Effect of Calibration and Validation Decisions on Streamflow Modeling for a Heterogeneous and Low Runoff–Producing River Basin in India

Adhikary, Partha Pratim; Sena, D.R.; Dash, Ch. Jyotiprava; Mandal, Uday; Nanda, Shruti; Madhu, M.; Sahoo, Dillip Kumar; Mishra, P. K.

doi:10.1061/(asce)he.1943-5584.0001792

Cited by 23 publications

(7 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…O SWAT é um software de domínio público, desenvolvido com o objetivo de predizer os impactos das mudanças no uso e manejo do solo atual e futuro, através da simulação de cenários, na produção de água, sedimentos, nutrientes e agroquímicos em bacias hidrográficas (Arnold et al, 1998;Gassman et al, 2007). No entanto, para que os resultados disponibilizados pelo modelo possam ser utilizados com confiabilidade, torna-se necessário calibrá-lo, ou seja, ajustar os dados simulados aos dados observados (Abbaspour et al, 2017;Paz et al, 2018;Premand et al, 2018;Ashikary et al, 2019).…”

Section: Introductionunclassified

Calibração hidrológica do modelo SWAT em bacia hidrográfica caracterizada pela expansão do cultivo da cana-de-açúcar

Martins

Moraes

et al. 2020

Rev. Bras. Geog. Fis.

View full text Add to dashboard Cite

A dificuldade na gestão de recursos hídricos aliada à dinâmica do uso e ocupação do solo em bacias hidrográficas agrícolas são fatores relevantes para a conservação da água e solo. A gestão de bacias hidrográficas, bem como o monitoramento de cenários de expansão agrícola e mudança no uso do solo, podem se beneficiar de ferramentas de modelagem hidrossedimentológica, como o SWAT (Soil and Water Assessment Tool). Entretanto, para que os resultados obtidos sejam confiáveis, os modelos precisam ser calibrados. Objetivou-se, neste trabalho, calibrar e validar o modelo SWAT, para a variável vazão, tendo como base a bacia hidrográfica do Ribeirão do Pinhal, Limeira -São Paulo, que se caracteriza pela expansão da cana-de-açúcar sobre áreas citrícolas. Dados de vazão de um posto fluviométrico localizado no exutório da bacia foram utilizados para a calibração e validação, a partir de séries temporais diferentes. Utilizou-se o software QSWAT para a simulação hidrológica e o SWAT-CUP para a calibração e validação do modelo. O modelo foi calibrado e validado resultando nos seguintes índices estatísticos NSE=0,64; PBIAS=15,2 e RSR=0,60 para calibração e NSE=0,68 PBIAS=-2,8 e RSR=0,56 para a validação. O ajuste de parâmetros do SWAT (USLE_P, USLE_C, CN2) e do calendário de operações da cana-de-açúcar em acordo com a situação real da bacia foi necessário para a calibração do modelo. Os resultados indicam que o modelo SWAT subestima as vazões extremas, no entanto, dentro de faixa aceitável. O SWAT, após a calibração, pode ser utilizado na gestão de recursos hídricos na bacia do Ribeirão do Pinhal.Hydrological calibration of the SWAT model in a watershed characterized by the expansion of sugarcane cultivationA B S T R A C TThe difficulty in water resources management combined with the dynamics of land use and occupation in agricultural watersheds are relevant factors for water and soil conservation. River basin management, as well as monitoring scenarios of agricultural expansion and land-use change, can benefit from hydrossedimentological modeling tools such as the SWAT (Soil and Water Assessment Tool). However, for the results to be reliable, the models must be calibrated. The objective of this study was to calibrate and validate the SWAT model for the flow variable, based on the Ribeirão do Pinhal watershed, Limeira-São Paulo, which is characterized by the expansion of sugarcane over citrus areas. Flow data from a fluviometric station located in the basin's outfall were used for calibration and validation from different time series. QSWAT software was used for hydrological simulation and SWAT-CUP for model calibration and validation. The model was calibrated and validated resulting in the following statistical indices NSE = 0.64; PBIAS = 15.2 and RSR = 0.60 for calibration and NSE = 0.68 PBIAS = -2.8 and RSR = 0.56 for validation. Adjustment of SWAT parameters (USLE_P, USLE_C, and CN2) and the sugarcane operation schedule according to the actual basin situation was necessary for model calibration. The results indicate that the SWAT model underestimates the extreme flow rates, however, within an acceptable range. After calibration, the SWAT can be used to manage water resources in the Ribeirão do Pinhal basin.Keywords: Hydrologic simulation; land use; flow rate.

show abstract

Section: Introductionunclassified

Calibração hidrológica do modelo SWAT em bacia hidrográfica caracterizada pela expansão do cultivo da cana-de-açúcar

Martins

Moraes

et al. 2020

Rev. Bras. Geog. Fis.

