red and NIR reflectance and NDVI were crosscorrelated with soil water, sand, clay and site elevation across a distance of 60 to 80 meters. Cross-correlation analysis of spectral reflectance, soil texture and site elevation could be useful for an in-season adjustment in water and N fertilizer application. Moreover authors pointed out on the possibility to use cross-correlation distance between NDVI and site elevation as a distance of variable N application over heterogeneous fields. Wright (2003) investigated the spectral signatures of wheat under different N rates, and the response to a midseason application at heading. VIs were computed (RVI, NDVI, DVI, GNDVI) and spectral data were compared with pre-anthesis tissue samples and post-harvest grain quality. The author found that imagery and tissue samples were significantly correlated with pre-anthesis tissue samples and post-harvest grain quality. The second application of N at heading improved protein only marginally. GNDVI was significantly correlated with nitrogen content of plants. VIs used in the study, whether from satellite or aircraft correlated well with preseason N and plant tissue analysis, but had lower correlation with protein. Osborne et al. (2002a; 2002b) demonstrated that hyperspectral data in distinguishing difference in N and P at the leaf and canopy level, but the relationship were not constant over all plant growth stages. Adams et al. (2000) have detected Fe, Mn, Zn and Cu deficiency in soybean using hyperspectral reflectance techniques and proposing a Yellowness Index (Adam et al., 1999) that 45 B. Basso et al. - Rivista Italiana di Agrometeorologia 36-53 (1) 2004 evaluated leaf chlorosis based on the shape of the reflectance spectrum between 570 nm and 670 nm. Pest Management Remote sensing has also shown great potential for detecting and identifying crop diseases (Hatfield and Pinter, 1993) and weeds. Visible and NIR bands can be useful for detecting healthy plants versus infected plants because diseased plant react with changes in LAI, or canopy structure. Malthus and Madeira (1993) using hyperspectral information in visible and NIR bands, were able to detect changes in remotely sensed reflectance before disease symptoms were visible to the human eye. Weed management represent an important agronomic practice to growers. Weeds compete for water, nutrient, light and often reduce crop yield and quality. Decisions concerning their control must be made early in the crop growth cycle. Inappropriate herbicide application can also have the undesirable effect on the environment and a side effect to the crop. With the advent of precision agriculture, there has been a chance from uniform application to the adoption of herbice-ready crop and to apply herbicide only when and where needed. This kind of approach is economically efficient and environmentally sound but site-specific herbicide management requires spatial information on the weeds. The discrimination between crops and weeds is usually accomplished based on the differences in the visible/NIR spectral signatures of crops and specific weeds (Gausman et al., 1981; Brown et al., 1994) or by acquiring images when weed coloring is particularly distinctive. Richardson et al. (1985) demonstrated that multispectral aerial video images could be used to distinguish uniform plot of Johnsongrass and pigweed from sorghum, cotton and cantaloupe plots. Several other authors have utilized spectral imagery to separate crops from weeds based on spectral signatures of species and bare soil (Hanks and Beck, 1998) or based on the leaf shape determine by the machine vision technology (Franz et al., 1995; Tian et al., 1999). Basso et al. 2004 (unpublished data) used the handheld radiometer CropScan to determine if a wheat field with various level of pappy (Papever Rhoeas) infestation could be detected by the multispectral radiometer . The study showed that the reflectance in the Red and NIR of the highly infested areas with pappy of the durum wheat field was significantly different from the no infestations of lower levels of weed presence (Figure 5). Remote sensing can also be used to determine herbicide injury to the crop for insurance purposes (Hickman et al., 1991; Donald, 1998a, Donald 1999b). To improve application efficiency of herbicides, Sudduth and Hummel (1993) developed a portable NIR spectrophotometer for use in estimating soil organic matter as part of the estimation procedure for the amount of herbicide to be sprayed. Several studies have also been carried out using remote sensing for identifying and managing insects, mite and nematode populations. Such studies have been able to demonstrate that remote sensing is able to detect actual 46 changes in plant pigments caused by pest presence, damages by pest and to identify areas susceptible to infestation. Riedell and Blackmer (1999) infested wheat seedlings with aphids and after 3 weeks they measured the reflectance properties of individual leaves. The leaves of the infected plants had lower chlorophyll concentration and displayed significant changes in reflectance spectra at certain wavelengths (500 to 525, 625 to 635 and 680 to 695 nm). This study in combination with others (Cook et al., 1999; Elliot et al., 1999; Willers et al., 1999) suggests the potential usefulness of canopy spectra for identifying outbreaks in actual field situations and to guide field scouts to specific areas for directed sampling. Site specific pesticide application can reduce the impact of toxic chemicals on the environment by 40 percent (Dupont et al., 2000). Roots may sense difficult condition in the soil and thence send inhibitory signals to the shoots which harden the plants against the consequences of a deteriorating or restriction environment, especially if the water supply is at risk. TIR can provide early, sometimes previsual, detection of diseases that interfere with flow of water from the soil through the plant to the atmosphere. Pinter et al (1979) found that a cotton plant whose roots were infected with the soil-borne fungus Pythium displayed sunlit leaf temperature that were 3 to 5° C warmer than adjacent healthy plants. TIR was also used for detecting root diseases in red clover under irrigated conditions (Oliva et al., 1994). Plant Population Plant population is an important variable that influences the final yield (Ritchie and Wei, 2000). Plant stands are affected by soil parameters, weather, field slope, aspect, seedling diseases, tillage etc. Remote sensing imagery taken after emergence at the proper spatial resolutions can be used to determine plant populations. Basso et al. (2001) through NDVI were able to identify 3 management zones were plant population of soybean was