Ecohydrology of the Caatinga Tropical Dry Forest
Abstract
The overarching goal of this study is to improve our understanding of ecohydrological process in tropical dry forests. Tropical dry forests provide ideal natural settings for ecohydrology research because of their distinct rainfall seasonality and high diversity of tree species and morphologies. This study takes place in the largest contiguous tropical dry forests of the Americas, the Caatinga biome of northeast Brazil. The objective of Chapter II was to determine if woody vegetation alters the temporal structure of soil water. I used a continuous wavelet transform examine soil water variability across different soil depths and time scales. Results indicate that soil water in the forested Caatinga more variable (high frequency changes) at short times and less variable at long time scales (low frequency changes) compared to the pasture site, due to differences in vegetation cover and soil properties. The objective of Chapter III was to use plant functional types to better understand plant water-use and conservation strategies. Field data included pre-dawn and mid-day leaf water potential, leaf thickness, specific leaf area, leaf δ¹³C, and stable water isotopes, δD and δ¹⁸O, of stem water and soil water. Results show that the plant functional type defined for the Caatinga dry forest are appropriate for understanding general differences in plant response to water stress. The objective of Chapter IV was to better understand the rate of water use for tree species of high vs. low wood density as a response to water availability, i.e. soil moisture, and demand, i.e. vapor pressure deficit. Results indicate that high vs. low wood density species are differentially sensitive to water availability and demand. The results of this dissertation have important implications for understanding how dry forests might respond to anthropogenic influences, such as land use change and climate change.
Citation
Wright, Cynthia Louise (2020). Ecohydrology of the Caatinga Tropical Dry Forest. Doctoral dissertation, Texas A&M University. Available electronically from https : / /hdl .handle .net /1969 .1 /191921.