Towards the selection of a longer record of quantitative precipitation forecast dataset from TIGGE ensembles for India during the monsoon season

Abstract

AbstractChanging precipitation patterns and increased extreme events make a reliable forecast of summer monsoon precipitation crucial in India. We need a longer record of forecast data to develop a streamflow forecasting system. To select a suitable dataset from the THORPEX Interactive Grand Global Ensemble (TIGGE) archives, we applied two criteria: (i) the length of the dataset should be at least 10 years of continuous record, and (ii) the forecast product should have at least 20 ensemble members for a lead time of minimum 5 days. We evaluate the ensemble quantitative precipitation forecasts (QPFs) obtained from the four selected international agencies over the Indian region throughout the monsoon season (June to September) from 2011 to 2020. We specifically looked at the accuracy of QPFs in 22 river basins in forecasting normal and extreme precipitation events. Data from the Integrated Multi-satellite Retrievals for Global Precipitation Measurement (IMERG) is used as observation data for the same period. We aim to assign the best QPF in each river basin. The proficiency of QPFs is evaluated using six criteria from deterministic, dichotomous, and probabilistic error statistics. The error values are classified into three categories – low, moderate, and high. A Forecast Reliability Index is formulated using the given four QPFs, and three categories for each of the six error statistics to answer (a) which QPF shows better performance in which river basin and (b) whether any conclusion can be made on the overall performance of a QPF for all the River Basins of India.Keywords: Quantitative Precipitation ForecastsTIGGE archiveForecast Reliability IndexExtreme precipitationRiver basinsDisclaimerAs a service to authors and researchers we are providing this version of an accepted manuscript (AM). Copyediting, typesetting, and review of the resulting proofs will be undertaken on this manuscript before final publication of the Version of Record (VoR). During production and pre-press, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal relate to these versions also. AcknowledgementsWe thank to the Editor, Associate Editor, and two anonymous Reviewers for their comments and suggestions on this manuscript. This research was completed thanks to the support of the Science and Engineering Research Board (SERB). Department of Science and Technology, Government of India (project number CRG/2022/004006) awarded to Sanjeev Kumar Jha.