highly different due to the different position in the landscape. Moran et al. (2003) stated that development of new sensitive sensors that remove soil background is on the way and accurate assessment of plant density should improve. Selection of growth traits The use of morphological and physiological traits as indirect selection criteria for grain yield is an alternative to breeding approach. Future wheat yield improvements may be gained by increasing total dry matter production (TDM). VIs have been proposed as an appropriate and nondestructive method to assess total dry matter and LAI. Aparacio et al., (2000) and (2002) investigated whether VIs could accurately identify TDM and LAI in durum wheat and serve as indirect selection criteria in breeding programs. They found that the best growth stages for growth traits appraisal were stages 65 and 75 of the Zadock scale. VIs accurately tracked changes in LAI when data were analyzed across a broad range of different growth stages, environment and genotypes. Since B. Basso et al. - Rivista Italiana di Agrometeorologia 36-53 (1) 2004 4000 2 R = 0.94 3500 Yield (kg ha-1) VIs lack of predictive ability for specific environment/growth stages combinations, their value as indirect genotypes selection criteria for TDM or LAI was limited. Ma et al., (2001) showed that canopy reflectance measured between R4 and R5 stages in soybean adequately discriminates high from low yielding genotypes and providing a reliable and fast indicator for screening and ranking soybean genotypes based on the relationship between NDVI and grain yield. 3000 2500 2000 1500 1000 Yield Estimation Remote sensing can provide valuable information of yield assessment and show spatial variation across the field. There are two approaches for yield estimation, the first is a direct method in which predicitions are derived directly from remote sensing measurements (Figure 6). The second method is an indirect one, where remotely sensed data are incorporated into simulation model for crop growth and development either as within season calibration checks of model output (LAI, biomass) or in a feedback loop used to adjust model starting conditions (Maas, 1988). The direct method for prediction yield using remote sensing can be based on reflectance or thermal-based. Both methods have been applied with case of successes on various crops like corn, soybean, wheat, alfalfa (Tucker et al, 1979; Tucker et al., 1981; Idso et al., 1977; Pinter et al., 1981). Hatfield (1981) in his survey of 82 different varieties of wheat was not able to find a consistent relationship between spectral indices and yield. Hatfield (1983b) coupled frequent spectral reflectance and thermal observation in a more physiological method to predict yields in wheat and sorghum. This method requires TIR daily measurements during grain filling period to estimate crop stress. Shanahan et al. (2001) demonstrated that the time of corn pollination was not a good growth stage to estimate yield because of the various that can cause tassel emergence dates to vary. Yang et al. (2000) found similar results, concluding that images from images taken at grain filling can provide good relationships between VIs and yield. Reliability of imagery for use in yield estimation decreases as the time before harvest increases because there is more opportunities for factors like various nature of stresses to influence yield. Aase and Siddoway (1981) had cautioned that the relationships of spectral indices to yield were depedent upon normal grain-filling conditions for the crop. Similar results were found by Basso et al. (2004) (personal communication, unpublished data) where the NDVI images on a rainfed durum wheat field showed different correlation to yield depending on the time of the image selected (Fig.7). In this specific case, spatial variability of soil texture and soil water uptake by plants affected by drought varied at anthesis presenting different scenarios from the one predicted by the NDVI estimation. 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 NDVI Fig. 6 – Correlation between NDVI derived from an image taken at flowering and yield for a soybean crop in Michigan USA. Fig. 6 – Correlazione tra NDVI in fioritura e produzione per una coltura di soia nello stato del Michigan, USA Combining remote sensing with crop modeling Crop models provide the ability to simulate different management options under different weather conditions, while remote sensing allows for identification of spatial patterns. Remote sensing data can directly be used for within season model calibration (Maas, 1993), model validation (Fisher, 1994). Once information on stable and dynamic variable are collected, remote sensing in combination with crop growth model can be used to quantify the spatial and temporal crop variability (Fig. 7 pag. 48). Clearly, the goal of crop simulation is to explain the spatial variability of crop performance mapped with grain yield monitoring systems and to help guide in management decisions related to the site-specific management of crop inputs. It is also clear that crop simulations cannot be performed everywhere given that the cost and the availability of detailed inputs would be prohibitive. A more balanced approach to spatial application of crop simulation models would be to delineate zones within the field representing areas of similar crop performance. One approach may be to obtain vegetation indexes derived from remote sensed imagery during critical times during the growing season, classify the images for target sampling, delineate spatial patterns and use the results of the target sampling as inputs for the models. Model validation can be then be performed at selected sites within these delineated management zones. Such an approach would facilitate the challenge of using crop models in precision farming by obtaining spatial inputs to simulate variations of crop yields across the field, as well as to decide where to use field averages for some factors along with spatially variable inputs for others. Basso et al. (2001) conducted a study to examine a new procedure for spatial validation of crop models for use in precision farming. This procedure used the CROPGRO-Soybean model to validate management zones across the field that were delineated using a NDVI classification procedure. Airborne false color composite images in the blue, red, green and NIR portion of the spectrum were taken at selected time during the season at 47 B. Basso et al. - Rivista Italiana di Agrometeorologia 36-53 (1) 2004 1 meter pixel resolution. The images provided spatial information about the condition of the crop throughout the season. Each image was used to generate NDVI maps of the