{"title":"Future changes in surface solar radiation over India: A bias-corrected and downscaled assessment of CMIP6 projections for renewable energy planning","authors":"Ashwin Vijay Jadhav, Rohini Lakshman Bhawar","doi":"10.1016/j.egycc.2025.100213","DOIUrl":null,"url":null,"abstract":"<div><div>Understanding future changes in surface solar radiation (SSR) is crucial for long-term renewable energy planning, especially in solar-rich countries such as India. However, high-resolution and bias-corrected SSR datasets for India remain scarce, limiting accurate assessment of future solar energy potential. In this study, we develop daily bias-corrected and statistically downscaled (BCD) dataset of SSR at a spatial resolution of 0.25°×0.25° for the Indian subcontinent. Simulations from nine Coupled Model Intercomparison Project Phase-6 (CMIP6) climate models are used for the historical period (1951–2014) and future period (2021–2100) under three Shared Socioeconomic Pathways (SSP1–2.6, SSP2–4.5, SSP5–8.5). The BCD was implemented using the Empirical Quantile Mapping (EQM) method at each fine-resolution grid point, improving local-scale variability representation. ERA5’s surface solar radiation downward (SSRD) is used as the reference dataset to downscale, bias correct, and validate the models. Evaluation of BCD dataset against observations showed substantial performance improvement, with correlation coefficients increasing from 0.65 to 0.81 (raw) to 0.89–0.94 (corrected) and reducing root mean square error across all models. These improvements in dataset reliability lay a strong foundation for analysing future SSR dynamics under different climate scenarios.</div><div>Future projections indicate a progressive decline in SSR with increasing emission intensity: modest changes under SSP1–2.6, with slight positive anomalies over parts of northern India, and pronounced reductions of up to –10 % under SSP2–4.5 and SSP5–8.5, particularly over central, southern, and eastern India. Decadal trends reveal average declines of –0.11 %, –0.52 %, and –0.60 % per decade under SSP1–2.6, SSP2–4.5, and SSP5–8.5, respectively, with a temporary slowdown in mid-century (2041–2060) under SSP2–4.5 followed by renewed intensification. The resulting high-resolution BCD dataset not only represents a technical advancement in the robust downscaling of CMIP6 SSR for India but also provides a practical resource to guide solar park siting, optimize grid integration strategies, and support climate-resilient energy policy development.</div></div>","PeriodicalId":72914,"journal":{"name":"Energy and climate change","volume":"6 ","pages":"Article 100213"},"PeriodicalIF":5.6000,"publicationDate":"2025-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Energy and climate change","FirstCategoryId":"1085","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2666278725000406","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"ENERGY & FUELS","Score":null,"Total":0}
引用次数: 0
Abstract
Understanding future changes in surface solar radiation (SSR) is crucial for long-term renewable energy planning, especially in solar-rich countries such as India. However, high-resolution and bias-corrected SSR datasets for India remain scarce, limiting accurate assessment of future solar energy potential. In this study, we develop daily bias-corrected and statistically downscaled (BCD) dataset of SSR at a spatial resolution of 0.25°×0.25° for the Indian subcontinent. Simulations from nine Coupled Model Intercomparison Project Phase-6 (CMIP6) climate models are used for the historical period (1951–2014) and future period (2021–2100) under three Shared Socioeconomic Pathways (SSP1–2.6, SSP2–4.5, SSP5–8.5). The BCD was implemented using the Empirical Quantile Mapping (EQM) method at each fine-resolution grid point, improving local-scale variability representation. ERA5’s surface solar radiation downward (SSRD) is used as the reference dataset to downscale, bias correct, and validate the models. Evaluation of BCD dataset against observations showed substantial performance improvement, with correlation coefficients increasing from 0.65 to 0.81 (raw) to 0.89–0.94 (corrected) and reducing root mean square error across all models. These improvements in dataset reliability lay a strong foundation for analysing future SSR dynamics under different climate scenarios.
Future projections indicate a progressive decline in SSR with increasing emission intensity: modest changes under SSP1–2.6, with slight positive anomalies over parts of northern India, and pronounced reductions of up to –10 % under SSP2–4.5 and SSP5–8.5, particularly over central, southern, and eastern India. Decadal trends reveal average declines of –0.11 %, –0.52 %, and –0.60 % per decade under SSP1–2.6, SSP2–4.5, and SSP5–8.5, respectively, with a temporary slowdown in mid-century (2041–2060) under SSP2–4.5 followed by renewed intensification. The resulting high-resolution BCD dataset not only represents a technical advancement in the robust downscaling of CMIP6 SSR for India but also provides a practical resource to guide solar park siting, optimize grid integration strategies, and support climate-resilient energy policy development.