AUTHORS:

Carneiro R.G., Borges C.K., Santos C.A., de Oliveira G., Stachlewska I.S., Fisch G., dos Santos C.A.C.

ABSTRACT:

The Amazon region stands out as Earth's largest and most prominent terrestrial convective center, significantly influencing precipitation patterns through the substantial net radiation at the Earth's surface (Rn). This radiation is further divided into sensible heat (H) and latent heat (LE) fluxes, underscoring the critical need to investigate the intricate exchanges of energy between ecosystems and the atmosphere while also scrutinizing the closure of the energy balance (EBC). Such studies are of paramount importance for gaining a comprehensive understanding of natural process dynamics. This research was undertaken to refine our comprehension of the temporal variability of energy fluxes, energy balance closure, and the phenomenon of evapotranspiration (ET) hysteresis in relation to key atmospheric variables, namely, Rn, air temperature, and vapor pressure deficit (VPD), within the central Amazon region. This investigation spanned a typical year (2014) and accounted for the influence of an El Niño Southern Oscillation (ENSO) event (2015). The findings illuminated that LE constituted the predominant component of the energy balance in both years, comprising approximately 70% of the total energy flux, except during the dry season of 2015 when H took precedence, accounting for 55% of the total energy exchange due to the ENSO event. The EBC exhibited an underestimation of turbulent fluxes (H + LE) of around 24% relative to the available energy (Rn - G). A more robust closure was observed during the dry period of 2015, reaching 82%, suggesting heightened uncertainty in LE estimates. In the context of ET, clockwise hysteresis loops were observed concerning air temperature and VPD. The relationship between ET and Rn displayed a linear response to solar radiation, with the intensity of ET hysteresis becoming more pronounced when correlated with air temperature and VPD.

Journal of South American Earth Sciences, 2023, vol. 132, art. 104686, doi: 10.1016/j.jsames.2023.104686