field and to identify spatial patterns across the field. The false color composite image taken on July 18 was selected for quantifying areas with similar reflectance by grouping areas into classes of similar NDVI values using supervised classification technique. Pixels of similar reflectance were queried across the field after trying various ranges of values able to reproduce the spatial patterns visible in the original false-color composite image. The reclassified NDVI map from 18 July image clearly showed spatial variability in soybean performance. Classification of the NDVI image indicated three classes of importance in this field. The model performance indicated that the NDVI reclassification procedure was appropriate and with multi-year simulation should allow for the characterization of management zones for this field. The use of site specific model inputs obtained with the NDVI-reclassification procedure has a major advantage since the power and application of simulation models in precision farming has been limited by data requirements at the sub-field scale. The site-specific inputs approach is scale-independent because the scale is controlled by the observed variation in the field and that is the scale at which the model is applied. temporal resolution imagery will increase. Remote sensing application in precision agriculture can be direct, but most likely, is indirect. Rapid response is required to provide information about the condition of the current crop in time to make management input corrections to accomplish maximum yield. Images only show the spatial variability of plant condition, thus suggestions can only be made if the causes of the variability are understood. The current limitation for image-based applications is due to sensors attributes. Crop spatial variability is usually observed at a scale that only is too fine for the sensors currently available on the orbiting satellites. Airborne images are already limiting this problem but the temporal and spectral aspects still need to be improved. The success of remote sensing in agriculture will be measured by the type of information that is provided to the farmer, how quickly the information is delivered and the fee that is charged for the information. Because the potential market for remote sensing is great, competition in farming business should help make the success a reality. The future brings tremendous prospects for remote sensing applications especially if integrated with crop simulation models. Future challenges and Opportunities Acknowledgements VIs are often used as synonymous of plant vigour. This is not necessarily correct because broad wave band VIs do not have the capability for identifying the factor responsible for a specific type of stress. Narrow band indices such as the Photochemical Reflectance Index (PRI), Water Band Index (WBI), and the Normalized Pigment Chlorophyll Ratio Index (NPCI) are examples of reflectance indices correlated with physiological stresses such as nutrient and water deficts (Penuelas et al, 1994). A step forward was taken by the Environmental and Plant Dynamic Unit of Soil and Water Conservation Laboratory of the USDA-ARS in Phoenix Az, with the work conducted by Clarke et al., (2001), Pinter et al., (2003) Moran et al., (2003) on the Canopy Chlorophyll Content Index (CCCI) based on a VI plus the reflectance in a narrow red edge band (~720 nm) to distinguish nutrient stress from other causes of reduced green biomass in cotton. Yield monitor and remote sensing images at harvest should display similar pattern but in many cases due to errors in yield monitor system this match does not occur. Shanahan et al. (2001) and Pinter et al. (2003) suggested that VIs can be used as an alternative to yield monitor systems. New ideas for commercialization are the combinations of pre-harvest remote sensing image and yield map to display the actual and realistic spatial variability of yield across the field. A major research challenge is to develop stress detection algorithms that are universal and perfom well over space, throughout the seasons and between years. Remote sensing technology will improve as spatial, spectral and 48 Thank Dr. Iannetta Massimo of C.R.ENEA-Casaccia, Rome for his support and Dr. Cafiero Giovanni for the technical assistance. B. Basso et al. - Rivista Italiana di Agrometeorologia 36-53 (1) 2004 Fig. 4 – Spectral reflectance envelopes for deciduous (broadleaved) and coniferous (needle-bearing) trees (Adopted by Kalensky and Wilson, 1975). Fig.4 – Curva di riflettanza spettrale di alberi cedui (a foglia larga) e conifere (ad aghi). (Da Kalensky e Wilson, 1975) Fig. 2 - Example of false color composite images of a corn crop in Michigan USA. Fig. 2 – Esempio di immagini a colori falsi composti di un campo di mais nello stato del Michigan USA 9050 9000 8950 8900 8850 8800 8750 6000 6050 6100 Fig. 7 – Examples of a NDVI map of a durum wheat crop in Puglia, Italy, to illustrate spatial variability within the field as affected by landscape position and soil properties. Fig. 7 – Esempio di una mappa di NDVI di frumento duro per illustrare la variabilità spaziale in funzione della topografia e proprietà del suolo. 49 B. Basso et al. - Rivista Italiana di Agrometeorologia 36-53 (1) 2004 References Aase, J.K., Siddoway, F.H., 1981. Spring wheat yield estimates from spectral reflectance measurements, IEEE Transactions on Geoscience and Remote Sensing, 19(2): 78-84. Adamsen, F.J., Pinter, P.J. Jr., Barnes, E.M., LaMorte, R.L., Wall, G.W., Leavitt, S.W. Kimball, B.A. 1999. Measuring Wheat Senescence with a digital camera, Crop Science, 39(3): 719-724. Aparicio, N., Villegas, D., Casadeus, J., L. Araus, J., Royo, C. 2000. Spectral Vegetation Indices as Nondestructive Tools for Determining Durum Wheat Yield. Agron. J. 92: 83-91. Aparicio, N., Villegas, D., J.L. Araus, J., Casadeus, J.Royo, C. 2002. Relationship between Growth Traits and Spectral Vegetation Indices in Durum Wheat. Crop Sci. 42-15471555. Baret, F., Guyot, G., and Major, D. J., 1989. TSAVI: a vegetation index which minimizes soil brightness effects on LAI and APAR estimation, in Proceedings of the 12th Canadian Symposium on Remote Sensing, IGARRS ’90, Vanouver, BC, Canada, 10-14 July, Vol. 3, pp. 1355-1358. Baret, F., Guyot, G. 1991. Potential and Limits of Vegetation Indicies for LAI and APAR Assesment. Remote Sens. Environ. 