An assessment of rainfall measurement based on TRMM products

Contrasts between TRMM 3B43 monthly data and rainfall observations of 720 stations in China are conducted based on a linear regression model. During January 1999 and December 2007, there is a significant correlation between TRMM data and the observed ones with an average r(2) 0.834. TRMM data performs better in the South and North, especially for flat regions. Limited by radar signal degradation due to heavy rain and low resolution of monitoring, TRMM data have better results in low-flow season than that in flood season. TRMM data cover all the places in middle and low latitudes. It is useful for long-term water resources planning, drought analysis in ungauged basins (PUB), and will be helpful for flood warning. Spatiotemporal data with higher resolution will greatly promote the development of hydrology in the future.