Projection the long-term ungauged rainfall using integrated Statistical Downscaling Model and Geographic Information...

Hydrological modeling accuracy is compromised in ungauged areas due to limited data availability. This limitation affects predictions of hydrologic extremes. Prior research has shown that ungauged regions receive less attention in climate studies. Established methods rely on gauged data for validation. However, ungauged areas lack such data sources. This gap motivated the development of alternative modeling approaches. No prior work had resolved the issue of projecting rainfall in ungauged regions with high accuracy. This paper introduces a novel approach to address this limitation.

The study aimed to evaluate the accuracy of long-term rainfall projections in ungauged areas using an integrated model. The specific problem is the lack of reliable data in ungauged regions. The motivation comes from the need to improve climate predictions in these areas. The study uses a combination of statistical and geographic tools. The goal is to assess the reliability of projected rainfall trends. The approach involves comparing gauged and ungauged stations. The study also tests the effectiveness of GIS interpolation methods. The findings aim to provide a validated model for ungauged rainfall projection.

The study employed the SDSM-GIS model to project rainfall changes in ungauged areas. Five climate predictors were selected from NCEP and CanESM2-RCP4.5 datasets. The SDSM was calibrated to produce reliable results with low %MAE. Statistical analyses were used to validate the model's performance. The GIS-Kriging method was applied for spatial interpolation. Monthly rainfall trends were compared between gauged and ungauged stations. The accuracy of the model was assessed using %MAE and R values. The results were compared against historical rainfall patterns.

The SDSM-GIS model produced rainfall projections with a %MAE of less than 23%. The R values indicated strong correlation between observed and projected data. The projected rainfall for Δ2030s showed a 14% decrease. All RCP scenarios indicated consistent long-term rainfall patterns. RCP8.5 showed the least change in rainfall intensity. The GIS-Kriging method successfully replicated trends at control stations. The accuracy of the model reached 84% when compared to control data. The results suggest that the model can reliably project rainfall in ungauged areas.

The study demonstrated that the SDSM-GIS model can generate reliable rainfall projections in ungauged areas. The model achieved a high accuracy rate of 84% when validated against control stations. The results suggest that the model is suitable for long-term climate predictions. The findings align with historical rainfall patterns and RCP scenarios. The model's performance was consistent across different RCPs. The GIS-Kriging method proved effective in spatial interpolation. The study confirms the model's ability to handle ungauged data. The results support the use of integrated statistical and geographic methods for climate modeling.