The research and application during the PhD was targeted to the conception, development, calibration and validation of a model that would allow to calculate, at a daily timescale, the basin hydrogeological water balance with greater accuracy than what you can be achieved with current models. In order to attain this purpose the procedure currently used in agriculture for irrigation water metering was extended to the entire basin. The procedure uses crop coefficient (Kc) to correct the potential evapotranspiration estimated by the equation of Penman Monteith; in particular we computed a daily basin unique crop coefficient which is the weighted average of all crop coefficients of the various land-use classes, coefficients varying during the year according to the phenological stages of the crops (initial, development, intermediate, final). FAO's notebook n°56 (Allen et al., 1998) was the primary source of values of Kc and durations of phenological stages but research was necessary to find the respective values for non-agricultural areas. The model contemplated a further correction of the potential evapotranspiration taking into account the actual surface area through the steepness coefficient (the ratio between actual and projected surface); precipitation values were also corrected but using the inverse of the coefficient of steepness; the calculation of the runoff was obtained using the method of the curve number (USDA, 1972) adapted for modeling in continuous (Williams, 2000). A soil water balance allowed, through various parameters, to estimate at a daily timescale the 3 unknowns parameters of the budget, i.e. the actual evapotranspiration, the variation of the soil water reserve and the deep percolation. In particular the deep percolation (DP) was the parameter that permitted to calibrate and validate the model since it was compared to trends and level values of the base-flow water stream, measured on a weekly frequency at the closure of the basin. The flow measurements were conducted for all the years in which the balances were calculated (2013 2014 2015 and 2016). Weekly measurements of piezometric levels of four wells were also carried out and the mean trend of the four acquifers has been compared with the DP as well. The area under investigation was the basin of the Fosso Santa Maria degli Angeli, a small basin of about 14 km2 located south of Urbino, Marche Region, central Italy. The data confirms the reliability of the model as multiannual values of modeled DP and measured base-flow differ of about only 10%; to annual timescale there is agreement between the trends of modeled DP, measured base-flow and measured groundwater levels, both during increase than decrease; in particular, the delay of the base-flow ascent against DP identifies the time necessary for the water to go through the underground path and arrive to the closing section of the basin. The model also permitted to well characterize the dynamics of water within the soil through the parameter of soil water reservoir variation (∆R). Flow measurements were also performed at the closing section of the middle-upper Metauro River basin, of which the Santa Maria degli Angeli is a left hand sub-basin tributary, in order to apply the hydrological model to a greater extension. For the purposes of the water balance study it was necessary the tracking of land-use maps for the years of investigation, and two maps for the years 1955 and 1997 in order to study the changes in the basin land-use were traced too. The daily phenological water balance is suitable for qualitative and quantitative studies of water resources at catchment level or other territorial demarcation and can be further improved through a better definition of the Kc values and durations of phenological stages, the use of double crop coefficient instead of the single crop coefficient and an exact attribution of the conduction of the agricultural areas (that is the data changing year by year among the land-use classes areas).
Il lavoro di ricerca e applicativo svolto durante il dottorato è stato mirato alla ideazione, allo sviluppo, alla calibrazione e alla validazione di un modello che consentisse di calcolare, a scala temporale giornaliera, il bilancio idrogeologico a livello di bacino con una maggior accuratezza rispetto a quella che si riesce a raggiungere con gli attuali modelli. A tal fine si è estesa a livello di bacino la procedura correntemente utilizzata in agricoltura per il dosaggio dell’acqua di irrigazione, procedura che impiega dei coefficienti colturali (Kc) per correggere la evapotraspirazione potenziale stimata con l’equazione di Penman Monteith; in particolare si è calcolato un coefficiente colturale unico giornaliero di bacino che è la media pesata di tutti i coefficienti colturali delle varie classi di uso del suolo, coefficienti che variano durante l’anno secondo le fasi fenologiche delle colture (iniziale, sviluppo, intermedia, finale). Il quaderno FAO n°56 (Allen et al., 1998) è stato la fonte primaria dei valori dei Kc e delle durate delle fasi fenologiche ma si è resa necessaria un’attività di ricerca per trovare i rispettivi valori per le aree non agricole. Il modello ha previsto un’ulteriore correzione della evapotraspirazione potenziale che tenesse conto della superficie reale del territorio attraverso il coefficiente di acclività (pari al rapporto tra superficie reale e proiettata); anche i valori di precipitazione sono stati corretti ma mediante l’inverso del coefficiente d’acclività; il calcolo del runoff è stato ottenuto mediante il metodo del curve number (USDA, 1972) adattato per la modellazione in continuo (Williams, 2000); infine è stato realizzato un bilancio idrico del suolo che ha consentito attraverso vari parametri di stimare a scala giornaliera le 3 incognite del bilancio, ovvero la evapotraspirazione reale, la variazione della riserva idrica, la percolazione profonda. In particolare la percolazione profonda (DP, deep percolation) è stato il parametro che ha consentito di calibrare e validare il modello poiché è stato confrontato a livello di trend e di valori con il deflusso di base del corso d’acqua del bacino, misurato settimanalmente alla chiusura del bacino mediante misure di portata per tutti gli anni in cui si è calcolato il bilancio (anni 2013, 2014, 2015 e 2016). Contemporaneamente sono state effettuate misure settimanali di