A spatiotemporal estimation method for hourly rainfall based on F-SVD in the recommender system

In this study, a spatiotemporal estimation method based on Funk singular value decomposition (F-SVD) that considers the correlation of rainfall is proposed to improve estimations from gauge observations. Hourly data several flood events are selected verify by comparing with Inverse Distance Weighting (IDW) and Ordinary Kriging (OK) in Hanjiang basin, China. The results show (1) F-SVD has best performance estimation, larger amount event, greater improvement as compared OK IDW; (2) through combination/integration F-SVD, accuracy IDW can be greatly improved. Therefore, employed practical estimate spatial distribution, which essential for regional hydrological modelling water resource analysis.