35 : 161-173. Barnes, E. .M., Sudduth,K.A., Hummel,J.W., Lesch,S.M., Corwin,D.L., Yang,C., Daughtry,C.S.T., and Bausch,W.C. 2003. Remote- and ground-based sensor techniques to map soil properties." Photogrammetric Engineering and Remote Sensing 69(6): 619-630 Basso, B., Ritchie, J.T., Pierce, F. J., Braga, R.P., Jones, J.W. 2001. Spatial Validation of Crop Models for Precision Agriculture. Agric. Syst. 68: 97-112. Batchelor, W.D., B. Basso, J.O. Paz. 2002. Examples of strategies to analyze spatial and temporal yield variability using crop models. European Journal Of Agronomy Vol. 18, pp. 141-158. Bauer, M.E. 1975. The role of remote sensing in determining the distribution and yield of crops, Advances in Agronomy, 27:271-304. Bausch W,C. 1995. Remote sensing of crop coefficients for improving the irrigation scheduling of corn. Agric Water Manage 27:55–68 Bausch W.C., Neale C.M.U. 1987. Crop coefficients derived from reflected canopy radiation: a concept. Trans ASAE 30(3):703–709 Bausch, W.C., Duke, H.R. 1996. Remote sensing of plant nitrogen status in corn, Transactions of the ASAE, 39(5):18691875. Blackmer, A. M., White, S. E. 1998. Using Precision Farming Tecnologies to Improbe Management of Soil and Fertiliser Nitrogen. Austr. J. Agric. Res. 49: 555-564. Blackmer, T. M., Schepers, J. S. 1995. Use of a chlorophyll meter to monitor N status and schedule fertigation of corn. J. Production Agriculture 8, 56-60. Blackmer, T.M., Schepers, J.S. 1994. Techniques for monitoring crop nitrogen status in corn. Commun. Soil Sci. Plant Anal. 25:1791–1800 Blackmer, T.M., Schepers, J.S., Meyer, G.E. 1996a. Analysis of aerial photography for nitrogen stress within corn fields, Agronomy Journal, 88(5):729-733. Blackmer, T.M., Schepers, J.S. Varvel, G.E., alter-Shea, E.A.1996. Nitrogen deficiency detection capabilities of light reflection from irrigated corn canopies. Agron. J. 88:1–5. Boissard, P., Pointel, J. G., Huet, P., 1993. Reflectance, green leaf area index and ear hydric status of wheat from anthesis until maturity. Int. J. Remote Sens. 14, 2713-2729. 50 Boote, K.J. 2000. Physiology and modeling kernel set in maize. CSSA Spec. Publ. 29. CSSA and ASA, Madison, WI. Brown, R. B., Steckler, J. P. G. A., Amderson, G. W., 1994. Remote sensing for identification of weeds in no-till corn. Trans. ASAE. 37, 297-302. Campbell, J. B. 1996. Introduction to Remote Sensing, second edition. The Guilford Press. New York. Carlson, T. N., Ripley, D. A. 1997. On the Relation between NDVI, Fractional Vegetation Cover, and Leaf Area Index. Remote Sens. Environ. 62: 241-252. Carlson, T.N., Taconet, O. Vidal, A. Gilles, R.R. Olioso, A. Humes, K. 1995. An overview of the workshop on thermal remote-sensing held at La-Londe-Les-Maures, France, September 20-24, 1993, Agricultural and Forest Meteorology, 77(3-4):141-151. Chappelle, E.W., Kim, M.S. McMurtrey, J.E. 1992. Ratio analysis of reflectance spectra (RARS) – An algorithm for the remote estimation of the concentrations of Chlorophyll-a, clorophyll-b, and carotenoids in soybean leaves, Remote Sens. Environ. 39(3):239-247. Choudhury, B. J., 1987. Relationships between vegetation indices, radiation absorption and net photosynthesis evaluated by a sensitivity analysis. Remote Sens. Environ.. 22, 209233. Choudhury BJ, Amhed NU, Idso SB, Reginato RJ, Daughtry CS (1994) Relations between evaporation coefficients and vegetation indices studied by model simulations. Remote Sens Environ 50:1-17 Clarke,T.R. 1997. An empirical approach for detecting crop water stress using multispectral airborne sensors. Hortechnology 7(1) 9-16 Clarke, T.R., Moran, M.S., Barnes, E.M. Pinter, P.J,Jr., Qi, J. 2001. Planar domain Indices: A method for measuring a quality of a single component in two-component pixels, Proceedings IEEE International Geoscience and Remote Sensing Symposium, 09-13 July, Sydney, Australia, unpaginated CD-ROM. Clevers, J. G. P. W. 1988. The derivation of a simplified reflectance model for the estimation of leaf area index, Remote Sensing of Environment, vol 35., pp. 53-70. Clevers, J.G.P.W. 1997. A simplified approach for yield prediction of sugar beet based on optical remote sensing data. Remote Sens. Environ. 61:221–228 Colwell, J.E. 1974. Vegetation canopy reflectance. Remote Sens. Environ. 3:175–183. Cook, C.G., Escobar, D.E., Everitt, J.H. Cavazos, I., Robinson, A.F. Davis, M.R.1999. Utilizing airborne video imagery in Kenaf management and production, Industrial Crops and Products, 9:205-210. Crippen, R. E. 1990.Calculating the Vegetation Index Faster. Remote Sens.Environ.34: 71-73. Cropscan Inc. 2000. User’s Manual. Rochester MN. Daughtry,C.S.T., Kustas,W.P., Moran, M.S., Pinter Jr.,P.J., Jackson,R.D., Nichols,W.D.,Brown,P.W., and Gay,L.W. 1990. Spectral estimates of net radiation and soil heat flux." Remote Sens. Environ 32:111-124 Daughtry, C. S. T., Gallo, K. P., Goward, S. N., Prince, S. D., Kurstas, W. P. 1992. Spectral Estimates of Absorbed Radiation and Phytomass Production in Corn and Soybean Canopies. Remote Sens. Environ. 39: 141-152. Daughtry, C.S.T., McMurtrey, J.E., Chappelle, E.W., Hunter, W.J., Steiner, J.L. 1996. Measuring crop residue cover using remote sensing techniques, Theoretical and Applied Climatology, 54(1-2):17-26. B. Basso et al. - Rivista Italiana di Agrometeorologia 36-53 (1) 2004 Daughtry, C. S. T., Walthall, C. L., Kim, M. S., de Colstoun, E. B., McMurtrey III, J. E. 2000. Estimating Corn Leaf Chlorophyll Concentration from Leaf and Canopy Reflectance. Remote Sens. Environ. 74 : 229-239. Daughtry, C.S.T., 2001. Discriminating crop residues from soil by shortwave infrared reflectance, Agronomy Journal, 93(1):125-131. Deering D. W., Eck, T. F., Otterman, J. 1990. Bidirectional reflectances of selected desert surfaces and their threeparameter soil characterization. Agri. Forest Meteor., 52. 71-93 Demetriades-Shah, T.H., Steven, M.D., Clark J.A. 1990. High resolution derivative spectra in remote sensing. Remote Sens. Environ. 33:55-64. Donald, W.W., 1998a. Estimated soybean (Glicine max) yield loss from herbicide damage using ground cover or rated stunting, Weed Science, 46:454-458. Donald, W.W., 1998b. Estimating relative crop yield loss resulting from herbicide damage using crop ground cover or rated stunting, with maize and sethoxydim as a case study, Weed Research, 38:425-431. Dupont, J.K., Campenella, R. , Seal, M.R., Willers, J.L., Hood, K.B. 2000. Spatially variable insecticide applications through remote sensing, Proceedings Beltwide Cotton Conferences, 04-08 January, San Antonio, Texas (National Cotton Council of America and The Cotton Foundation), 1:426-429. Elliot, N.C., Kieckhefer, R.W., Lee, J.H., French, B.W. 1999. Influence of within-field and landscape factors on aphid predator populations in wheat. Landscape Ecology 14:239252. Frannz, E., Gebhardt, M.R. Unklesby, K.B. 1991. The use of local spectral properties of leaves as an aid for identifying weed seedlings in digital images, Transactions of the ASAE, 34(2):682-687. Frannz, E., Gebhardt, M.R. Unklesby, K.B. 1995. Algorithms for extracting leaf boundary information from digital images of plant foliage, Transactions of the ASAE, 38(2):625633. Gardner, B.R., Nielsen, D.C. Shock, C.C., 1992a. Infrared thermometry and the crop water-stress index: 1. History, theory, and base-lines, Journal of Production Agriculture, 5(4):462-466. Gardner, B.R., Nielsen, D.C. Shock, C.C., 1992b. Infrared thermometry and the crop water-stress index: 2. Sampling procedures and interpretation, Journal of Production Agriculture, 5(4):466-475. Gausman, H. W. 1985. Plant leaf optical parameters in visible and near-infrared light. Graduate studies. Texas Tech Univ. No. 29. Texas Tech Press, Lubbock, Tex. 78 p. Gausman, H. W., Allen, W. A., Cardenas, R. (1969), Reflectance of cotton leaves and their structure. Remote Sens.Environ. 