Assessing MODIS evapotranspiration data for hydrological modelling in South Africa
- Mazibuko, Sbongiseni Christian
- Authors: Mazibuko, Sbongiseni Christian
- Date: 2017
- Subjects: Evapotranspiration , Evapotranspiration -- Measurement , Hydrologic models , Hydrologic models -- South Africa , MODIS (Spectroradiometer)
- Language: English
- Type: Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10962/8009 , vital:21334
- Description: Evapotranspiration as a major component of the water balance has been identified as a key factor in hydrological modelling. Water management can be improved by means of increased use of reliable methods for estimating evapotranspiration. The limited availability of measured climate and discharge data sets, particularly in the developing world, restricts the reliability of hydrological models in these regions. Furthermore, rapid changes in hydrological systems with increasing development mean uncertainties in water resource estimation are growing. These changes are related to the modification of catchment hydrological processes with increasing human activity. Dealing with data uncertainty and quantifying the impacts of catchment activities are significant challenges that scientists in the field of hydrology face today. Uncertainties in hydrometeorological data are associated with poor observation networks that provide data at point scales which are not adequately representative of the inherent heterogeneity within catchment processes. Using uncertain data in model applications reduces the predictive power of hydrological models as well as the ability to validate the model outcomes. This study examines the potential of using remote sensing-based evapotranspiration data to reduce uncertainty in the climatic forcing data and constraining the output of a rainfall-runoff hydrological model. It is common to use fixed seasonally variable potential evapotranspiration (PET) instead of temporally varying PET data as inputs to standard rainfall-runoff models. Part of the reason is that there are relatively few stations available to measure a variety of meteorological input data needed to compute PET, as well as the apparent lack of sensitivity of rainfall-runoff models to different types of PET inputs. As hydrometeorological data become more readily available through the use of earth observation systems, it is important to determine whether rainfall-runoff models are sensitive to time-varying PET derived from these earth observations systems. Further potential includes the use of actual evapotranspiration (ETa) from this type of data to constrain model outputs and improve model realism. It is assumed that a better representation of evapotranspiration demands could improve the efficiency of models, and this study explores some of these issues. The study used evapotranspiration estimates (PET and ETa) from the MOD16 global product with one of the most widely used hydrological models in South Africa. The investigation included applying the Pitman model in a number of case study catchments located in different climatic regions of the country. The main objectives of the study included (i) the establishment of behavioural model parameter sets that generate acceptable hydrological response under both naturalised and present-day conditions, (ii) the use of time-varying PET estimates derived from MOD16 data to force the model, and (iii) the use of MOD16 ETa estimates to constrain model-simulated ETa. Before examining the use of different PET forcing data in the model, a two-step modelling approached was employed both a single-run and an uncertainty version of the Pitman model. During the first step (using a single-run version), available information on catchment physical properties and regionalised groundwater recharge together with model calibration principles were used to develop model functionality understanding and establish initial parameter sets. The outcomes from the first step were used to define uncertain parameter ranges for the use in the uncertainty version of the Pitman model (second step). Further, catchment water uses were quantified to ensure comparability with present-day flow conditions represented by the stream flow records. The effects of forcing the Pitman model with MOD16-based time-varying PET data inputs were evaluated using static and dynamic sensitivity analysis approaches. In the static approach, parameter sets calibrated using fixed seasonal distributions of PET data remain unchanged when forcing the model with other forms of PET, whereas in the dynamic method, the model is recalibrated with changing PET inputs. In both approaches, model sensitivity was assessed by comparing objective function statistics of reference flow simulations with those simulations incorporating changing PET data inputs. The use of the MOD16 ETa data to constrain model- simulated evapotranspiration losses was conducted by calibrating the parameters such that the simulated-ETa matched the evapotranspiration loss estimated from the MOD16 data. Despite issues around model equifinality and significant uncertainty within water use information, the Pitman model simulations were generally satisfactory and compared with observed stream flow data where available. The use of time-varying PET data does not improve the efficiency of the model when both static and dynamic sensitivity approaches are used. This was highly expected with the static approach where fixed model parameter sets do not account for the changes in evapotranspiration demands. However, with the dynamic approach, it was difficult to conclude why the model efficiency did not improve given the flexibility of the model to achieve appropriate parameter sets to different forms of PET. The study noted that the insensitivity of the model to changes in PET demands could be due to uncertainties in the model structure and MOD16 data. Attempts to constrain the model-simulated actual evapotranspiration with MOD16 ETa estimates were hampered by large errors in the MOD16 data and resulted in the non-closure of the catchment annual water balance, even when likely errors in the other components of the water balance were accounted for. There is still a great deal of work that needs to be done to reduce uncertainties associated with the use of earth observation data in hydrological modelling. This study has identified some of the specific gaps within the application of evapotranspiration data from earth observation information. While the MOD16 data applied with the Pitman model did not achieve improved simulations, the study has demonstrated the enormous potential of the data product in the future should the identified uncertainties be resolved. Lastly, the investigation highlighted some of the possible model structural uncertainties specifically associated with the simplified soil-moisture accounting routines within the model. It is possible that amending the model structure through investigating the dynamics of the relationship between soil moisture and evapotranspiration losses would assist in the improved utilisation of earth observation products related to the MOD16 ET data.
