Rob Vertessy

Factors determining relations between stand age and catchment water balance in mountain ash forests

Abstract There is a well-documented empirical relationship between stand age and water yield for mountain ash ( Eucalyptus regnans ) forested catchments in the Central Highlands of Victoria, Australia. Catchments covered with old-growth stands of mountain ash yield almost twice the amount of water annually as those covered with re-growth stands aged 25 years. In this paper, we provide a mechanistic hydro-ecologic explanation for this phenomenon. We measured leaf area index (LAI), sapwood area index (SAI) and various water balance components in several mountain ash stands, ranging in age between 5 and 240 years. Sap flow measurements show that sap velocity does not vary appreciably amongst stands of different ages, but a systematic decline in SAI with age produces a concomitant decrease in stand transpiration. The decline in overstory SAI is accompanied by a decline in overstory LAI. Understory LAI increases as the overstory LAI decreases, but this layer transpires at only about 63% of the mountain ash rate on a per unit leaf area basis. Hence, while total stand LAI decreases are modest over time, the trend is for a significant decline in total stand transpiration as the forest ages. Rainfall interception also declined over time and there was some indication that interception per unit leaf area also declined. Such reductions can be explained by lesser turbulent mixing and elevated humidity around the bulk of the leaf area in the mature forest. There were small decreases in forest floor evaporation through time, though this only accounted for about 5–8% of the site water balance. Our water balance measurements agree qualitatively with empirical water yield relationships developed for mountain ash forests, though the magnitudes of change differ.

Predicting water yield from a mountain ash forest catchment using a terrain analysis based catchment model

Abstract The structure, capabilities and performance of a distributed parameter hydrologic model are described. The model, called Topog-Yield, permits a transient analysis of unsaturated-saturated flow and evapotranspiration to be performed across complex terrain using a one-dimensional framework. It is applied to a 0.32 km 2 mountain ash ( Eucalyptus regnans ) forest catchment in the central Victorian highlands, Australia. We compare observed and predicted daily runoff values for the site over a continuous 12 year period (1972–1983) when the catchment vegetation was in an undisturbed climax condition. All input parameter values were based on published or measured data, although some variables were adjusted within the range of known variability to yield a best fit between predicted and observed streamflow in the first year of simulation, 1972. Although the model was ‘calibrated’ for the first year, all variables other than climatic inputs remained fixed for the following 11 years. Modelled and observed daily runoff values compare well throughout the period of simulation, despite a wide range of climatic conditions. When modelled daily runoff values were lumped on a monthly basis, the model was able to explain 87% of the variation in observed monthly streamflows over the 12 year period. Modelled annual runoff was within ±5% of observed values for 6 of the 12 years of record. Annual runoff prediction errors exceeded ±10% of observed values in only 2 of the 12 years. By the end of the 12 year simulation, the model had over-predicted runoff by less than 5%. Input data requirements and model results are discussed in the light of a preliminary sensitivity analysis.

Large-scale modelling of forest hydrological processes and their long-term effect on water yield

A water balance model was used to simulate the long-term increases in water yield with forest age which are observed in the mountain ash (Eucalyptus regnans) forests of Victoria, Australia. Specifically, the hypothesis was tested that water yield changes could be explained by changes in evapotranspiration resulting from changes in leaf area index (LAI). A curve predicting changes in the total LAI of mountain ash forest was constructed from ground-based observations and their correlation with Landsat Thematic Mapper measurements of the transformed normalized difference vegetation index (TNDVI). A further curve for mountain ash canopy LAI was constructed from destructive LAI measurements and stem diameter measurements. The curves were incorporated within Macaque, a large-scale, physically based water balance model which was applied to three forested catchments (total area 145 km2). The model was used to evaluate the effect of changes in LAI on predicted stream flow over an 82-year period spanning the 1939 wildfires which burnt most of the area. The use of the LAI curves induced improvement in the predicted hydrographs relative to the case for constant LAI, but the change was not large enough to account for all of the difference in water yield between old-growth and regrowth forests. Of a number of possibilities, concomitant changes in leaf conductance with age were suggested as an additional control on stream flow. These were estimated using data on stand sapwood area per unit leaf area and coded into Macaque. The hydrograph predicted using both the LAI curves and a new leaf conductance versus age curve accurately predicted the observed long-term changes in water yield. We conclude that LAI is a partial control on long-term yield changes, but that another ‘water use efficiency per unit LAI’ control is also operative. Copyright

Plantations, river flows and river salinity

Summary Large-scale plantation development will exert additional pressure on a water resource system that is already under considerable stress. Tree planting will reduce river flows and recharge to groundwater and, in certain circumstances, may lead to short-term worsening of river salinity prior to any improvement. Reductions in flow will be particularly problematic during dry spells, when water resources are sorely stretched. Most of the likely hydrologic impacts of afforestation can be predicted using current catchment models, but new field data are needed to test and improve their accuracy. Reductions in river flow induced by afforestation can be minimised with careful planning, and various strategies to minimise impact are recommended. It is argued that a regulatory framework needs to be erected to control the development of new plantations in order to complement other policies to preserve water resources, such as the cap on diversions in the Murray-Darling Basin and recently introduced legislation on farm dams. Given the future need to allocate additional river flows to the environment, new allocations of water to plantations should be offset by up-front transfers of water from other uses. We argue that water use by plantations should be factored into the water economy of catchments.

