INTERCOMPARISON OF PROCESS-BASED PHYSICAL AND MATHEMATICAL HYDROLOGICAL MODELS IN DATA-SCARCE SEMI-ARID REGION OF ERITREA
Watershed models simulate natural hydrological and biogeochemical processes within watersheds as well as quantify the impact of human activities on these processes. Among them, rainfall-runoff models have been widely applied for generating hydrological responses using reanalysis datasets as forcing...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | Russian |
| Published: |
Russian Research Institute for Water Resources Integrated Management and Protection (RosNIIVKh)
2021-02-01
|
| Series: | Водное хозяйство России: проблемы, технологии, управление |
| Subjects: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Watershed models simulate natural hydrological and biogeochemical processes
within watersheds as well as quantify the impact of human activities on these processes. Among them, rainfall-runoff models have been widely applied for generating hydrological responses using reanalysis datasets as forcing variables in data-scare regions. In the present study, Soil and Water Assessment Tool model and rainfall-runoff model were employed to simulate streamflow from a small watershed with arid and semi-arid climate. As such, models that provide reliable streamflow predictions in the region as well as whose errors and
uncertainties are within acceptable ranges could be identified. The intercomparison of the models’ performances indicated that the Soil and Water Assessment Tool model relatively outperformed the rainfall-runoff model. However, while most of the statistical evaluations proved an acceptable performance of the Soil and Water Assessment Tool model, significant
amounts of uncertainties during calibration and validation procedures were noticed. Among the possible sources of errors, errors due to forcing variables were highly likely to be responsible for unsatisfactory performances of the selected models. In this regard, to minimize model uncertainty and thereupon improve its performance, ground-based data collection need to be boosted up. Besides, the study highlighted the need for further investigation on the
possible mechanisms of properly applying reanalysis datasets in arid and semi-arid regions. |
|---|---|
| ISSN: | 1999-4508 2686-8253 |