Stem water storage capacity and diurnal patterns of water use were studied in five canopy trees of a seasonal tropical forest in Panama. Sap flow was measured simultaneously at the top and at the base of each tree using constant energy input thermal probes inserted in the sapwood. The daily stem storage capacity was calculated by comparing the diurnal patterns of basal and crown sap flow. The amount of water withdrawn from storage and subsequently replaced daily ranged from 4 kg d -1 in a 0·20-mdiameter individual of Cecropia longipes to 54 kg d -1 in a 1·02-m-diameter individual of Anacardium excelsum, representing 9-15% of the total daily water loss, respectively. Ficus insipida, Luehea seemannii and Spondias mombin had intermediate diurnal water storage capacities. Trees with greater storage capacity maintained maximum rates of transpiration for a substantially longer fraction of the day than trees with smaller water storage capacity. All five trees conformed to a common linear relationship between diurnal storage capacity and basal sapwood area, suggesting that this relationship was species-independent and size-specific for trees at the study site. According to this relationship there was an increment of 10 kg of diurnal water storage capacity for every 0·1 m 2 increase in basal sapwood area. The diurnal withdrawal of water from, and refill of, internal stores was a dynamic process, tightly coupled to fluctuations in environmental conditions. The variations in basal and crown sap flow were more synchronized after 1100 h when internal reserves were mostly depleted. Stem water storage may partially compensate for increases in axial hydraulic resistance with tree size and thus play an important role in regulating the water status of leaves exposed to the large diurnal variations in evaporative demand that occur in the upper canopy of seasonal lowland tropical forests.

Source water used by woody perennials in a Brazilian savanna (Cerrado) was determined by comparing the stable hydrogen isotope composition (deltaD) of xylem sap and soil water at different depths during two consecutive dry seasons (1995 and 1996). Plant water status and rates of water use were also determined and compared with xylem water deltaD values. Overall, soil water deltaD decreased with increasing depth in the soil profile. Mean deltaD values were -35 per thousand for the upper 170 cm of soil and -55 per thousand between 230 and 400 cm depth at the end of the 1995 dry season. Soil water content increased with depth, from 18% near the surface to about 28% at 400 cm. A similar pattern of decreasing soil water deltaD with increasing depth was observed at the end of the 1996 dry season. Patterns consistent with hydraulic lift were observed in soil profiles sampled in 1995 and 1997. Concurrent analyses of xylem and soil water deltaD values indicated a distinct partitioning of water resources among 10 representative woody species (five deciduous and five evergreen). Among these species, four evergreen and one deciduous species acquired water primarily in the upper soil layers (above 200 cm), whereas three deciduous and one evergreen species tapped deep sources of soil water (below 200 cm). One deciduous species exhibited intermediate behavior. Total daily sap flow was negatively correlated with xylem sap deltaD values indicating that species with higher rates of water use during the dry season tended to rely on deeper soil water sources. Among evergreen species, minimum leaf water potentials were also negatively correlated with xylem water deltaD values, suggesting that access to more readily available water at greater depth permitted maintenance of a more favorable plant water status. No significant relationship between xylem water deltaD and plant size was observed in two evergreen species, suggesting a strong selective pressure for small plants to rapidly develop a deep root system. The degree of variation in soil water partitioning, leaf phenology and leaf longevity was consistent with the high diversity of woody species in the Cerrado.

The savannas of central Brazil (cerrado) form the second most extensive plant formation in South America. In Brazil, the 2·0 × 10 6 km 2 of land area covered by cerrado vegetation is exceeded only by the Amazonian rain forest (Ratter 1992). The cerrado is characterized by a markedly seasonal rainfall regime with a 4-5 month dry season, old oligotrophic soils and frequent fires. The fires occur either naturally, or as part of a land-management system applied to increase the abundance and palatability of the relatively shallow-rooted grasses for cattle (Skole et al. 1994). The impact of the resulting decrease in density of the more deeply rooted trees and shrubs on ecosystem water fluxes is unknown.Cerrado comprises three principal physiognomic subtypes ranging from cerradão, medium to tall woodlands with closed or semi-closed canopies, to cerrado sensu stricto, a savanna woodland with 3. The dependence of maximum whole-plant sap flow rates on sapwood area was similar among all four species during both the wet and dry seasons. When total daily sap flow on a leaf area basis was normalized by the daily average air saturation deficit (ASD), only one of the four species showed significantly greater water use during the wet season. 4. Although seasonal differences in regulation of transpiration were not pronounced, strong stomatal limitation of both maximum daily transpiration rates and total daily transpiration was evident during both the wet and dry seasons. Sap flow typically increased sharply in the morning, briefly attained a maximum value by about 09.30-10.30 h, then decreased sharply, despite steadily increasing solar radiation and atmospheric evaporative demand. 5.The total leaf area-specific apparent hydraulic conductance of the soil/leaf pathway (G t ) varied among plants and diurnally. The identical linear dependence of transpiration and stomatal conductance (g s ) on G t among the four study species suggested that stomatal adjustment to variation in G t limited transpiration over the entire range of G t observed. 6. When g s was normalized for daily variation in G t , about 80% of the remaining variation in g s was associated with variation in ASD. The results suggested that transpiration in these species was not limited by soil water availability per se, but by high atmospheric evaporative demand and hydraulic constraints possibly arising from their deep rooting habit.

