Resumo Modelos atmosféricos globais e regionais são importantes ferramentas para a gestão de recursos hídricos. Com a intenção de realizar previsão de vazões e subsidiar a operação de sistemas de abastecimento, esse estudo avaliou as previsões de precipitação de três modelos atmosféricos. Foram analisadas as previsões dos modelos ETA (resolução horizontal de 40 km, horizonte de 10 dias), BAM (resolução horizontal de 20 km, horizonte de 10 dias) e WRF (resolução horizontal de 5 km, horizonte de 3 dias) para as bacias do Sistema Cantareira, no Sudeste brasileiro. As previsões foram comparadas com observações de pluviômetros e de radar. Foram avaliados o coeficiente de correlação de Pearson, eficiência modificada de Kling-Gupta, coeficiente angular da regressão linear entre previsões e observações, REQM e PBIAS, além dos índices categóricos fBIAS, POD e FAR. Verificou-se que as correlações tendem a ser mais fortes nos primeiros dias de previsão (até o segundo dia à frente). As maiores correlações foram encontradas comparando valores acumulados para todo o horizonte de previsão. O modelo ETA apresentou forte tendência a subestimar as observações e o modelo BAM, de superestimá-las. O modelo WRF apresentou uma tendência leve de subestimação. A previsão de precipitação com modelos globais e regionais é um importante subsídio à operação de sistemas de reservatórios, entretanto, é essencial conhecer o comportamento dessas previsões a fim de minimizar imprecisões e maximizar seu valor na tomada de decisão.
The ever growing demand for water resources, along with concerns about the environmental, social and economical impact of the infrastructure expansion, reinforces the importance of management measures to ensure a better use of existing facilities. In this context, simulation and optimization models are important tools in water resources management. Although widely discussed in academia, such models face some resistance in its use by the system operators. This research proposes a decision support system to aid the operation of reservoir systems based on rainfall forecast from regional climate models. Whenever possible, the system used public access tools and it was designed to be easily replicated. The methodology was applied to the Cantareira system in order to optimize the downstream flows towards the PCJ basins, according to 2017's regulation. Rainfall forecast from the ETA, WRF and BAM climate models were analyzed. The rainfall was used as input for the SMAP runoff model, producing flow forecasting. These flow series served as input to a water allocation model, providing guidelines for the operation. The rainfall forecasts present considerable differences from the observed data, with higher correlation for the first days of forecast. Correlation is also improved when considering the cumulative precipitation along the forecast range. The proposed tool presented itself as a useful way to model the problem and could be easily applied to other cases.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.