Combining satellite precipitation and long‐term ground observations for hydrological monitoring in China

Abstract Satellite real‐time precipitation enables hydrological monitoring in China where the near‐real‐time ground observations are not readily available. However, the inconsistency between the real‐time satellite precipitation and gauge‐based retrospective data may introduce large systematic bias in near‐real‐time hydrological monitoring. Here we attempted to integrate the Tropical Rainfall Measuring Mission (TRMM) real‐time precipitation (3B42RTV7) into a 62 year gauge‐based retrospective product, the IGSNRR (Institute of Geographical Sciences and Natural Resources Research) dataset through matching their cumulative probability functions toward a near‐real‐time hydrological monitoring consistent with the long‐term retrospective simulations. A nearly 11 year period from March 2000 to December 2010 was taken as the training period to establish the satellite‐gauge precipitation relationship, which was employed in the period of 2011–2013 to evaluate the performance of the adjustment. The results show that the adjusted 3B42RTV7 matches well with IGSNRR precipitation, while the unadjusted data tend to overestimate precipitation. Forced by the adjusted 3B42RTV7, the Variable Infiltration Capacity model can reproduce the IGSNRR‐derived hydrographs and high/low flows better than the model forced by the unadjusted data. The percentiles of the adjusted hydrological estimates in the 62 year estimates from IGSNRR are used for near‐real‐time assessment of hydrological extremes. The hydrological monitoring assisted by the adjusted satellite precipitation, which enables the employment of the long‐term ground observations, is able to capture more detailed drought information than that before adjustment. Our experiment suggests that the satellite real‐time precipitation, after adjustment, can generate the current hydrological conditions which can be directly compared with the long‐term climatology, and thus facilitates near‐real‐time diagnosis and detection of hydrological extremes.