CONICET | Buscador de Institutos y Recursos Humanos

In this sensitivity study, we used the Rossby Centre regional atmosphere model RCA3 [1] modified by including the surface database Ecoclimap and by adjusting the atmospheric physics to improve the performance of the model for tropical and subtropical climates (RCA3-E). The continental model domain had a horizontal resolution of 0.5º and 24 levels in the vertical. In the tropical and subtropical South America, a warm-season precipitation maximum associated with the South American Monsoon System (SAMS) dominates the seasonal precipitation cycle. The ongoing deforestation in South America can decrease soil moisture in the region which, in turn, can potentially modify the monsoon rainfall. Previous studies of the effect of modified soil moisture on SAMS have led to quite opposite results [2, 3]. On the one hand, a dryer soil can lead to higher air column temperatures because evapotranspiration (latent heat flux; evaporation from surfaces and transpiration by plants) decreases and, therefore, a larger portion of outgoing energy will be in the form of warm air rising (sensible heat flux). This increases the thermal gradient between the continent and the ocean which can produce stronger inflow of the Atlantic trade winds over the continent, bringing moisture to the monsoon region and producing an early onset of the monsoon. On the other hand, some studies [e.g. 2] have shown that destabilisation of the atmosphere through latent heat flux influences the large-scale circulation triggering the inflow of trade winds. A dry disturbance resulting in weaker latent heat fluxes may, therefore, lead to a later onset of the monsoon. We explored the influence of anomalous soil moisture in late austral winter on the development of SAMS through two ensembles of simulations. The first ensemble was initialised with extremely dry and the other one with extremely wet soil moisture conditions over the whole continent. Members of each ensemble differed only in initialisation dates. Our study covered the monsoon of only one summer, from August 1992 to March 1993. However, the surface and dynamical processes of SAMS act independently of the large-scale conditions [3, 4] and could therefore be similar in other summers. Our results showed that soil moisture anomalies induce both large-scale and local precipitation responses. The difference between the wet and dry ensemble in the partitioning of surface fluxes (the relationship of latent heat flux to sensible heat flux) induced a large difference between the ensembles in air column temperature over the central part of Amazonia. In the dry ensemble, the continental air temperature was higher and brought in stronger Atlantic trade winds over the northern part of the continent that were blocked and turned anti-clockwise to the south by the Andes mountains. Moisture convergence for dry initial conditions was therefore larger than for wet conditions east of the northern Andes and in southern Amazonia, producing more rainfall over these regions during spring (Sep-Oct). In summer (Dec-Jan), precipitation was stronger in the wet ensemble than in the dry one in central Amazonia. Because no difference was observed in moisture convergence (a large-scale phenomenon) in this region among the two ensembles, an explanation could be local precipitation recycling: the region remained wetter since springtime precipitation was similar in both wet and dry ensembles although the soil moisture conditions differed. Our results are only preliminary. A more thorough analysis on different time scales is in progress. n the atmosphere through latent heat flux influences the large-scale circulation triggering the inflow of trade winds. A dry disturbance resulting in weaker latent heat fluxes may, therefore, lead to a later onset of the monsoon. We explored the influence of anomalous soil moisture in late austral winter on the development of SAMS through two ensembles of simulations. The first ensemble was initialised with extremely dry and the other one with extremely wet soil moisture conditions over the whole continent. Members of each ensemble differed only in initialisation dates. Our study covered the monsoon of only one summer, from August 1992 to March 1993. However, the surface and dynamical processes of SAMS act independently of the large-scale conditions [3, 4] and could therefore be similar in other summers. Our results showed that soil moisture anomalies induce both large-scale and local precipitation responses. The difference between the wet and dry ensemble in the partitioning of surface fluxes (the relationship of latent heat flux to sensible heat flux) induced a large difference between the ensembles in air column temperature over the central part of Amazonia. In the dry ensemble, the continental air temperature was higher and brought in stronger Atlantic trade winds over the northern part of the continent that were blocked and turned anti-clockwise to the south by the Andes mountains. Moisture convergence for dry initial conditions was therefore larger than for wet conditions east of the northern Andes and in southern Amazonia, producing more rainfall over these regions during spring (Sep-Oct). In summer (Dec-Jan), precipitation was stronger in the wet ensemble than in the dry one in central Amazonia. Because no difference was observed in moisture convergence (a large-scale phenomenon) in this region among the two ensembles, an explanation could be local precipitation recycling: the region remained wetter since springtime precipitation was similar in both wet and dry ensembles although the soil moisture conditions differed. Our results are only preliminary. A more thorough analysis on different time scales is in progress. n e.g. 2] have shown that destabilisation of the atmosphere through latent heat flux influences the large-scale circulation triggering the inflow of trade winds. A dry disturbance resulting in weaker latent heat fluxes may, therefore, lead to a later onset of the monsoon. We explored the influence of anomalous soil moisture in late austral winter on the development of SAMS through two ensembles of simulations. The first ensemble was initialised with extremely dry and the other one with extremely wet soil moisture conditions over the whole continent. Members of each ensemble differed only in initialisation dates. Our study covered the monsoon of only one summer, from August 1992 to March 1993. However, the surface and dynamical processes of SAMS act independently of the large-scale conditions [3, 4] and could therefore be similar in other summers. Our results showed that soil moisture anomalies induce both large-scale and local precipitation responses. The difference between the wet and dry ensemble in the partitioning of surface fluxes (the relationship of latent heat flux to sensible heat flux) induced a large difference between the ensembles in air column temperature over the central part of Amazonia. In the dry ensemble, the continental air temperature was higher and brought in stronger Atlantic trade winds over the northern part of the continent that were blocked and turned anti-clockwise to the south by the Andes mountains. Moisture convergence for dry initial conditions was therefore larger than for wet conditions east of the northern Andes and in southern Amazonia, producing more rainfall over these regions during spring (Sep-Oct). In summer (Dec-Jan), precipitation was stronger in the wet ensemble than in the dry one in central Amazonia. Because no difference was observed in moisture convergence (a large-scale phenomenon) in this region among the two ensembles, an explanation could be local precipitation recycling: the region remained wetter since springtime precipitation was similar in both wet and dry ensembles although the soil moisture conditions differed. Our results are only preliminary. A more thorough analysis on different time scales is in progress. nn