The final aim of this work is to provide a statistical evaluation of the accuracy level of different evapotranspiration calculation models compared to measured data and to find, if possible, trends that would favor one method over another based on the characteristics of the case study, with particular attention to more arid areas, where a more accurate estimate of evapotranspiration would allow for more conscious use of water resources. A statistical comparison is made between various evapotranspiration (ET) calculation methods over time and in different regions of the world, starting from data collected by Eddy Covariance micrometeorological stations, which are capable of measuring energy exchanges between vegetation and the atmosphere, and comparing them with estimates provided by existing software that offer open-source data and have very different characteristics in terms of temporal and spatial resolution, as well as theoretical models on which their foundations are based. On these assumptions, the following models are selected for this comparison: GLEAM, ERA5-Land, ETMonitor, and MODIS. The data to be compared come from seventeen micrometeorological stations distributed around the world, belonging to the ICOS and FLUXNET2015 networks, within the span years of 2010-2023, which is not the same for all stations. These are chosen to provide a reasonable representation of different ecosystems, with a significant range of aridity index values, between 0.1225 and 1.5367, after validating the data through verification of the energy balance closure for both incoming and outgoing energy. The evapotranspiration data used is the average of the measured value, the corrected value assuming a measurement error in the latent heat of evapotranspiration, and the corrected value, using the Bowen ratio formula, distributing the error both on the latent heat and on the sensible heat measurements.
L’obiettivo finale del lavoro è dare una valutazione su base statistica del livello di accuratezza dei diversi modelli di calcolo della evapotraspirazione rispetto ai dati misurati e di trovare, se possibile, delle linee di tendenza che facciano preferire un metodo all’altro rispetto alle caratteristiche del caso di studio, con particolare attenzione alle zone più aride ovvero dove, una stima sempre più corretta della evapotraspirazione, permetterebbe un utilizzo più consapevole della risorsa idrica Si opera una comparazione statistica tra diversi metodi di calcolo della evapotraspirazione (ET) nel tempo, e in diversi luoghi del globo, a partire dai dati raccolti dalle stazioni micrometeorologiche Eddy Covariance, in grado di misurare scambi di energia tra vegetazione e atmosfera, confrontandoli con le stime fornite da software già esistenti, che offrono dati open source e caratteristiche molto differenti tra loro in termini di risoluzione temporale e spaziale oltre che di modelli teorici su cui fondano le basi. Su questi presupposti, per questa comparazione, si selezionano i seguenti modelli: GLEAM, ERA5-Land, ETMonitor e MODIS. I dati da confrontare provengono da diciassette stazioni micrometeorologiche distribuite in tutto il mondo, appartenenti alle reti ICOS e FLUXNET2015, nell’arco temporale 2010-2023, non uguale per tutte le stazioni. Queste sono scelte in modo da avere una discreta rappresentatività di differenti ecosistemi, con una forbice notevole dell’indice di aridità, con valori compresi tra 0.1225 e 1.5367, dopo averne validato i dati tramite verifica della chiusura del bilancio energetico in entrata e in uscita. Il dato di evapotraspirazione utilizzato è la media tra il valore misurato, il valore corretto ipotizzando un errore di misura dell’energia latente di evapotraspirazione ed il valore corretto, tramite formula di Bowen, distribuendo l’errore sia sulla misura dell’energia latente sia del calore sensibile.
Confronto tra modelli di calcolo dell'evapotraspirazione in diverse situazioni climatiche e di ecosistemi
SALA, RICCARDO
2023/2024
Abstract
The final aim of this work is to provide a statistical evaluation of the accuracy level of different evapotranspiration calculation models compared to measured data and to find, if possible, trends that would favor one method over another based on the characteristics of the case study, with particular attention to more arid areas, where a more accurate estimate of evapotranspiration would allow for more conscious use of water resources. A statistical comparison is made between various evapotranspiration (ET) calculation methods over time and in different regions of the world, starting from data collected by Eddy Covariance micrometeorological stations, which are capable of measuring energy exchanges between vegetation and the atmosphere, and comparing them with estimates provided by existing software that offer open-source data and have very different characteristics in terms of temporal and spatial resolution, as well as theoretical models on which their foundations are based. On these assumptions, the following models are selected for this comparison: GLEAM, ERA5-Land, ETMonitor, and MODIS. The data to be compared come from seventeen micrometeorological stations distributed around the world, belonging to the ICOS and FLUXNET2015 networks, within the span years of 2010-2023, which is not the same for all stations. These are chosen to provide a reasonable representation of different ecosystems, with a significant range of aridity index values, between 0.1225 and 1.5367, after validating the data through verification of the energy balance closure for both incoming and outgoing energy. The evapotranspiration data used is the average of the measured value, the corrected value assuming a measurement error in the latent heat of evapotranspiration, and the corrected value, using the Bowen ratio formula, distributing the error both on the latent heat and on the sensible heat measurements.File | Dimensione | Formato | |
---|---|---|---|
2025_4_Sala.pdf
accessibile in internet per tutti
Descrizione: Confronto tra modelli di calcolo dell'evapotraspirazione in diverse situazioni climatiche e di ecosistemi
Dimensione
14.71 MB
Formato
Adobe PDF
|
14.71 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/236465