View full text Add to dashboard Cite

show abstract

“…The LSTM‐All model achieved comparable results to the monthly calibrated SWAT model for five out of six watersheds (which are the same as the present study, with the NSE difference of 0.07–0.12) and outperformed the monthly calibrated VIC model (Figure 8b,c and Table S5). The daily LSTM model performed poorly for the watershed (Alladupalli and Singhavaram, Pennar catchment) in a semi‐arid climate zone compared to the monthly calibrated SWAT and VIC model (Adhikary et al, 2019). This suggests that in semi‐arid climate zones where precipitation has higher variability and hydrologic processes are complex, the LSTM model may not be suitable, and a process‐based model is a viable option.…”

Section: Resultsmentioning

confidence: 97%

“…S5). The daily LSTM model performed poorly for the watershed (Alladupalli and Singhavaram, Pennar catchment) in a semi-arid climate zone compared to the monthly calibrated SWAT and VIC model (Adhikary et al, 2019). This suggests that in semi-arid climate zones where precipitation has higher variability and hydrologic processes are complex, the LSTM model may not be suit- and À41% to À15%, respectively.…”

Section: Effect Of Data Length On Model Performancementioning

confidence: 97%

How to enhance hydrological predictions in hydrologically distinct watersheds of the Indian subcontinent?

Mangukiya

Sharma

2023

Hydrological Processes

View full text Add to dashboard Cite

Accurate hydrological predictions are required to prepare for the impacts of climate change, especially in India, which experiences frequent floods and droughts. However, the complex hydrological processes of its distinct watersheds and limited data make it challenging to deliver highly‐performant hydrologic predictions using conventional models. Moreover, it remains uncertain where the limits of predictability are and whether recently‐popular deep learning approaches can offer significant improvements. Here, we tested the first instance of the hydrologic model based on long short‐term memory (LSTM) for 55 Indian watersheds, using a new dataset comprising forcing, attributes, and discharge data. Our results show that the LSTM model provides much‐improved performance compared to conventional models in India, providing a median Nash‐Sutcliffe efficiency (NSE) of 0.56. The LSTM model trained on all the watersheds is more favourable to those trained on individual or homogeneous watersheds, as it benefits from a broader range of hydrological processes and patterns in the input data. However, the LSTM model performs poorly for non‐perennial, large, and semi‐arid climate zone watersheds due to its inability to simulate the complex hydrological processes specific to these environments. Integrating lagged observations with the LSTM model (referred to as DI‐LSTM) improved the predictions in such watersheds and enhanced the median NSE to 0.76 by capturing the temporal dependencies and historical patterns that influence hydrological processes. Overall, the contrast of model performance across watersheds suggests major limitations could be associated with the quality of forcing data, and the slow flow or groundwater processes are highly important in the Indian subcontinent. Notably, both LSTM and DI‐LSTM models performed reasonably well for predictions in ungauged watersheds. The findings of this study demonstrate that data‐sparse countries, too, can benefit from big‐data deep learning and point out further avenues toward model improvements.

show abstract

“…With the use of distinct sets of sensitive parameter values, the multisite calibration methodology under a physio-climatically heterogeneous environment shows discrepancies in the hydrological behavior of the basin's upstream and downstream sections [16]. Therefore, the model's performance at different outlets may be compromised, thereby balancing the targeted threshold of the objective function during the multisite calibration approach [12].…”

Section: Introductionmentioning

confidence: 99%

Metrics Assessment and Streamflow Modeling under Changing Climate in a Data-Scarce Heterogeneous Region: A Case Study of the Kabul River Basin

Akhtar

Borgemeister

Tischbein

et al. 2022

Water

View full text Add to dashboard Cite

Due to many uncertainties in hydrological data and modeling, the findings are frequently regarded as unreliable, especially in heterogeneous catchments such as the Kabul River Basin (KRB). Besides, statistical methods to assess the performance of the models have also been called into doubt in several studies. We evaluated the performance of the Soil and Water Assessment Tool (SWAT) model by statistical indicators including the Kling-Gupta efficiency (KGE), Nash–Sutcliffe efficiency (NSE), and the coefficient of determination (R2) at single and multi-outlets in the KRB and assessed the streamflow under changing climate scenarios i.e., Representative Concentration Pathways (RCP) 4.5 and 8.5 (2020–2045). Because of the heterogeneous nature of the KRB, NSE and R2 performed poorly at multi-outlets. However, the KGE, as the basic objective function, fared much better at single-outlet. We conclude that KGE is the most crucial metric for streamflow evaluation in heterogeneous basins. Similarly, the mean and maximum annual streamflow is projected to decrease by 15.2–15.6% and 17.2–41.8% under the RCP 4.5 and 8.5, respectively.

show abstract

Effect of Calibration and Validation Decisions on Streamflow Modeling for a Heterogeneous and Low Runoff–Producing River Basin in India

Cited by 23 publications

References 44 publications

Calibração hidrológica do modelo SWAT em bacia hidrográfica caracterizada pela expansão do cultivo da cana-de-açúcar

Calibração hidrológica do modelo SWAT em bacia hidrográfica caracterizada pela expansão do cultivo da cana-de-açúcar

How to enhance hydrological predictions in hydrologically distinct watersheds of the Indian subcontinent?

Metrics Assessment and Streamflow Modeling under Changing Climate in a Data-Scarce Heterogeneous Region: A Case Study of the Kabul River Basin

Contact Info

Product

Resources

About