1:19–22. Gausman, H. W., Allen, W. A., Escobar, D. E., Rodrigues,Z. Z., and Carenas, R. (1971), Age effects of cotton leaves on light reflectance, transmittance, and absorptance and on water content and thickness. Agron. J. 63:465–469. Gausman, H. W., Everitt, J. H., Gerbermann, A. H. Bowen, R. L. 1977a. Canopy reflectance and film image relations among three south Texas rangeland plants. J. Range Manage. 30:449-450. Gausman, H. W., Menges, R. M., Escobar, D. E. Everitt, J. H. Bowen R. L. 1977b. Pubescence affects spectra and imagery of silverleaf sunflower (Helianthus argophyllus). Weed Sci. 25:437-440. Gausman, H.W., Menges, R.M., Richardson, A.J, Walter, H.R. Rodriguez, R., Tamez, S. 1981. Optical-parameters of leaves of 7 weed species, Weed Science, 29(1):24-26. Gilabert, M. A., Gonzàles-Piqueras, J., Garcìa-Haro, F. J., Melià, J. 2002. A Generalized soil-adjusted vegetation index. Remote Sens. Environ.. 82, 303-310. Gitelson, A.A., Kaufman, Y.J. Merzlyak, M.N.. 1996. Use of green channel in remote sensing of global vegetation from EOS-MODIS. Remote Sens. Environ 58:289–298 Hanks, J. E.. Beck J. L. 1998. Sensor-controlled hooded sprayer for row crops. Weed Technol. 12:308-314. Hatfield, J.L., 1981. Spectral behaviour of wheat yield variety trials, Photogrammetric Engineering & Remote Sensing, 47:1487-1491. Hatfield, J.L., Vauclin, M., Vieira, S.R., Bernard, R. 1984c. Surface temperature variability patterns within irrigated fields, Agricultural Water Management, 8:429-437. Hatfield,J.L., Pinter Jr.,P.J. Remote sensing for crop protection. Crop Protection 12(6):403 414 (1993). Hatfield,J.L., Perrier,A., Jackson,R.D. Estimation of evapotranspiration at one time-of-day using remotely sensed surface temperatures. Agricultural Water Management 7(13):341-350 Hatfield,J.L., Pinter Jr.,P.J., Cumpton,M.C.,Webb,W.M. 1998. Data collection routines on the polycorder for use with infrared thermometers and multiband radiometers. Computers and Electronics in Agriculture 2:317-322 Hickman, M.V., Everitt, J.H., Escobar, D.E., Richardson, A.J. 1991. Aerial-photography and videography for detecting and mapping Dicamba injury patterns. Weed Technology, 5(4):700-706. Howell, T.A., Hatfield, J.L..Yamada, H Davis, K.R. 1984a. Evaluation of cotton canopy temperature to detect crop water stress, Transactions of the ASAE, 27(1):84:88. Huete, A.R., Jackson, R.D., Post. D.F. 1985. Spectral response of a plant canopy with different soil backgrounds. Remote Sens. Environ. 17:37–53 Huete, A.R. 1987. Soil-dependent spectral response in a developing plant canopy. Agron. J. 79:61 68 Huete, A. R. 1988. A Soil-Adjusted Vegetation Index (SAVI). Remote Sens. Environ. 25 : 295-309. Huete, A. R., 1989. Soil influences in remotely sensed vegetation-canopy spectra. In: Asrar, G. (Ed.), Theory and Application of Optical Remote Sensing. Wiley Series in Remote Sensing, Wiley New York, pp. 107-141. Huete, A. R., Jackson, R. D., 1988. Soil and Atmosphere influences on the spectra of partial canopies. Remote Sens. Environ.. 25, 89-105. Huete, A. R., Hua, G., Qu, J., Chebouni, A., Leeuwen, W. J. O. G., 1992. Normalization of multidirectional red and NIR reflectances with the SAVI. Remote Sens. Environ.. 40, 120. Hunsaker,D.J., Pinter Jr.,P.J., Barnes,E.M., Kimball,B.A. Estimating cotton evapotranspiration crop coefficients with a multispectral vegetation index. Irrigation Science 22:95104 (2003) Idso, S.B., 1982. Non-water-stressed baselines – A Key to measuring and interpreting plant water-stress, Agricultural Meteorology,27 (1-2):59-70. Idso, S.B., Jakson, R.D., Reginato, R.J. 1977a. Remote-sensing for agicultural water management and crop yield production, Agricultural Water Management, 1(4):299310. Idso, S.B., Jakson, R.D. Reginato, R.J. 1977b. Remote-sensing of crop yields. Science, 196(4285):19-25. Idso, S.B., Jakson, R.D. Pinter, P.J. Jr., Reginato R.J., Hatfield, J.L. 1981. Normalizing the stress-degree-day parameter foe environmental variability, Agricultural Meteorology, 24(1):45-55. Idso,S.B., Hatfield,J.L., Jackson,R.D., Reginato,R.J. 1979. Grain yield prediction: Extending the stress degree day 51 B. Basso et al. - Rivista Italiana di Agrometeorologia 36-53 (1) 2004 approach to accommodate climatic variability. Remote Sensing Environ 8(3):267-272. Idso,S.B., Hatfield,J.L., Reginato,R.J., Jackson,R.D. 1978. Wheat yield estimation by albedo measurement. Remote Sensing Environ 7:273-276 Idso,S.B., Pinter Jr.,P.J., Hatfield,J.L., Jackson,R.D., Reginato,R.J. 1979. A remote sensing model for the prediction of wheat yields prior to harvest. Journal of Theoretical Biology 77:217-228 Idso, S.B., Reginato ,R.J., Hatfield, J.L., Walker,G.K., Jackson, R.D., and Pinter Jr.,P.J. A generalization of the stressdegree-day concept of yield prediction to accommodate a diversity of crops. Agricultural Meteorology 21:205-211 Jackson, R. D. 1983. Spectral indices in n-space. Remote Sens. Environ. 13 : 409-421. Jackson, R. D., Teillet, P. M., Slater, P. M., Fedosejevs, G., Jasinski, M. F., Aase, J. K., Moran, M. S., 1990. Bidirectional measurements of surface reflectance for view angle corrections of oblique imagery. Remote Sensing. Environ.. 32, 189-202. Jackson, R.D., 1982. Canopy temperature and crop water stress, Advances in irrigation, Volume 1 (D.I. Hillel, editor). Academic Press, New York, N.Y., pp. 43-85. Jackson, R.D., M.S. Moran, L.W. Gay, and L.H. Raymond, 1987. Evaluating evaporation from field crops using airborne radiometry and ground-based meteorological data, Irrigation Science, 8(2):81-90. Jackson, R.D., Idso, S.B. Reginato, R.J, Pinter, P.J. Jr. 1980. Remotely sensed crop temperatures and reflectances as inputs to irrigation scheduling, Proceedings ASCE Irrigation and Drainage Division Specialty Conferences, 23-25 July, Boise, Idaho, pp. 390-397. Jackson, R.D., Kustas, W.P. Choundhury, B.J. 1988. A reexamination of the crop water-stress index, Irrigation Science, 9(4):309-317. Janiski, M. F., Eagleson, P. S., 1989. The structure of redinfrared scattergrams of semivegetated landscapes, IEEE Trans. Geosci. Remote Sens.. 27, 441-451. Jordan, C.F. 1969. Derivation of leaf area index from quality measurements of light on the forest floor. Ecology, 50, 663-666. Kaufman, Y. J., Tantrè, D. 1992. Atmospherically resistant vegetation index (ARVI) for EOS-MODIs, IEEE Trans. Geosci. Remote Sens.. 30 (2), 261-270. Kimes, D. S., Newcomb, W. W., Tucker, C. J. 1985. Directional reflectance factor distributions for cover types of Northern Africa. Remote Sens. Environ.. 