Land-Atmosphere coupling strength in GLDAS-2 and among CLARIS-LPB regional models:a statistical approach

PABLO C. SPENNEMANN; A. CELESTE SAULO

Punta del Este

Encuentro; CLARIS LPB M48 Final Meeting; 2012

Land?atmosphere feedbacks have been identified as one of the key sources of uncertainty in climate models and are thus of central importance for climate change projections (Koster et al. 2004a; Friedlingstein et al. 2006; Seneviratne et al. 2006a; Betts 2007; Pitman et al. 2009). Despite these uncertainties, soil moisture initialization is promising for seasonal forecasting in regions where the influence of soil states on the atmosphere is relevant and soil moisture memory is large (see, e.g., Koster et al. 2004a; Seneviratne et al. 2006b; Koster et al. 2010). Understanding these feedbacks is therefore of crucial relevance for climate research and applications. The objective of this work is to document the coupling strength in GLDAS-2 data set, using a statistical approach and characterize this coupling strength among the CLARIS-LPB models. Due to the noisy character of λp is hard to drive conclusions. Focusing on LPB it can be mentioned that models do not necessary reproduce the ?observed? coupling strength, particularly during non-ENSO years. During ENSO years there is more coherence between GLDAS-Era Interim and most models over an area near the coast, around 30ºS. Interestingly, the amount of precipitation variance that might be attributed to previous month of soil moisture variance is important. The coupling strength over LPB is moderate and its sign changes betwen non-ENSO years and all years.Although at very local scales, there are indications that some portions of PP variability are modulated by SM variability in the previous month.