H13H-1665
Characteristics and Error Modeling of GPM Satellite Rainfall Estimates over Different Regions of Brazil
Monday, 14 December 2015
Poster Hall (Moscone South)
Romulo Augusto Juca Oliveira, National Institute for Space Research (INPE), Cachoeira Paulista, Brazil, Daniel A Vila, INPE National Institute for Space Research, Sao Jose dos Campos, Brazil, Viviana Maggioni, George Mason University Fairfax, Fairfax, VA, United States and Carlos A. Morales, Universidade de São Paulo, São Paulo, Brazil
Abstract:
This study aims to investigate, over the different regions of Brazil, the error characteristics and uncertainties (random and systematic errors components) in satellite-based precipitation estimates by comparing the Goddard Profiling Algorithm (GPROF), through different sensors from GPM database (such as GMI, TMI, SSMI/S, AMSR2, MHS, among others), and Integrated Multi-satellitE Retrievals for GPM (IMERG) algorithms. The analyses are made with other ground (S- and X-band dual polarization weather radar) and space (e.g., TRMM-PR and GPM-DPR [at Ku-band] active radars) based rainfall estimates as references at instantaneous timescales and respecting their temporal limitations. The Precipitation Uncertainties for Satellite Hydrology (PUSH) framework is used for the analysis and uncertainties characterization and error modeling. Specially, this study are focused on specific regions of Brazil, where the campaigns of the CHUVA project occurred (CHUVA/GoAmazon [IOP1 and 2] in Amazon and over southern Brazil where the S-band dual polarization radars (e.g., the FCTH radar) are located.
