Evaluating monthly and daily gridded station-based precipitation datasets; challenges and recommendations.

Gridded precipitation datasets are essential products in climatology, hydrology, and climate impact studies. These datasets represent the high spatial and temporal variability of this variable over continuous areas and periods, and yet, due to the complex characteristics of precipitation, it is challenging to obtain accurate estimates. Consequently, the creation of a gridded dataset from observations requires a thorough and accurate application of quality control, reconstruction, and gridded techniques. However, most gridded datasets created and published from the mid-1990s to the present use a broad variety of techniques, methods, and results, which can change the final representativeness of the data. Therefore, it is key to provide general guidelines for the development of future, more robust gridded datasets based on data characteristics, geographic factors, and advanced statistical techniques. Here, we identify gaps and challenges for near-future perspectives and provide guidelines for implementing better approaches based on the performance of 48 products. Finally, we concluded that scientists should adopt tailored methods to improve the representativeness and uncertainty of the estimates.