Monitoramento e modelagem da interceptação da precipitação e da umidade do solo em fragmento de Mata Atlântica

Abstract

Due to the influence of forests on the hydrologic cycle, the awereness of ecohydrologic processes in remaining areas of threatened ecosystems is extremely relevant. Although forests have great importance in the environmental context, studies about hydrologic relationships that encompass edaphic, topographic, climate, pedological and vegetative factors are practically nonexistent, especially in tropical ecosystems. Under native forests, investigations are incipient due to the complexity of field observation and to time-space analysis, for they require an adequate sample plan. So in this sense, we aim to study the precipitation partition in detail and monitor soil moisture (θ) in order to analyze the space-time variability and time stability in an ecosystem located in the southern part of Minas Gerais, typical of the Brazilian Atlantic Forest. That ecosystem is a remaining area of the Tropical Mountain Semideciduous Forest. For examining this, we investigated external (E), internal (IP), effective (EP), and exceedingly internal precipitations (EIP), throughfall and stemflow, the canopy’s and the trunks’ storage capacity, aiming at modeling I’s diary totals bewteen Sept./2012 and Mar./2015. For this, we selected 32 trees where sets of monitoring equipment were installed in order for us to examine the aforementioned ecohydrologic variables. The temporal stability of θ was evaluated in the depths (z) (0.10, 0.20, 0.30, 0.40 e 1.0 m), which allowed us to investigate whether the stability indicators depend on z, EP, saturated soil’s hydraulic conductivity (k0), leaf area index (LAI) and landscape representation indexes, aiming at indentifying representative aspects of θ’s behavior. In this case, we interpreted θ on a monthly basis in the x and y directions. Also, we took time reference into consideration, in the 32 spots. We noticed that IP and TR showed greater variability in rainy seasons and that the number of spots for this monitoring detected seasonal variations. The TSC did not display any correlation with the canopy’s area, height, DAP and LAI, whereas the EI could be associated with the number of events, EP’s intensity and volume, wind speed and direction, and the LAI. Liu and Gash’s models demonstrated good results in the simulation of I for this ecosystem. Gash underestimated I in 21%, with evaporation generated by Gash’s (1979) equation, besides being stable in long periods of drought. We noticed that Liu overestimated I in 11.5%, with evaporation generated by P-M’s equation, being more stable in the drought periods. Based on those results, we recommend the Gash model for the modeling of I in environments similar to TMSF, which is one of the main forest formations in Southeast Brazil. We perceived a decrease in the variability of the stability’s statistics in the greater depths. There was greater temporal stability of θ in the terrain’s lowest areas, less stability in medium altitudes, and no temporal stability in the highest regions. We concluded that the topographic characteristics affect stability and identification of stable spots. The spots chosen for the monitoring in the depths deferred according to their location. We noticed that the representation indexes can be incorporated in the temporal stability’s analyses and be associated with the soil’s and forest’s specificities, which helps in the selection of representative spots.