University of Twente Student Theses
Local and catchment scale validation of soil hydraulic pedotransfer functions for an Indonesian watershed
Oldhoff, R.J.J. (2015) Local and catchment scale validation of soil hydraulic pedotransfer functions for an Indonesian watershed.
|
PDF
5MB |
| Abstract: | Hydrological models are important tools used for many different purposes. In order to accurately model the hydrological processes in a catchment, information on the soil properties are of great importance. Data on the soil hydraulic properties can be obtained by conducting field work, this is however costly and time consuming. An alternative to field measurements is to use pedotransfer functions (PTFs). A PTF is an empirical relationship between easily obtainable soil characteristics and a soil hydraulic parameter. PTFs have been developed for a range of parameters. For this thesis, PTFs for the saturated hydraulic conductivity (Ks) and the available water content (AWC) are researched. Models are often very sensitive to the hydraulic conductivity and moderately sensitive to AWC, which makes an accurate estimation of these parameters important. A problem with PTF application is that PTFs are empirically determined relations for a specific area. Using a PTF for an area with other climatological and geographical characteristics can result in poor performance. The success of extrapolation of a PTF depends on the comparability of the soils. Tropical soils often have a different composition and have different hydraulic behaviour compared to temperate soils. Application of temperate soil PTFs on tropical soils might result in poor performance. Furthermore, not a lot of tropical soil PTFs are available from literature. The objective of this research is to determine whether Ks and AWC can be accurately approximated using PTFs, by analysing their performance at both the local scale and catchment scale for an Indonesian region. Four published PTFs for Ks and AWC are validated on a data set containing 91 soil samples that were collected during field work in four sub-catchments of the Upper Bengawan Solo catchment, located on Java Indonesia. This showed that the AWC is predicted very poorly, with R2 values below zero for all selected PTFs. For Ks PTFs better results were found. Two PTFs, the Wösten and Rosetta-3 PTFs, predict the Ks moderately accurate, with R2 values of 0.28 and 0.39, respectively. New PTFs for both AWC and Ks were made for the dataset, using Multiple Linear Regression. For the best performing PTFs for AWC and Ks, R2 values of 0.37 and 0.55 were found, respectively. Though these are not very high R2 values, they are significantly higher than the published PTFs. The new PTFs are sufficiently accurate for Ks and AWC estimation at the local scale. The SWAT model was set up for the Keduang, a sub-catchment of the Upper Bengawan Solo catchment. With a monthly time step, the catchment outflow was modelled. Eleven cases were defined. One based on measured inputs, six using the Digitalized Soil Map of the World (DSMW) and the new PTFs, the Wösten and Rosetta-3 Ks PTFs, and the Van den Berg AWC PTF for soil inputs. One case using the FAO DSMW in combination with lookup tables for Ks and AWC as input, a case using the FAO DSMW where the Ks and AWC were calibrated and finally two uncalibrated cases. Uncalibrated model results are moderately accurate, with Nash Sutcliffe (NS) values of 0.52 and 0.54. For Ks the model outputs indicate that the model accuracy is not significantly different when using measured values as opposed to PTFs. For each Ks PTF case a NS value of around 0.84 was obtained. Even though the model is especially sensitive to Ks, the small difference in PTF estimated values and measured values of Ks result in equal model accuracy for the different cases. The use of AWC PTFs resulted in slightly lower NS values, though still the differences in model accuracy are low. For the Keduang the tested PTFs can be used as an alternative to field measurements for hydrological modelling. To conclude, at the local scale PTF accuracy is not very high, but at the catchment scale they perform well. At the local scale the Wösten and Rosetta-3 PTFs can be used to predict Ks. AWC PTFs show insufficient accuracy at the local scale. At the catchment scale, the Wösten and Rosetta-3 Ks PTFs and the Oldhoff AWC and Ks PTFs are validated. It is recommended to use the Oldhoff PTFs in the Upper Bengawan Solo catchment. More research is needed on the effect of PTF input on hydrological state variables, such as soil moisture content. The effect of catchment soil heterogeneity also requires more research. |
| Item Type: | Essay (Master) |
| Faculty: | ET: Engineering Technology |
| Subject: | 56 civil engineering |
| Programme: | Civil Engineering and Management MSc (60026) |
| Link to this item: | https://purl.utwente.nl/essays/68820 |
| Export this item as: | BibTeX EndNote HTML Citation Reference Manager |
Show download statistics for this publication
Show download statistics for this publicationRepository Staff Only: item control page