- Full Text:
- Date Issued: 2017
- Authors: Mazibuko, Sbongiseni Christian
- Date: 2017
- Subjects: Evapotranspiration , Evapotranspiration -- Measurement , Hydrologic models , Hydrologic models -- South Africa , MODIS (Spectroradiometer)
- Language: English
- Type: Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10962/8009 , vital:21334
- Description: Evapotranspiration as a major component of the water balance has been identified as a key factor in hydrological modelling. Water management can be improved by means of increased use of reliable methods for estimating evapotranspiration. The limited availability of measured climate and discharge data sets, particularly in the developing world, restricts the reliability of hydrological models in these regions. Furthermore, rapid changes in hydrological systems with increasing development mean uncertainties in water resource estimation are growing. These changes are related to the modification of catchment hydrological processes with increasing human activity. Dealing with data uncertainty and quantifying the impacts of catchment activities are significant challenges that scientists in the field of hydrology face today. Uncertainties in hydrometeorological data are associated with poor observation networks that provide data at point scales which are not adequately representative of the inherent heterogeneity within catchment processes. Using uncertain data in model applications reduces the predictive power of hydrological models as well as the ability to validate the model outcomes. This study examines the potential of using remote sensing-based evapotranspiration data to reduce uncertainty in the climatic forcing data and constraining the output of a rainfall-runoff hydrological model. It is common to use fixed seasonally variable potential evapotranspiration (PET) instead of temporally varying PET data as inputs to standard rainfall-runoff models. Part of the reason is that there are relatively few stations available to measure a variety of meteorological input data needed to compute PET, as well as the apparent lack of sensitivity of rainfall-runoff models to different types of PET inputs. As hydrometeorological data become more readily available through the use of earth observation systems, it is important to determine whether rainfall-runoff models are sensitive to time-varying PET derived from these earth observations systems. Further potential includes the use of actual evapotranspiration (ETa) from this type of data to constrain model outputs and improve model realism. It is assumed that a better representation of evapotranspiration demands could improve the efficiency of models, and this study explores some of these issues. The study used evapotranspiration estimates (PET and ETa) from the MOD16 global product with one of the most widely used hydrological models in South Africa. The investigation included applying the Pitman model in a number of case study catchments located in different climatic regions of the country. The main objectives of the study included (i) the establishment of behavioural model parameter sets that generate acceptable hydrological response under both naturalised and present-day conditions, (ii) the use of time-varying PET estimates derived from MOD16 data to force the model, and (iii) the use of MOD16 ETa estimates to constrain model-simulated ETa. Before examining the use of different PET forcing data in the model, a two-step modelling approached was employed both a single-run and an uncertainty version of the Pitman model. During the first step (using a single-run version), available information on catchment physical properties and regionalised groundwater recharge together with model calibration principles were used to develop model functionality understanding and establish initial parameter sets. The outcomes from the first step were used to define uncertain parameter ranges for the use in the uncertainty version of the Pitman model (second step). Further, catchment water uses were quantified to ensure comparability with present-day flow conditions represented by the stream flow records. The effects of forcing the Pitman model with MOD16-based time-varying PET data inputs were evaluated using static and dynamic sensitivity analysis approaches. In the static approach, parameter sets calibrated using fixed seasonal distributions of PET data remain unchanged when forcing the model with other forms of PET, whereas in the dynamic method, the model is recalibrated with changing PET inputs. In both approaches, model sensitivity was assessed by comparing objective function statistics of reference flow simulations with those simulations incorporating changing PET data inputs. The use of the MOD16 ETa data to constrain model- simulated evapotranspiration losses was conducted by calibrating the parameters such that the simulated-ETa matched the evapotranspiration loss estimated from the MOD16 data. Despite issues around model equifinality and significant uncertainty within water use information, the Pitman model simulations were generally satisfactory and compared with observed stream flow data where available. The use of time-varying PET data does not improve the efficiency of the model when both static and dynamic sensitivity approaches are used. This was highly expected with the static approach where fixed model parameter sets do not account for the changes in evapotranspiration demands. However, with the dynamic approach, it was difficult to conclude why the model efficiency did not improve given the flexibility of the model to achieve appropriate parameter sets to different forms of PET. The study noted that the insensitivity of the model to changes in PET demands could be due to uncertainties in the model structure and MOD16 data. Attempts to constrain the model-simulated actual evapotranspiration with MOD16 ETa estimates were hampered by large errors in the MOD16 data and resulted in the non-closure of the catchment annual water balance, even when likely errors in the other components of the water balance were accounted for. There is still a great deal of work that needs to be done to reduce uncertainties associated with the use of earth observation data in hydrological modelling. This study has identified some of the specific gaps within the application of evapotranspiration data from earth observation information. While the MOD16 data applied with the Pitman model did not achieve improved simulations, the study has demonstrated the enormous potential of the data product in the future should the identified uncertainties be resolved. Lastly, the investigation highlighted some of the possible model structural uncertainties specifically associated with the simplified soil-moisture accounting routines within the model. It is possible that amending the model structure through investigating the dynamics of the relationship between soil moisture and evapotranspiration losses would assist in the improved utilisation of earth observation products related to the MOD16 ET data.