Source water used by plants of several species in a semi-evergreen lowland tropical forest on Barro Colorado Island, Panama, was assessed by comparing the relative abundance of deuterium, D, versus hydrogen, H (stable hydrogen isotope composition, δD) in xylem sap and in soil water at different depths, during the dry season of 1992. Ecological correlates of source water were examined by comparing xylem water δD values with leaf phenology, leaf water status determined with a pressure chamber, and rates of water use determined as mass flow of sap using the stem heat balance method. Soil water δD values decreased sharply to 30 cm, then remained relatively constant with increasing depth. Average δD values were-13‰, for 0-30 cm depth and-36.7‰ for 30-100 cm depth. Soil water δD values were negatively associated with soil water content and soil water potential. Concurrent analyses of xylem water revealed a high degree of partitioning of water resources among species of this tropical forest. Xylem water δD of deciduous trees (average=-25.3±1.4‰) was higher than that of evergreen trees (average=-36.3±3.5‰), indicating that evergreen species had access to the more abundant soil water at greater depth than deciduous species. In evergreen shade-tolerant and high-light requiring shrubs and small trees, δD of xylem water was negatively correlated with transpiration rate and leaf water potential indicating that species using deeper, more abundant water resources had both higher rates of water use and more favorable leaf water status.

Environmental and physiological regulation of transpiration were examined in several gap-colonizing shrub and tree species during two consecutive dry seasons in a moist, lowland tropical forest on Barro Colorado Island, Panama. Whole plant transpiration, stomatal and total vapor phase (stomatal + boundary layer) conductance, plant water potential and environmental variables were measured concurrently. This allowed control of transpiration (E) to be partitioned quantitatively between stomatal (g ) and boundary layer (g) conductance and permitted the impact of invividual environmental and physiological variables on stomatal behavior and E to be assessed. Wind speed in treefall gap sites was often below the 0.25 m s stalling speed of the anemometer used and was rarely above 0.5 m s, resulting in uniformly low g (c. 200-300 mmol m s) among all species studied regardless of leaf size. Stomatal conductance was typically equal to or somewhat greater than g . This strongly decoupled E from control by stomata, so that in Miconia argentea a 10% change in g when g was near its mean value was predicted to yield only a 2.5% change in E. Porometric estimates of E, obtained as the product of g and the leaf-bulk air vapor pressure difference (VPD) without taking g into account, were up to 300% higher than actual E determined from sap flow measurements. Porometry was thus inadequate as a means of assessing the physiological consequences of stomatal behavior in different gap colonizing species. Stomatal responses to humidity strongly limited the increase in E with increasing evaporative demand. Stomata of all species studied appeared to respond to increasing evaporative demand in the same manner when the leaf surface was selected as the reference point for determination of external vapor pressure and when simultaneous variation of light and leaf-air VPD was taken into account. This result suggests that contrasting stomatal responses to similar leaf-bulk air VPD may be governed as much by the external boundary layer as by intrinsic physiological differences among species. Both E and g initially increased sharply with increasing leaf area-specific total hydraulic conductance of the soil/root/leaf pathway (G ), becoming asymptotic at higher values of G. For both E and g a unique relationship appeared to describe the response of all species to variations in G. The relatively weak correlation observed between g and midday leaf water potential suggested that stomatal adjustment to variations in water availability coordinated E with water transport efficiency rather than bulk leaf water status.