Savanna vegetation-fire-climate relationships differ among continents.

Ecologists have long sought to understand the factors controlling the structure of savanna vegetation. Using data from 2154 sites in savannas across Africa, Australia, and South America, we found that increasing moisture availability drives increases in fire and tree basal area, whereas fire reduces tree basal area. However, among continents, the magnitude of these effects varied substantially, so that a single model cannot adequately represent savanna woody biomass across these regions. Historical and environmental differences drive the regional variation in the functional relationships between woody vegetation, fire, and climate. These same differences will determine the regional responses of vegetation to future climates, with implications for global carbon stocks.

Land-use changes alter radiative energy and water vapor fluxes of a tall-grass Andropogon field and a savanna-woodland continuum in the Orinoco lowlands.

Changes in land use in the Orinoco lowlands affect the daily trends of energy and water vapor fluxes. We analyzed these fluxes along a disturbance gradient beginning from a cultivated tall-grass Andropogon field (S1) and extending over three savanna sites with increasing woody cover over herbaceous vegetation. The savanna sites encompass a herbaceous savanna (S2), a tree savanna (S3) and a woodland savanna (S4). In the wet season, there were differences in the radiation budget: seasonally averaged albedo for S1 (0.17) exceeded that of S2-S4 (0.13-0.14). Eddy covariance fluxes indicate that the partitioning of the daily net radiation (Rn) into sensible and latent heat (lambda E) fluxes depends on land use. During the wet season, evapotranspiration (i.e., lambda E) over the S1-S4 sites accounted for a variable fraction of Rn (i.e., 0.75, 0.52, 0.67 and 0.68, respectively). Therefore, the Bowen ratio was typically below 1. As the dry season progressed, the lambda E/Rn ratio decreased markedly with increasing air and canopy temperatures and air humidity mole fraction deficit. The maximum evaporation rate over the S1-S4 sites was 3.2, 2.5, 3.5 and 4.1 mm day(-1), respectively, and the annual values were 721, 538, 771 and 732 mm year(-1), respectively, equivalent to 49, 65, 52 and 88% of the rainfall. Soil water content fell from a maximum above 0.28 in the wet season to 0.030, 0.026, 0.030 and 0.028 m(3) m(-3) at sites S1-S4, respectively, in the dry season. Leaf area index was greatly reduced as herbaceous vegetation dried out.

Land-use changes alter CO2 flux patterns of a tall-grass Andropogon field and a savanna-woodland continuum in the Orinoco lowlands.

Land use changes in the savannas of the Orinoco lowlands have resulted in a mosaic of vegetation. To elucidate how these changes have affected carbon exchanges with the atmosphere, we measured CO2 fluxes by eddy covariance and soil CO2 efflux systems along a disturbance gradient beginning with a cultivated tall-grass Andropogon field (S1) and extending over three savanna sites with increasing woody cover growing above native herbaceous vegetation. The savanna sites included a herbaceous savanna (S2), a tree savanna (S3) and a woodland savanna (S4). During the wet season, maximum diurnal net ecosystem exchange (NEE) over the S1-S4 sites was 6.6-9.3, 6.6-7.9, 10.6-11.3 and 9.3-10.6 micromol m(-2) s(-1), respectively. The rate of CO2 uptake over S1 was lower than that for C4 grasses elsewhere because of pasture degradation. Soil respiration and temperature were exponentially related when soil water content (theta) was above 0.083 m(3) m(-3); however, soil respiration declined markedly as theta decreased from 0.083-0.090 to 0.033-0.056 m(3) m(-3). There were bursts of CO2 emission when dry soils were rewetted by rainfall. During the wet season, all sites constituted carbon sinks with maximum net daily ecosystem production (NEP) of 2.1, 1.7, 2.1 and 2.1 g C m(-2) day(-1), respectively. During the dry season, the savanna sites (S2-S4) became carbon sources with maximum emission fluxes of -0.5, -1.4 and -1.6 g C m(-2) day(-1), respectively, whereas the tall-grass field (S1) remained a carbon sink with a maximum NEP of 0.3 g C m(-2) day(-1) at the end of the season. For all measurement periods, annual NEP of sites S1-S4 was 366, 6, 116 and 139 g C m(-2), respectively. Comparisons of carbon source/sink dynamics across a wide range of savannas indicate that savanna carbon budgets can change in sign and magnitude. On an annual basis, gross primary production over the S1-S4 stands was 797, 803, 136 and 1230 g C m(-2), respectively. Net primary productivity (NPP) of the S1-S4 stands, calculated from eddy covariance measurements as the daily sum of NEE and day and night heterotrophic respiration was 498, 169, 181 and 402 g C m-2 year-1, respectively. These values were slightly higher than NPP based on harvest measurements (432, 162, 176 and 386 g C m(-2) year(-1), respectively), presumably because fine roots were incompletely harvested. Soil water content limited carbon uptake at all sites, and water-use efficiency (WUE) was related to rainfall dynamics. During the dry season, all sites except the cultivated tall-grass Andropogon field (S1) had a negative WUE. Although our results are specific to the Orinoco vegetational mosaic, the effects of land-use practices on the controls and physiological functions of the studied ecosystems may be generalized to other savannas.

Water flux through a semi-deciduous forest grove of the Orinoco savannas.

Water relations were analysed in a semi-deciduous forest grove occurring in the oxisols of the Orinoco savannas. This grove has a shallow unconsolidated ironstone cuirass, which is overlaid by a sandy loam layer (0.0-0.5 m) that contains more than 90% of the grove forest root phytomass. Evapotranspiration and through drainage were calculated by using data from the soil profile as related to physical characteristics of the site root zone, hydraulic conductivity, volumetric water content and potential hydraulic gradient. Mean annual evapotranspiration was 783 mm year-1 and annual through drainage below the root zone was 14% (162 mm year-1) of the gross rainfall. This drainage recharged the 42% of the annual saturation deficit of the water table. Similar mean annual evapotranspiration (770 mm year-1) was also calculated by using the water balance components. The mean daily coupling omega factor (Ω) between the grove canopy and the surrounding atmosphere indicated that a high degree of coupling (Ω=0.14±0.16) occurs in the grove and evapotranspiration was mainly controlled by surface conductance. As the dry season proceeded, the soil saturation deficit (δθ) increased rapidly resulting in a threshold surface conductance (0.030-0.005 m s-1) for δθ ranging from 0.05 to 0.10. Hypotheses to explain the omnipresence of perennial species in the wide range of physical conditions in neotropical savannas are discussed.