Atmosphere and hydrological controls of the evapotranspiration over a floodplain forest in the Bananal Island region, Amazonia

Abstract

[1] This article discusses seasonal and interannual variations of the evapotranspiration (ET) rates in Bananal Island floodplain, Brazil. Measurements included ET and sensible heat flux using the eddy covariance method, atmospheric forcings (net radiation, Rn, vapor pressure deficit, VPD, wind speed and air temperature), soil moisture profiles, groundwater level and flood height, taken from November 2003 to December 2006. For the hydrological years (October-September) of 2003/2004, 2004/2005 and 2005/2006, the accumulated precipitation was 1692, 1471, 1914 mm and the accumulated ET was 1361, 1318 and 1317 mm, respectively. Seasonal analyses indicated that ET decreased in the dry season (average 3.7 mm day À1 ), despite the simultaneous increase in Rn, air temperature and VPD. The increase of ET in the wet season and particularly in the flood period (average 4.1 mm day À1 ) showed that the free water surface evaporation strongly influenced the energy exchange. Soil moisture, which was substantially depleted during the dry season, and adaptative vegetation mechanisms such as leaf senescence contributed to limit the dry season ET. Strong drainage within permeable sandy soils helped to explain the soil moisture depletion. These results suggest that the Bananal flooding area shows a different pattern in relation to the upland Amazon forests, being more similar to the savanna strictu senso areas in central Brazil. For example, seasonal ET variation was not in phase with Rn; the wet season ET was higher than the dry season ET; and the system stored only a tiny memory of the flooding period, being sensitive to extended drought periods.

Mechanisms of water supply and vegetation demand govern the seasonality and magnitude of evapotranspiration in Amazonia and Cerrado

“…2 shows that equatorial forests exhibit a seasonal cycle of E peaking with net radiation during the dry season, transitional southern forests show either a flat seasonal cycle (due to less seasonality in available light) or a slight dry season depression (some degree of water limitation), and Cerrado demonstrates a strong dry season depression (both due to reductions in light and water). These results corroborate those of previous work which showed a general trend of increasing water limitation from north to south (Hasler and Avissar, 2007;Juarez et al, 2007;Borma et al, 2009;da Rocha et al, 2002da Rocha et al, , 2009Fisher et al, 2009). While BAN and RJA differed slightly in their respective seasonalities of E (BAN has a more pronounced dry season depression compared to RJA), overall the individual site E seasonalities corresponded to the mean E seasonality of the grouped sites (see Appendix D of the supplement for individual site seasonalities).…”

“…While both models with these modifications simulate the overall seasonality of E well, the observations indicated slight reductions in E during the dry season in June through September, which were best captured by SiB3. Surprisingly, the water budget analysis for the BAN site (Borma et al, 2009) revealed that observed seasonal patterns of soil moisture storage and groundwater flux were not consistent with either of the deep soil/deep roots or groundwater formulations (Fig. 4f,h and j).…”

“…A large influx of groundwater into the system is inferred during the month of November because soil water increases by nearly double the incoming precipitation, even when the soil moisture measurements are not extrapolated beyond the measurement domain (Table 3). The abrupt influx of groundwater (Q g↑ ) into this system occurs not because of soil type or depth, but because of this site's proximity to a floodplain (Borma et al, 2009), and no further net influx of groundwater after November is recorded because the soil quickly becomes and remains saturated throughout the flooding period. This highlights the importance of modeling groundwater fluctuations as a 2-dimensional topographically driven process, in which orientation in relation to drainage basins makes a big difference (Fan and Miguez-Macho, 2010).…”

Vegetational changes during the last millennium inferred from a palynological record from the Bananal Island, Tocantins, Brazil

“…It is considered the largest fluvial island in the world (Borma et al 2009;Dias et al 2011;Devi 2012). The island is a floodplain temporarily flooded during the rainy season by precipitation of local waters, being classified as a seasonal wetland (Valente 2007).…”

The Role of Physical and Political Factors on the Conservation of Native Vegetation in the Brazilian Forest-Savanna Ecotone

“…The dry season is from May to September (mean relative humidity of 40%), while the wet season is from October to April (mean relative humidity of 90%) (Borma et al, 2009). Plinth and Concretionary soils are the dominant soil types of the region, where can also be found Hydromorphic soils, Ox soils and Quartz sands (SEPLAN, 2008).…”