livello piezometrico in quattro pozzi e l’andamento mediato dei quattro è stato altresì confrontato con la DP. Il territorio oggetto di indagine è stato il bacino del Fosso Santa Maria degli Angeli, piccolo bacino di circa 14 km2 posto a sud di Urbino, centro-Italia. I dati confermano la bontà del modello poiché i valori pluriennali di DP e deflusso differiscono appena del 10%; a scala temporale annuale c’è accordo tra i trends di DP, deflusso di base e falde acquifere, sia nelle fasi di salita che di discesa; in particolare il ritardo nella risalita del deflusso di base rispetto alla DP individua il tempo necessario all’acqua per effettuare il percorso ipogeo fino alla sezione di chiusura del bacino. Il modello ha poi consentito di ben caratterizzare le dinamiche idriche all’interno del suolo tramite il parametro della variazione della riserva idrica (∆R). Sono state inoltre effettuate misure di portata alla sezione di chiusura del bacino del fiume medio-alto Metauro, di cui il Santa Maria degli Angeli è sottobacino in sinistra idrografica, allo scopo di applicare il modello a un bacino di estensione maggiore. Ai fini del bilancio il lavoro ha anche previsto la realizzazione di carte di uso del suolo per gli anni di indagine, nonché due relative agli anni 1955 e 1997 al fine di studiare le variazioni di uso del suolo nel bacino. Il bilancio idrogeologico giornaliero su base fenologica è adatto per studi qualitativi e quantitativi delle risorse idriche a livello di bacino o di altra delimitazione territoriale e può essere ulteriormente migliorato attraverso una migliore definizione dei valori dei Kc e delle durate delle fasi fenologiche, l’utilizzo del double crop coefficient in luogo del single crop coefficient e una esatta attribuzione della tipologia di conduzione alle aree agricole (che è il dato che varia di anno in anno tra le superfici delle classi di uso del suolo).
Proposta di una nuova metodologia su base fenologica per la valutazione del bilancio idrogeologico giornaliero di bacino: calibrazione e validazione del modello attraverso misure sperimentali nel medio e alto bacino del Fiume Metauro
BARTOLUCCI, MICHELE
2018
Abstract
The research and application during the PhD was targeted to the conception, development, calibration and validation of a model that would allow to calculate, at a daily timescale, the basin hydrogeological water balance with greater accuracy than what you can be achieved with current models. In order to attain this purpose the procedure currently used in agriculture for irrigation water metering was extended to the entire basin. The procedure uses crop coefficient (Kc) to correct the potential evapotranspiration estimated by the equation of Penman Monteith; in particular we computed a daily basin unique crop coefficient which is the weighted average of all crop coefficients of the various land-use classes, coefficients varying during the year according to the phenological stages of the crops (initial, development, intermediate, final). FAO's notebook n°56 (Allen et al., 1998) was the primary source of values of Kc and durations of phenological stages but research was necessary to find the respective values for non-agricultural areas. The model contemplated a further correction of the potential evapotranspiration taking into account the actual surface area through the steepness coefficient (the ratio between actual and projected surface); precipitation values were also corrected but using the inverse of the coefficient of steepness; the calculation of the runoff was obtained using the method of the curve number (USDA, 1972) adapted for modeling in continuous (Williams, 2000). A soil water balance allowed, through various parameters, to estimate at a daily timescale the 3 unknowns parameters of the budget, i.e. the actual evapotranspiration, the variation of the soil water reserve and the deep percolation. In particular the deep percolation (DP) was the parameter that permitted to calibrate and validate the model since it was compared to trends and level values of the base-flow water stream, measured on a weekly frequency at the closure of the basin. The flow measurements were conducted for all the years in which the balances were calculated (2013 2014 2015 and 2016). Weekly measurements of piezometric levels of four wells were also carried out and the mean trend of the four acquifers has been compared with the DP as well. The area under investigation was the basin of the Fosso Santa Maria degli Angeli, a small basin of about 14 km2 located south of Urbino, Marche Region, central Italy. The data confirms the reliability of the model as multiannual values of modeled DP and measured base-flow differ of about only 10%; to annual timescale there is agreement between the trends of modeled DP, measured base-flow and measured groundwater levels, both during increase than decrease; in particular, the delay of the base-flow ascent against DP identifies the time necessary for the water to go through the underground path and arrive to the closing section of the basin. The model also permitted to well characterize the dynamics of water within the soil through the parameter of soil water reservoir variation (∆R). Flow measurements were also performed at the closing section of the middle-upper Metauro River basin, of which the Santa Maria degli Angeli is a left hand sub-basin tributary, in order to apply the hydrological model to a greater extension. For the purposes of the water balance study it was necessary the tracking of land-use maps for the years of investigation, and two maps for the years 1955 and 1997 in order to study the changes in the basin land-use were traced too. The daily phenological water balance is suitable for qualitative and quantitative studies of water resources at catchment level or other territorial demarcation and can be further improved through a better definition of the Kc values and durations of phenological stages, the use of double crop coefficient instead of the single crop coefficient and an exact attribution of the conduction of the agricultural areas (that is the data changing year by year among the land-use classes areas).File | Dimensione | Formato | |
---|---|---|---|
phd_uniurb_269933.pdf
accesso aperto
Tipologia:
DT
Licenza:
Creative commons
Dimensione
5.52 MB
Formato
Adobe PDF
|
5.52 MB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.