Evapotranspiration partitioning of tropical deciduous forests of North America: A model-based comparison of two eddy-covariance sites with contrasting hydroclimatic regimes

Globally, Tropical Deciduous Forests (TDF) are among the least hydrologically studied ecosystems despite their importance for sustaining vast biodiversity and carbon sequestration potential. These ecosystems are highly controlled by water availability almost fully delivered by precipitation (P) with >80% of these water inputs returned to the atmosphere as evapotranspiration (ET) and with a dominant proportion from plant transpiration (T). However, assessing the magnitude and temporal variation of soil evaporation (E) and plant transpiration (T) in TDF during succession can help to move forward a more comprehensive hydrological theory where biological and abiotic mechanisms that control the partition of ET in disturbed ecosystems are better understood. In this work, a two-year period of Eddy-Covariance (EC) observations were integrated with two different model approaches to assess the seasonal variations of ET partition in two TDF sites under contrasting hydroclimatic regimes. One model approach is based on the Maximum Entropy Production (MEP) theory while the other one is based on the definition of the ecosystem underlying water use efficiency (uWUE). One of the EC monitoring sites is an early succession TDF located in Northwestern Mexico where water-limited ecosystems are strongly controlled by the seasonal precipitation inputs of the North American Monsoon. The second site is a secondary TDF located in the Yucatan Peninsula where water availability is strongly controlled by both seasonal rainfall inputs and groundwater access and TDF can switch from water to energy limited conditions during the rainy season depending on the atmospheric modulation of incoming energy. Our results suggest that both, the hydroclimatic conditions within the same biome and the seasonal variation of rainfall intensity affect the temporal and interannual variation of ET components and ecosystem hydrological dynamics. Knowledge generated by this work would offer information to scientists, stakeholders and decision makers for the design of policies that improve the management of water resources in these types of ecosystems.