Water balance of 31 Mile Brook, Western Australia

[摘要] Reductions in rainfall in Perth’s water supply catchments of around 15% since 1975, have resulted in water yield falling to only 58% of the long-term average. During the local drought of 2001-2002 some streams in the jarrah forest ceased flowing for the first time in living memory (Water Corporation, 2005), and in 2010 inflow to the Perth reservoirs was less than 2% of the long-term average (data courtesy of the Water Corporation).Understanding the water balance of the jarrah forest that covers most of these catchments is important if we are to develop improved understanding of how the catchments work and thereby improve management to balance the water supply needs, and needs of aquatic and terrestrial ecosystems and the commercial forests.This aims of this project were to:1.Quantify the contribution of different vegetation components to the catchment water balance at 31 Mile Brook.2.Quantify the impact of rainfall on catchment moisture storage.We undertook a field experiment to quantify the water balance of a small catchment in the northern jarrah forest. Overstorey transpiration was measured using the heat ratio method in four plots, one each at the top and bottom of two hillside transects.A ‘notching’ treatment was applied to small areas within the catchment but had little impact on the measurement plots.The understorey and ground evapotranspiration was estimated at the same four sites using the Bowen Ratio method based on temperature and humidity profiles through the canopy and measurements of available energy.Ground surface and low vegetation evapotranspiration was measured using a specially built dome chamber in which the change in absolute humidity was used as an estimate of evaporation over the known surface area.These measurements were scaled up to estimate the catchment water balance from 2004 to 2010 using Landsat-derived leaf area index.Finally catchment water storage change was calculated as the residual of the water balance from rainfall minus total evapotranspiration.Key findings were:Rainfall and RunoffMean annual rainfall at 31 Mile Brook from 2004 to 2010 was 947 mm, well short of the long-term average of circa 1100 mm.The period included two notable drought years: 748 mm in 2006 and 524 mm in 2010.Mean run-off over the period was only 61 mm year-1.Rainfall interceptionA good relationship was found between leaf area index and canopy interception of rainfall. Interception increased from 10% of rainfall for an L of one to 20% for an L of 2.5.Mean annual interception was 121 mm.Overstorey transpirationThe average overstorey transpiration per unit L at the four intensive sites at 31 Mile Brook was 311 mm year-1. There was no evidence that stands in lower parts of the landscape with access to relatively shallow groundwater were transpiring more water per unit ground area than smaller regrowth trees on hill slopes. In fact, there was less overstorey transpiration at the blackbutt (E. megacarpa) dominated plot, which was poorly stocked, had small L and contained numerous large trees. The mean annual sap velocities of trees in our study (8-11 cm hr-1) were typical of eucalypts based on previous studies. Mean annual overstorey transpiration was 427 mm which, combined with canopy interception, resulted in overstorey evapotranspiration of 548 mm year-1 on average.Ground and understorey evaporationThere was no relationship between understorey cover and overstorey leaf area index. Ground evaporation was positively correlated with reference evaporation in winter-spring but negatively correlated with reference evaporation in summer-autumn, and totalled 130 mm year-1.We estimated an understorey leaf area index of 0.45, and calculated that the annual transpiration per unit leaf area index of understorey was approximately 550 mm year-1.Annual evaporation of combined understorey and ground was 400 mm year-1.Change in catchment moisture storageWe calculated th

[发布日期] 2012-08-13 [发布机构] CSIRO

[学科分类] 地球科学(综合)