Insights on modeling the hydrometeorology of the Amazon
Renato
Ramos da
Silva, Duke University, renato@duke.edu
(Presenting)
Roni
Avissar, Duke University, avissar@duke.edu
Observations alone cannot be used to explain the complex mechanisms involved in the hydrometeorology of the Amazon and models are needed to bridge the gap in our understanding. Here, we evaluate the performance of the Regional Atmospheric Modeling System (RAMS) in simulating the hydrometeorology of this region during the rainy season. Model results are compared with observations from several instruments (weather stations, meteorological towers, radiosondes, raingauges, radars and satellites) collected during the WetAMC/TRMM campaign of January and February 1999 in Rondonia. We find that RAMS is capable of simulating correctly the meteorology, including the averaged daily rainfall, of that region. Several sensitivity tests were performed to evaluate the impacts of grid resolution, initial conditions of soil and atmospheric moisture, maximum stomatal conductance, and tree trunk heat storage on the model performance. Overall, the results show that better initial conditions and higher vertical resolution provide significant improvements to the model’s performance. Atmospheric humidity forced at the lateral boundaries appears to be particularly important to correctly simulate the onset and accumulation of daily rainfall, emphasizing the needs for very-high quality large-scale forcing conditions.