18, 1-19. Leone, A.P., Menenti, M., Sorrentino, G. 2001. Reflectance Spectometry to Study Crop Response to Soil Salinity. Ital. J. Agron., 4,2, 75-85. Li, H., Lascano, R.J., Barnes, E.M., Booker, J. L., Wilson, T., Bronson, K.F. Segarra E. 2001. Multispectral reflectance of cotton related to plant growth, soil water and texture, and site elevation. Agron. J. 93:1327–1337 Lillesand, T. M., Kiefer, R. W., 1994. Remote sensing and Image Interpretation, 3rd edition. Wiley. New York. Ma, B. L., Morrison, M. J., Dwyer, L. M. 1996. Canopy Light Reflectance and Field Greenness to Assess Nitrogen Fertilization and Yield of Maize. Agron. J. 88: 915-920. Maas, S.J., 1993. Within-season calibration of modelled wheat growth using remote-sensing and field sampling, Agron. J., 85(3):669-672. Maas, S. J., 1993b. Parameterized model of gramineous crop growth: II. Within-season simulation calibration. Agron. J.. 85, 354-358. Maas, S. 1998. Estimating cotton canopy ground cover from remotely sensed scene reflectance. Agron. J. 90:384–388 52 Mahan, J.R., Upchurch, D.R. 1988. Maintenance of constant leaf temperature by plants. 1. Hypothesis limited homeothermy, Environmental and Experimental Botany, 28:351357. Miller J.R. , Hare E.W., Wu J., 1990. Quantitative characterization of the vegetation red edge reflectance. I. An inverted-Gaussian reflectance model. Int. J. Remote Sens., 11, 1755-1773. Moran, M. S., Inoue, Y., Barnes, E. M. 1997. Opportunities and Limitations for image-Based Remote Sensing in Precision Crop Management. Remote Sens. Environ. 61: 319-346. Moran, M. S., Maas, S. J., Pinter, P. J. Jr., 1995. Combining remote sensing and modelling for estimating surface evaporation and biomass production. Remote Sens. Rev.. 12, 335-353. Moran, M.S., 1994. Irrigation management in Arizona using satellites and airplanes, Irrigation Science, 15:35-44. Moran, M.S., Pinter, P.J. Jr., Clothier, B.E. Allen, S.G. 1989a. Effect of water-stress on the canopy architecture and spectral indices of irrigated alfalfa, Remote Sens. Environ. 29:251-261. Moran, M.S., Jackson, R.D., Raymond, L.H. Gay, L.W., Slater, P.N. 1989b. Mapping surface-energy balance components by combining Landsat thematic mapper and ground-based meteorological data, Remote Sens. Environ. 30(1):77-87. Moran, M.S., Clarke, T.R. Inoue, Y., Vidal, A. 1994. Estimating crop water-deficit using the relation between surfaceair temperature and spectral vegetation index, Remote Sens. Environ. 49:246-263. Moran,M.S., Fitzgerald,G.J., Rango, A., Walthall, C.L., Barnes, E.M., Bausch,W.C., Clarke,T.R., Daughtry,C.S., Everitt,J.H., Hatfield,J.L., Havstad,K.M., Jackson,T.J., Kitchen,N.R., Kustas,W.P., Mcguire,M.R., Pinter Jr.,P.J., Sudduth,K.A., Schepers,J.S., Schmugge,T.J., Starks,P.J., and Upchurch,D.R. 2003. Sensor development and radiometric correction for agricultural applications. Photogrammetric Engineering and Remote Sensing 69:705-718 Muller, E., James, M., 1994. Seasonal variation and stability of soil spectral patterns in a fluvial landscape. Int. J. Remote Sens.. 15, 1885-1900. Nagler, P.L. Daughtry, C.S. Goward, S.N. 2000. Plant litter and soil reflectance. Remote sens.Environ. 71:207-215. Oliva R.N., Steiner J.J., Young W.C., III. Red clover seed production: I. Crop water requirements and irrigation time. Crop Sci. 1994;34:178-184 Osborne, S.L., Schepers, J.S., Francis, D.D., Schlemmer M.R. 2002. Detection of phosphorus and nitrogen deficiencies in corn using spectral radiance measurements. Agron. J. 94:1215–1221 Pearson, R. L., Miller, L. D. 1972. Remote mapping of standing crop biomass for estimation of the productivity of the shortgrass Prairie, Panwnee National Grasslands, Colorado, in Proc. 8th Int. Symp. On Remote Sens. Of Environ., ERIM, Ann Arbor, MI, pp. 1357-1381. Penuelas, J., Filella, I. 1998. Visible and Near-Infrared Reflectance Techniques for Diagnosing Plant Physiological Status. Trends in Plant Science Vol. 3, No 4: 151-155. Peterson, T.A., Blackmer, T.M., Francis, D.D, Schepers, J.S. 1993. Using a chlorophyll meter to improve N management. Nebguide G93-1171A. Coop. Ext. Serv., Univ. of Nebraska, Lincoln, NE. Pinter, P.J. Jr., Ritchie, J.C., Hatfield, J.L. Hart, G.F. 2003. The Agricultural Research Service’s remote sensing program: An example of interagency collaboration, Photogrammetric Engineering and Remote Sensing, 69(6):615618. B. Basso et al. - Rivista Italiana di Agrometeorologia 36-53 (1) 2004 Pinter, Jr., P.J., Stanghellini, M.E., Reginato, R.J., Idso, S.B. Jenkins A.D., Jackson, R.D. 1979. Remote detection of biological stresses in plants with infrared thermometry, Science, 205(4406):585-586. Pinter, Jr., P.J., R.D. Jackson, Idso, S.B., Reginato R.J. 1981. Multidate spectral reflectances as predictors of yield in water stressed wheat and barley. Int. J. Remote Sens. 2(1):4348. Pinty, B., Verstraete, H. M. 1992. GEMI”: a non-linear index to monitor global vegetation from satellites. Vegetation. 101, 15-20. Price, J.C. 1990. On the information-content of soil reflectance spectra. Remote Sens. Environ. 33 (2): 113-121. Qi, J., Chehbouni, A., Huete, A. R., Kerr, Y. H. 1994. Modified Soil Adjusted Vegetation Index (MSAVI)," Remote Sens. Environ vol. 48, pp. 119-126. Qi, J., Huete, A. R., Moran, M. S., Chehbouni, A., and Jackson R. D., 1993. Interpretation of vegetation indices derived from multi-temporal SPOT images. Remote Sens. Environ.. 44, 89-101. Qi, J., Kerr, Y., and Chehbouni, A. 1994 "External Factor Consideration in Vegetation Index Development," in Proc. of Physical Measurements and Signatures in Remote Sensing,ISPRS, 723- 730. Qi, J., Moran, M.S. Cabot, F. Dedieu, G. 1995. Normalization of sun/view angle effects using spectral albedo-based vegetation indexes. Remote Sens. Environ 52(3):207-217. Qi, J., Huete,A.R., Moran,M.S., Chehbouni,A., and Jackson,R.D. Interpretation of vegetation indices derived from multitemporal SPOT images. Remote Sens. Environ. 44:89-101 Railyan, V. Ya., Korobov, R. M., 1993. Red edge structure of canopy reflectance spectra of Triticale. Remote Sens. Environ.. 46, 173-182. Raun, W.R., Solie, J.B., Johnson, G.V. Stone, M.L., Mullen, R.W., Freeman, W.E. Thomason, and E.V. Lukina. 