- Full Text:
- Date Issued: 2017
Linking satellite and point micrometeorological data to estimate : distributed evapotranspiration modelling based on MODIS LAI, Penman-Monteith and functional convergence theory
- Authors: Weideman, Craig Ivan
- Date: 2014
- Subjects: Plants -- Water requirements -- South Africa , Evaporation (Meteorology) -- Measurement , Satellite meteorology , Micrometeorology , Evapotranspiration , MODIS (Spectroradiometer)
- Language: English
- Type: Thesis , Masters , MSc
- Identifier: vital:4779 , http://hdl.handle.net/10962/d1012078 , Plants -- Water requirements -- South Africa , Evaporation (Meteorology) -- Measurement , Satellite meteorology , Micrometeorology , Evapotranspiration , MODIS (Spectroradiometer)
- Description: Recent advances in satellite sensor technology and micrometeorological instrumentation for water flux measurement, coupled with the expansion of automatic weather station networks that provide routine measurements of near-surface climate variables, present new opportunities for combining satellite and ground-based instrumentation to obtain distributed estimates of vegetation water use over wide areas in South Africa. In this study, a novel approach is tested, which uses satellite leaf area index (LAI) data retrieved by the Moderate Resolution Imaging Spectroradiometer (MODIS) to inform the FAO-56 Penman-Monteith equation for calculating reference evaporation (ET₀) of vegetation phenological activity. The model (ETMODIS) was validated at four sites in three different ecosystems across the country, including semi-arid savanna near Skukuza, mixed community grassland at Bellevue, near Pietermaritzburg, and Groenkop, a mixed evergreen indigenous forest near George, to determine potential for application over wider areas of the South African land surface towards meeting water resource management objectives. At Skukuza, evaluated against 170 days of flux data measured at a permanent eddy covariance (EC) flux tower in 2007, the model (ETMODIS) predicted 194.8 mm evapotranspiration relative to 148.9 mm measured fluxes, an overestimate of 31.7 %, (r² = 0.67). At an adjacent site, evaluated against flux data measured on two discrete periods of seven and eight days in February and May of 2005 using a large aperture scintillometer (SLS), ETMODIS predicted 27.4 mm and 6.7 mm evapotranspiration respectively, relative to measured fluxes of 32.5 and 8.2 mm, underestimates of 15.7 % and 18.3 % in each case (r² = 0.67 and 0.34, respectively). At Bellevue, evaluated against 235 days of evapotranspiration data measured using a surface layer scintillometer (SLS) in 2003, ETMODIS predicted 266.9 mm evapotranspiration relative to 460.2 mm measured fluxes, an underestimate of 42 % (r² = 0.67). At Groenkop, evaluated against data measured using a SLS over three discrete periods of four, seven and seven days in February, June and September/October respectively, ETMODIS predicted 9.7 mm, 10.3 mm and 17.0 mm evapotranspiration, relative to measured fluxes of 10.9 mm, 14.6 mm and 23. 9 mm, underestimates of 22.4 %, 11.2 % and 24.1 % in each case (r² = 0.98, 0.43 and 0.80, respectively). Total measured evapotranspiration exceeded total modelled evapotranspiration in all cases, with the exception of the flux tower site at Skukuza, where evapotranspiration was overestimated by ETMODIS by 31.7 % relative to measured (EC) values for the 170 days in 2007 where corresponding modelled and measured data were available. The most significant differences in measured versus predicted data were recorded at the Skukuza flux tower site in 2007 (31.7 % overestimate), and the Bellevue SLS flux site in 2003 (42 % underestimate); coefficients of determination, a measure of the extent to which modelled data are able to explain observed data at validation periods, with just two exceptions, were within a range of 0.67 – 0.98. Several sources of error and uncertainty were identified, relating predominantly to uncertainties in measured flux data used to evaluate ETMODIS, uncertainties in MODIS LAI submitted to ETMODIS, and uncertainties in ETMODIS itself, including model assumptions, and specific uncertainties relating to various inputs; further application of the model is required to test these uncertainties however, and establish confidence limits in performance. Nevertheless, the results of this study suggest that the technique is generally able to produce estimates of vegetation water use to within reasonably close approximations of measurements acquired using micrometeorological instruments, with r² values within the range of other peer-reviewed satellite remote sensing-based approaches.
