Potencial energético da chuva e da precipitação interna em floresta tropical seca

Abstract

Water scarcity in dry tropical regions is a fact that can be aggravated by climate change. Therefore, the characterization of rainfall events is necessary for a better evaluation of hydrological processes. To fill this gap, the present study was based on the following objectives: (i) to determine the minimum time between events - rain MIET; (ii) identify how the vegetation of the dry rain forest - FTS defines the internal precipitation - TF at different stages of canopy development; (iii) investigate the attenuating power of FTS in the erosivity power of rain; (iv) evaluate the prediction generated by the adjustments of two models in the literature to estimate kinetic energy (EC) and erosivity potential (EI30) of total precipitation (P) and TF; (v) Propose a new P and TF EC model based on data measured in field conditions. Twelve consecutive years of P data in an interval of 5 minutes were collected and grouped into different (MIETs) - 15 min, 1 h, 2 h, 3 h, 6 h, 12 h and 24 h in semiarid region of northeastern Brazil. To evaluate the energy distribution of raindrops at different stages of canopy development, we installed two disdrometers, one to register - P and the other under the FTS canopy to register - TF. Data were collected from December/2019 to May/2021, totaling 95 natural events of P and TF. The results confirm that the inclusion or exclusion of single-end events (EPU) in the choice of the most appropriate MIET affected the number of rainfall events and their characteristics; the number of single-end events decreased to a 3-hour MIET, but showed no difference above a 6h MIET. The 6-hour MIET is the most appropriate to characterize rainfall distribution in this tropical semiarid region. At the beginning of the rainy season when leaf density is low, a larger fraction of the rain was converted into TF, which decreases as leaf density increases. For events greater than 3 mm, the number of ST drops was always higher than p and with smaller diameters, regardless of the stage of canopy development. This result indicates fragmentation of raindrops by the canopy of vegetation. The relationship between EC and mean intensity of P and Measured ST is better expressed as a function of time (J m-2 h-1), through a linear model (R2 > 0.98 - P = 0.000). Among the models used, it was observed that the Wischmeier and Smith (WS) model was the one that most underestimated the EC and EI30 values of P and TF. The proposed model of EC by temporal variation of intensity showed the best fit to the measured data of EC and EI30 of P and TF compared to the WS model. Faced with this non-adequacy of the EC models to the P and TF data, we propose a new model, calibrated with 50% of the measured data of height and average intensity of P and then validated with the other 50%. It was found that the new model underestimates the EC by 5% (R2 = 0.98) for P and overestimates it by 8% (R2 = 0.97) for TF, proving to be an alternative in studies of erosion processes in semiarid regions with similar characteristics. In addition, the FTS canopy reduces the measured EC by 25% and the EI30 by 39%, proving that the canopy of tree vegetation protects the soil surface from the impact of raindrops, as well as being an effective natural ground cover to protect against soil erosion.