2001. Improving nitrogen use efficiency in cereal grain production with optical sensing and variable-rate application. Agron. J. 94:815–820. Reginato, R.J., and J. Howe, 1985. Irrigation Scheduling using crop indicators Journal of Irrigation and Drainage Engineering. ASCE, 111(2):125-133. Richardson, A. J., Everitt, J. H. 1992. Using spectra vegetation indices to estimate rangeland productivity, Geocarto International. vol. 1, pp. 63-69. Richardson, A. J. Wiegand, C. L. 1977. Distinguishing vegetation from soil background Information. Photogrammetric Engineering and Remote Sensing, vol. 43, pp. 1541-1552. Richardson, A. J., Wiegand, C. L., Gausman, H. W., Cuellar, J. A. Gerbermann A. H. 1975. Plant, soil, and shadow reflectance components of row crops. Photogrammetric. Engineering and Remote Sensing 41:1401-1407. Richardson, A.J., Meneges, R.M. Nixon, P.R. 1985. Distinguishing weed from crop plants using video remote-sensing. Photogrammetric Engineering & Remote Sensing, 51(11):1785-1790. Ridao E., Conde, J. R. Inés Mìnguez, M. 1998. Estimating fAPAR from Nine Vegetation Indices for Irrigated and Nonirrigated Faba Bean and Semileafless pea Canopies. Remote Sens. Environ.. 66, 87-100. Ritchie, J.T., Wei, J. 2000. Models of kernel number in maize. p. 75–88. In M.E. Westgate and K.J. Boote (ed.) Physiology and modeling kernel set in maize. CSSA Spec. Publ. 29. CSSA and ASA, Madison, WI. Rondeaux, G., Steven, M., Baret, F. 1996. Optimization of SoilAdjusted Vegetation Indicies Remote Sens. Environ. 55 : 95-107. Rouse, J. W., Haas, R. H., Schell, J. A., and Deering, D. W. 1973. Monitoring vegetation systems in the great plains with ERTS," _Third ERTS Symposium, NASA SP-351_, vol. 1, pp.309-317. Schepers, 2001. In-season prediction of potential grain yield in winter wheat using canopy reflectance, Agronomy Journal, 93(1):131-138. Schepers, J.S.,Francis, D.R. 1998. Precision agriculture– What's in our future. Commun. Soil Sci. Plant Anal. 29:1463–1469. Schepers, J.S., Francis, D.R., Vigil, M., Below, F.E. 1992. Comparison of corn leaf nitrogen concentration and chlorophyll meter readings, Commun. Soil Sci. Plant Anal. 23(17-20):2173-2187. Schepers, J.S., Blackmer, T.M., Wilhelm, W.W., Resende, M. 1996. Transmittance and reflectance measurements of corn leaves from plants with different nitrogen and water supply. Journal of Plant Physiology, 148(5):523-529. Sellers, P. J., 1985. Canopy reflectance, photosynthesis and transpiration. Int. J. Remote Sens.. 6, 1335-1372. Serrano, L., Filella, I., Penuelas, J. 2000. Remote Sensing of Biomass and Yield of Winter Wheat under Different Nitrogen Supplies. Crop Sci. 40: 723-731. Shanahan, J. F., Schepers, J. S., Francis, D.D., Varvel, G. E., Wilhelm, W. W., Tringe, J. M., Schelemmer, M. R., Major, D. J. 2001. Use of Remote Sensing Imagery to Estimate corn Grain Yield. Agron. J. 93: 583589. Strahler, A. 1997, Vegetation canopy reflectance modeling – Recent developments and remote sensing perspectives, Remote Sensing Reviews, vol. 15, 179-194. Tian, L., J.F. reid, Hummel, J.W. 1999. Development of a precision sprayer for site-specific sorghum growth and yield variations using air-borne multispectral digital imagery. Trans. ASAE, 43(6):1927-1938. Tucher, C.J., B.N. Holben, Jr., J.H. Elgin, and J.E. McMurtrey III, 1981. Remote sensing of total dry matter accumulation in winter wheat, Remote Sens. Environ, 13:461. Tucker, C.J. 1979. Red and photographic infrared linear combinations for monitoring vegetation.Remote Sens. Environ. 8:127–150. Turner, N.C., Prasertsak, P., Setter, T. L. 1994. Plant Spacing, Density, and Yield of Wheat Subjected to Postanthesis Water Deficits. Crop Sci. 34:741-748 Verhoef, W., 1984. Light scattering by leaf layers with application to canopy reflectance modelling: the SAIL model. Remote Sens. Environ.. 16, 125-141. Wang, D., Poss, J.A.,.Donovan, T.J, Shannon, M.C., Lesch, S.M. 2002b. Biophysical properties and biomass production of elephant grass under saline conditions, Journal of Arid Environments, 52(4):447-456 Wheeler, T.A., Kaufman, H.W., Baugh, B. ,Kidd, P. Schuster, G., Siders. 1999. Comparisons of variable and single-rate applications of aldicarb on cotton yield in fields infested with Meloidogyne incognita. J. Nematology 31:700–708. Wiegand, C. L., Richardson A. J., Escobar, D. E., Gerbermann, A. H., 1991. Vegetation Indicies in Crop Assesments. Remote Sens. Environ.. 35, 105-119. Wiegand, C.L., Gerbermann, A.H. , Gallo, K.P., Blad, B.L., Dusek, D.1990. Multisite analyses of spectral biophysical data for corn, Remote Sens. Environ, 35:105-119. Wiegand, C.L., Richardson, A. J., Escobar, D. E., Gerbermann, A. H. 1991. Vegetation Indices in Crop Assesments. Remote Sens. Environ. 35 : 105-119. Wiegand,C.L., Maas,S.J., Aase,J.K., Hatfield,J.L., Pinter Jr.,P.J., Jackson,R.D., Kanemasu,E.T. and Lapitan,R.L. 1992. Multisite analyses of spectral-biophysical data for wheat. Remote Sens.Environ. 42:1-21 53 B. Basso et al. - Rivista Italiana di Agrometeorologia 36-53 (1) 2004 Willers, J.L., Seal, M.R. Luttrel, R.G. 1999. Remote sensing, line-intercept sampling for tarnished plant bugs (Heteroptera: Miridae) in Mid-South cotton, Journal of Cotton Science, 3:160-170. Wright, D. L. Jr, 2003. Using Remote Sensing to Manage Wheat Grain Protein. Final report for Utah State University Affiliated Research Ceneter, Earth Science Applications Directorate NASA Science Enterprise. Pp. 1-8. 54 Yoshioka, H., Miura, T., Huete, A. R., Ganapol B. D., 2000. Analysiscof Vegetation Isolines in Red-NIR Reflectance Space. Remote Sens. Environ.. 74, 313-326. Zipoli, G., Grifoni, D., 1994. Panicle contribution to bidirectional reflectance factors of a wheat canopy. Int. J. Remote Sens.. 15, 3309-3314. Guida per gli Autori I manoscritti inviati devono riferirsi a materiale originale, non pubblicato altrove, se non in fase strettamente preliminare. Un autore per ogni pubblicazione dovrà impegnasi ad acquistare 25 reprint. Almeno uno degli autori, per ogni pubblicazione, dovrà essere socio dell’AIAM. Invio dei manoscritti I manoscritti devono essere inviati (i) in formato cartaceo (2 copie) e (ii) via E-mail a: Roberto Confalonieri Dipartimento di Produzione Vegetale Università degli Studi di Milano via Celoria 2 20133 Milano E-mail: [email protected] Tel. 02 50316592 Fax 02 50316575 L'invio dei manoscritti presuppone l'approvazione degli stessi da parte di tutti gli Autori. I manoscritti accettati per la pubblicazione devono essere spediti esclusivamente su supporto magnetico o via E-mail. I file possono essere spediti su floppy disk, su ZIP disk o su CD ROM. I file possono essere in formato .doc, .rtf per quanto riguarda il testo, .jpg per quanto riguarda le figure. Preparazione dei manoscritti