- Full Text:
- Date Issued: 2014
- Authors: Weideman, Craig Ivan
- Date: 2014
- Subjects: Plants -- Water requirements -- South Africa , Evaporation (Meteorology) -- Measurement , Satellite meteorology , Micrometeorology , Evapotranspiration , MODIS (Spectroradiometer)
- Language: English
- Type: Thesis , Masters , MSc
- Identifier: vital:4779 , http://hdl.handle.net/10962/d1012078 , Plants -- Water requirements -- South Africa , Evaporation (Meteorology) -- Measurement , Satellite meteorology , Micrometeorology , Evapotranspiration , MODIS (Spectroradiometer)
- Description: Recent advances in satellite sensor technology and micrometeorological instrumentation for water flux measurement, coupled with the expansion of automatic weather station networks that provide routine measurements of near-surface climate variables, present new opportunities for combining satellite and ground-based instrumentation to obtain distributed estimates of vegetation water use over wide areas in South Africa. In this study, a novel approach is tested, which uses satellite leaf area index (LAI) data retrieved by the Moderate Resolution Imaging Spectroradiometer (MODIS) to inform the FAO-56 Penman-Monteith equation for calculating reference evaporation (ET₀) of vegetation phenological activity. The model (ETMODIS) was validated at four sites in three different ecosystems across the country, including semi-arid savanna near Skukuza, mixed community grassland at Bellevue, near Pietermaritzburg, and Groenkop, a mixed evergreen indigenous forest near George, to determine potential for application over wider areas of the South African land surface towards meeting water resource management objectives. At Skukuza, evaluated against 170 days of flux data measured at a permanent eddy covariance (EC) flux tower in 2007, the model (ETMODIS) predicted 194.8 mm evapotranspiration relative to 148.9 mm measured fluxes, an overestimate of 31.7 %, (r² = 0.67). At an adjacent site, evaluated against flux data measured on two discrete periods of seven and eight days in February and May of 2005 using a large aperture scintillometer (SLS), ETMODIS predicted 27.4 mm and 6.7 mm evapotranspiration respectively, relative to measured fluxes of 32.5 and 8.2 mm, underestimates of 15.7 % and 18.3 % in each case (r² = 0.67 and 0.34, respectively). At Bellevue, evaluated against 235 days of evapotranspiration data measured using a surface layer scintillometer (SLS) in 2003, ETMODIS predicted 266.9 mm evapotranspiration relative to 460.2 mm measured fluxes, an underestimate of 42 % (r² = 0.67). At Groenkop, evaluated against data measured using a SLS over three discrete periods of four, seven and seven days in February, June and September/October respectively, ETMODIS predicted 9.7 mm, 10.3 mm and 17.0 mm evapotranspiration, relative to measured fluxes of 10.9 mm, 14.6 mm and 23. 9 mm, underestimates of 22.4 %, 11.2 % and 24.1 % in each case (r² = 0.98, 0.43 and 0.80, respectively). Total measured evapotranspiration exceeded total modelled evapotranspiration in all cases, with the exception of the flux tower site at Skukuza, where evapotranspiration was overestimated by ETMODIS by 31.7 % relative to measured (EC) values for the 170 days in 2007 where corresponding modelled and measured data were available. The most significant differences in measured versus predicted data were recorded at the Skukuza flux tower site in 2007 (31.7 % overestimate), and the Bellevue SLS flux site in 2003 (42 % underestimate); coefficients of determination, a measure of the extent to which modelled data are able to explain observed data at validation periods, with just two exceptions, were within a range of 0.67 – 0.98. Several sources of error and uncertainty were identified, relating predominantly to uncertainties in measured flux data used to evaluate ETMODIS, uncertainties in MODIS LAI submitted to ETMODIS, and uncertainties in ETMODIS itself, including model assumptions, and specific uncertainties relating to various inputs; further application of the model is required to test these uncertainties however, and establish confidence limits in performance. Nevertheless, the results of this study suggest that the technique is generally able to produce estimates of vegetation water use to within reasonably close approximations of measurements acquired using micrometeorological instruments, with r² values within the range of other peer-reviewed satellite remote sensing-based approaches.
- Full Text:
- Date Issued: 2014
- «
- ‹
- 1
- ›
- »