Developing a framework for an early warning system of seasonal temperature and rainfall tailored to aquaculture in Bangladesh

The occurrence of high temperature and heavy rain events during the monsoon season are a major climate risk affecting aquaculture production in Bangladesh. Despite the advances in the seasonal forecasting, the development of operational tools remains a challenge. This work presents the development o...

Descripción completa

Detalles Bibliográficos
Autores principales: Montes, Carlo, Acharya, Nachiketa, Hossain, Peerzadi Rumana, Amjath-Babu, Tharayil Shereef, Krupnik, Timothy J., Hassan, S.M. Quamrul
Formato: Journal Article
Lenguaje:Inglés
Publicado: Elsevier 2022
Materias:
Acceso en línea:https://hdl.handle.net/10568/126494
_version_ 1855539890236686336
author Montes, Carlo
Acharya, Nachiketa
Hossain, Peerzadi Rumana
Amjath-Babu, Tharayil Shereef
Krupnik, Timothy J.
Hassan, S.M. Quamrul
author_browse Acharya, Nachiketa
Amjath-Babu, Tharayil Shereef
Hassan, S.M. Quamrul
Hossain, Peerzadi Rumana
Krupnik, Timothy J.
Montes, Carlo
author_facet Montes, Carlo
Acharya, Nachiketa
Hossain, Peerzadi Rumana
Amjath-Babu, Tharayil Shereef
Krupnik, Timothy J.
Hassan, S.M. Quamrul
author_sort Montes, Carlo
collection Repository of Agricultural Research Outputs (CGSpace)
description The occurrence of high temperature and heavy rain events during the monsoon season are a major climate risk affecting aquaculture production in Bangladesh. Despite the advances in the seasonal forecasting, the development of operational tools remains a challenge. This work presents the development of a seasonal forecasting approach to predict the number of warm days (NWD) and number of heavy rain days (NHRD) tailored to aquaculture in two locations of Bangladesh (Sylhet and Khulna). The approach is based on the use of meteorological and pond temperature data to generate linear models of the relationship between three-monthly temperature and rainfall statistics and NWD and NHRD, and on the evaluation of the skill of three operational dynamical models from the North American Multi-Model Ensemble (NMME) project. The linear models were used to evaluate the forecasts for two seasons and 1-month lead time: May to July (MJJ), forecast generated in April, and August to October (ASO), forecast generated in July. Differences were observed in the skill of the models predicting maximum temperature and rainfall (Spearman correlation, Root Mean Square Error, Bias statistics, and Willmott’s Index of Agreement,), in addition to NWD and NHRD from linear models, which also vary for the target seasons and location. In general, the models show higher predictive skill for NWD than NHRD, and for Sylhet than in Khulna. Among the three evaluated NMME models, CanSIPSv2 and GFDL-SPEAR exhibit the best performance, they show similar features in terms of error metrics, but CanSIPSv2 presents a lower interannual standard deviation.
format Journal Article
id CGSpace126494
institution CGIAR Consortium
language Inglés
publishDate 2022
publishDateRange 2022
publishDateSort 2022
publisher Elsevier
publisherStr Elsevier
record_format dspace
spelling CGSpace1264942025-11-06T13:09:30Z Developing a framework for an early warning system of seasonal temperature and rainfall tailored to aquaculture in Bangladesh Montes, Carlo Acharya, Nachiketa Hossain, Peerzadi Rumana Amjath-Babu, Tharayil Shereef Krupnik, Timothy J. Hassan, S.M. Quamrul fishery production risk management aquaculture early warning systems climate change The occurrence of high temperature and heavy rain events during the monsoon season are a major climate risk affecting aquaculture production in Bangladesh. Despite the advances in the seasonal forecasting, the development of operational tools remains a challenge. This work presents the development of a seasonal forecasting approach to predict the number of warm days (NWD) and number of heavy rain days (NHRD) tailored to aquaculture in two locations of Bangladesh (Sylhet and Khulna). The approach is based on the use of meteorological and pond temperature data to generate linear models of the relationship between three-monthly temperature and rainfall statistics and NWD and NHRD, and on the evaluation of the skill of three operational dynamical models from the North American Multi-Model Ensemble (NMME) project. The linear models were used to evaluate the forecasts for two seasons and 1-month lead time: May to July (MJJ), forecast generated in April, and August to October (ASO), forecast generated in July. Differences were observed in the skill of the models predicting maximum temperature and rainfall (Spearman correlation, Root Mean Square Error, Bias statistics, and Willmott’s Index of Agreement,), in addition to NWD and NHRD from linear models, which also vary for the target seasons and location. In general, the models show higher predictive skill for NWD than NHRD, and for Sylhet than in Khulna. Among the three evaluated NMME models, CanSIPSv2 and GFDL-SPEAR exhibit the best performance, they show similar features in terms of error metrics, but CanSIPSv2 presents a lower interannual standard deviation. 2022-04 2023-01-03T12:13:17Z 2023-01-03T12:13:17Z Journal Article https://hdl.handle.net/10568/126494 en Open Access application/pdf Elsevier Montes, C., Acharya, N., Rumana Hossain, P., Amjath Babu, T. S., Krupnik, T. J., & Quamrul Hassan, S. M. (2022). Developing a framework for an early warning system of seasonal temperature and rainfall tailored to aquaculture in Bangladesh. Climate Services, 26, 100292. https://doi.org/10.1016/j.cliser.2022.100292
spellingShingle fishery production
risk management
aquaculture
early warning systems
climate change
Montes, Carlo
Acharya, Nachiketa
Hossain, Peerzadi Rumana
Amjath-Babu, Tharayil Shereef
Krupnik, Timothy J.
Hassan, S.M. Quamrul
Developing a framework for an early warning system of seasonal temperature and rainfall tailored to aquaculture in Bangladesh
title Developing a framework for an early warning system of seasonal temperature and rainfall tailored to aquaculture in Bangladesh
title_full Developing a framework for an early warning system of seasonal temperature and rainfall tailored to aquaculture in Bangladesh
title_fullStr Developing a framework for an early warning system of seasonal temperature and rainfall tailored to aquaculture in Bangladesh
title_full_unstemmed Developing a framework for an early warning system of seasonal temperature and rainfall tailored to aquaculture in Bangladesh
title_short Developing a framework for an early warning system of seasonal temperature and rainfall tailored to aquaculture in Bangladesh
title_sort developing a framework for an early warning system of seasonal temperature and rainfall tailored to aquaculture in bangladesh
topic fishery production
risk management
aquaculture
early warning systems
climate change
url https://hdl.handle.net/10568/126494
work_keys_str_mv AT montescarlo developingaframeworkforanearlywarningsystemofseasonaltemperatureandrainfalltailoredtoaquacultureinbangladesh
AT acharyanachiketa developingaframeworkforanearlywarningsystemofseasonaltemperatureandrainfalltailoredtoaquacultureinbangladesh
AT hossainpeerzadirumana developingaframeworkforanearlywarningsystemofseasonaltemperatureandrainfalltailoredtoaquacultureinbangladesh
AT amjathbabutharayilshereef developingaframeworkforanearlywarningsystemofseasonaltemperatureandrainfalltailoredtoaquacultureinbangladesh
AT krupniktimothyj developingaframeworkforanearlywarningsystemofseasonaltemperatureandrainfalltailoredtoaquacultureinbangladesh
AT hassansmquamrul developingaframeworkforanearlywarningsystemofseasonaltemperatureandrainfalltailoredtoaquacultureinbangladesh