Gli articoli possono essere in Italiano o in Inglese. Per la stesura del manoscritto seguire l'ordine: - Titolo evitare titoli troppo lunghi1. Ad esempio nel caso di un articolo che parla delle relazioni esistenti fra la circolazione a scala sinottica e le migrazioni dei lepidotteri nottuidi è preferibile il titolo “Circolazione sul Mediterraneo e migrazioni stagionali dei lepidotteri nottuidi” al titolo “Analisi delle strutture circolatorie a scala sinottica sull’area del Mediterraneo e dei relativi effetti sulle migrazioni stagionali dei lepidotteri nottuidi” Nel caso l’articolo sia in lingua italiana dovrà essere inviata anche una versione del titolo in lingua inglese; il contrario si avrà nel caso di articolo in lingua inglese) - Autori - Indirizzo completo degli Autori ed ente di appartenenza (includere i numeri di telefono, di fax e l'indirizzo di posta elettronica; indicare chiaramente l'Autore di riferimento per la corrispondenza con l'Editor) - Riassunto (non deve superare le 400 parole. Nel caso l’articolo sia in lingua italiana dovrà essere inviata anche una versione del riassunto in lingua inglese; il contrario si avrà nel caso di articolo in lingua inglese) - Parole chiave (normalmente da 3 a 6; escludere le parole già incluse nel titolo. Nel caso l’articolo sia in lingua italiana dovrà essere inviata anche una versione delle parole chiave in lingua inglese; il contrario si avrà nel caso di articolo in lingua inglese) - Introduzione - Materiali e Metodi - Risultati - Discussione - Conclusioni - Ringraziamenti - Appendice - Bibliografia - Tabelle (precedute dalla didascalia) - Figure (precedute dalla didascalia) Nel caso l’articolo sia in lingua italiana dovrà essere inviata anche una versione delle didascalie in lingua inglese; il contrario si avrà nel caso di articolo in lingua inglese. In alternativa i paragrafi "Risultati" e "Discussione" possono essere accorpati, come anche i paragrafi “Discussione” e Conclusioni”. I paragrafi "Ringraziamenti" e "Appendice" sono facoltativi. Utilizzare sempre l'allineamento a sinistra e non scrivere su due o più colonne. Non utilizzare collegamenti tra varie parti del testo. Limitare l'uso di sottoparagrafi all'interno di quelli sopra elencati solo ai casi in cui questi siano strettamente necessari. Utilizzare interlinea doppia e lasciare ampi margini (a destra, a sinistra, sopra e sotto). Numerare tutte le pagine e le righe in ogni pagina. Non numerare i paragrafi. Una versione definitiva ed impaginata degli articoli pronti per la pubblicazione sarà inviata agli Autori per la segnalazione di eventuali errori verificatisi nel processo di editing e/o di eventuali correzioni. Tabelle Le tabelle devono essere in bianco e nero. Le tabelle possono essere inviate anche su file di testo o, preferibilmente, su foglio elettronico excel. Tenere presente che la tabella avrà, una volta pubblicata, solo le righe (di delimitazione tra campi) orizzontali. Le tabelle devono essere autoesplicative nel senso che la relativa didascalia dovrà essere in grado di fornire tutte le spiegazioni necessarie alla comprensione dei contenuti della tabella stessa. In aggiunta possono essere presenti note che compariranno a fondo tabella. Le tabelle devono essere numerate in accordo all'ordine con il quale sono citate nel testo. Tutte le tabelle devono essere citate nel testo. Figure Le figure devono essere in bianco e nero. La stampa di figure a colori comporta un costo a carico degli Autori di € 50 per pagina a colori. Le figure devono essere autoesplicative nel senso che la relativa didascalia dovrà essere in grado di fornire tutte le spiegazioni necessarie alla comprensione dei contenuti della figura. Le figure devono essere numerate in accordo all'ordine con il quale sono citate nel testo. Tutte le figure devono essere citate nel testo. Si prega infine di limitare al minimo la presenza di scritte sulle immagini. Tenere ben presente che le figure, al momento dell’impaginazione, potranno essere ridotte fino al 50% delle loro dimensioni originali e che, anche in questo caso, dovranno conservare una perfetta leggibilità. Riferimenti bibliografici Tutti i riferimenti citati nel testo devono essere riportati nel paragrafo "Bibliografia" e tutti i riferimenti presenti in "Bibliografia" devono essere citati nel testo. Nel testo i cognomi degli Autori devono essere citati senza il nome e senza l’iniziale (es: Paperino, 2003 e non Paolino Paperino, 2003 e nemmeno P. Paperino, 2003). Se nel testo viene citato un lavoro di due Autori, utilizzare la notazione "Paperone e Paperino, 2003" (non utilizzare "and" o "&" tra i due nomi). Se nel testo viene citato un lavoro con più di due Autori, utilizzare la notazione "Paperone et al., 2003". Si raccomanda inoltre di non utilizzare questa notazione nel paragrafo "Bibliografia": in questo caso tutti gli Autori devono infatti essere elencati. Nel caso nel testo si faccia riferimento a più lavori nello stesso punto, questi dovranno essere elencati in ordine cronologico, dal meno recente al più recente (esempio; Paperone, 2002; Paperino, 2004). Nel paragrafo "Bibliografia" i riferimenti sono da citare in ordine alfabetico (iniziale del cognome del primo Autore). Nel caso l'Autore di un lavoro compaia con co-Autori in altri lavori, citare prima i lavori nei quali compare come singolo Autore (ordinandoli cronologicamente), poi i lavori in cui compare con un solo co-Autore (ordinandoli cronologicamente) e infine quelli in cui compare con più di un co-Autore (ordinandoli cronologicamente). Per lavori dello stesso Autore(i) pubblicati nello stesso anno, utilizzare la notazione 2003a, 2003b, ecc. Per le riviste, le citazioni dovranno essere stese come segue: Bianchi, R., Colombo, B., Ferretti, N., 2003. Titolo dell'articolo. Nome della rivista, numero della rivista, pagine. Per Atti di Convegni: Bianchi, R., Colombo, B., Ferretti, N., 2003. Atti del "Nome del Convegno", luogo presso il quale si è tenuto il Convegno, data(e) nelle quali si è tenuto il convegno, pagine. Per libri: Bianchi, R., Colombo, B., Ferretti, N., 2003. Titolo del libro. Da chi è stato pubblicato, Luogo presso il quale è avvenuta la pubblicazione. Per libri con più Autori: Bianchi, R., Colombo, B., Ferretti, N., 2003. Titolo del capitolo. In: Baudo, P. (Ed.), Titolo del volume. Da chi è stato pubblicato, Luogo presso il quale è avvenuta la pubblicazione, pagine. Può essere citato materiale "in stampa". Non sono accettati riferimenti a lavori presentati in occasione di convegni, seminari e simili come poster o comunicazioni orali e non pubblicati nei relativi atti. 55 Rivista Italiana di Agrometeorologia Periodico quadrimestrale edito da Associazione Italiana di Agrometeorologia (AIAM) www.agrometeorologia.it