The GAR(1) model with fragment method for hydrological drought risk assessment in semiarid regions / O modelo GAR(1) com método de fragmento para avaliação de risco de seca hidrológica em regiões semiáridas
The Northeast Brazil has been recognized as an area that will suffer greatly from the effects of variability and climate change. This will lead to a reduction of precipitation and streamflow in the region, causing greater pressure on the scarce water resources of the region, especially on the water...
| Autores: | , |
|---|---|
| Tipo de recurso: | artículo |
| Estado: | Versión publicada |
| Fecha de publicación: | 2019 |
| País: | Brasil |
| Institución: | Instituto Superior de Educação Vera Cruz (VeraCruz) |
| Repositorio: | Revista Veras |
| Idioma: | portugués |
| OAI Identifier: | oai:ojs2.ojs.brazilianjournals.com.br:article/3666 |
| Acceso en línea: | https://ojs.brazilianjournals.com.br/ojs/index.php/BRJD/article/view/3666 |
| Access Level: | acceso abierto |
| Palabra clave: | hydrological drought intermittent rivers reservoir operation optimization. |
| Sumario: | The Northeast Brazil has been recognized as an area that will suffer greatly from the effects of variability and climate change. This will lead to a reduction of precipitation and streamflow in the region, causing greater pressure on the scarce water resources of the region, especially on the water stored in the reservoirs. Optimization of the design and operation of multipurpose reservoir systems depends on the ability of synthetic streamflow generation models to reproduce the typical intermittent characteristics of semi-arid rivers. A Gamma Autoregressive – GAR(1) model have been tested and applied, for generating annual flows, coupled with the Fragment Method to disaggregate the annual flows to monthly ones. This coupled model was applied to four typical intermittent basins of the NE-Brazil, with drainage area varying from 410 to 5.695 Km². In order to analyze the performance of the model not only the statistical parameters (mean, variance, lag-1 serial correlation, etc.) of the historical and generated series were examined, but also a storage analysis by mean of the Sequent-Peak-Algorithm (SPA) was performed and additionally the preservation of the droughts and floods characteristics (duration, severity and magnitude) of the historical series was analyzed. This model was able to preserve the statistical parameters of the historical time series. However, when the generated synthetic flows were used to a storage analysis the model was not adequate to reproduce, particularly, the persistence (long periods of low and high flow) encountered in the historical series, which is fundamental by the reservoir design and for hydrological drought risk assessment. |
|---|