Improved methods to assess water yield changes from paired-catchment studies: application to the Maroondah catchments

The statistical methods underpinning analysis of streamflow data from paired-catchment studies have not changed much since the 1960s. Whilst such analyses are widespread in hydrologic practice and research, attention is rarely given to the problems of heteroscedacity, seasonality, serial correlation, and non-normally distributed variates. Each of these problems can potentially invalidate the basic assumptions upon which traditional statistical methods are based. We describe methods to contend with some of these problems and apply them to mountain ash (Eucalyptus regnans) forested catchments in the Maroondah Basin, south-eastern Australia. A seasonal regression model with lag-one auto-regressive (AR1) error was developed to predict monthly streamflow at treated catchments based on streamflow data from a control catchment. It is particularly well-suited to situations where little pre-treatment data is available. Differences between observed and predicted streamflow were used to quantify the effect of forest treatment on streamflow. Results from two catchment groups broadly matched the trend predicted by a previous regional model, with 2–3 year increases in streamflow, followed by decreases over the following one or two decades. A third group of catchments also showed initial increases, but expected subsequent decreases in streamflow were offset by the flow-increasing effects of insect infestations.

Stormflow generation and flowpath characteristics in an Amazonian rainforest catchment

The Amazon basin covers an area of roughly 7 × 106 km2 and encompasses diverse soil – landscape types with potentially differing hydrological behaviour. This study was conducted in the Ultisol landscape of the western Amazon basin in Peru. Processes of stormflow generation were investigated on an event basis in a first-order rainforest catchment to establish a causal link between soil physical and precipitation characteristics, hillslope flowpaths and stormflow hydrograph attributes. A sharp decrease in soil hydraulic conductivity with depth and high rainfall intensity and frequency favour rapid near-surface flowpaths, mainly in the form of saturation-excess overland flow and return flow. The latter results in an almost random occurrence of overland flow, with no obvious topographic control. Hillslope flowpaths do not vary much with respect to the hydrograph attributes time of rise, response time, lag time and centroid lag time. They have the same response time as streamflow, but a somewhat lower time of rise and significantly shorter lag times. The recession constant for hillslope hydrographs is about 10 min, in contrast to the streamflow recession constants of 28, 75 and 149 min. Stormflow generation in this Ultisol rainforest catchment differs strongly from that reported for Oxisol rainforest catchments. These two soilscapes may define a spectrum of possible catchment hydrological behaviour in the Amazon basin. Copyright

Distributed modeling of storm flow generation in an Amazonian rain forest catchment: Effects of model parameterization

We describe a process-based storm flow generation model, Topog_SBM consisting of a simple bucket model for soil water accounting, a one-dimensional kinematic wave overland flow scheme, and a contour-based element network for routing surface and subsurface flows. Aside from topographic data and rainfall the model has only six input parameters: soil depth (z), saturated hydraulic conductivity at the soil surface (K0), the rate of decay in K0 with depth (m), the Manning surface roughness parameter (n), the maximum (saturated) soil water content (θs), and the minimum (residual) soil water content (θr). However, the model is fully distributed, so these values can vary in magnitude across space. The model was applied to La Cuenca, a very small rainforest catchment in western Amazonia that has been well characterized in several hydrometric and hydrochemical investigations. Total runoff, peak runoff, time of rise, and lag time were predicted for 34 events of varying magnitudes and antecedent moisture conditions. We compared results for eight different model parameterizations or “sets”; four of these were freely calibrated to yield the best possible model fit to runoff data, whereas the other four were constrained (in various ways) by the use of actual K0 data gathered for the catchment. The eight sets were calibrated on either one of three events or on the three events jointly to illustrate the importance of calibration event selection on model performance. Model performance was evaluated by comparing observed and predicted (1) storm flow hydrograph attributes and (2) spatiotemporal patterns of overland flow occurrence across the catchment. The model generally predicted the right amount of runoff but usually underpredicted the peak runoff rate and overpredicted the time of rise. The “best” parameterization could credibly predict hydrographs for only about half of the events. Significant, and sometimes gross, errors were encountered for about one fourth of the events modeled, raising concerns in our minds about the a priori simulation of events that diverge too far from the conditions that the model was calibrated for. For the best parameterization we were able to predict an overland flow frequency distribution that accorded with field observations, though the model almost always overpredicted the spatial extent of overland flow. We concluded that model performance for the La Cuenca conditions could be enhanced by adding a “fast” subsurface flow pathway and/or by modifying the K0 versus depth decay function.

The sensitivity of a catchment model to soil hydraulic properties obtained by using different measurement techniques

Most studies on the use of physically based hydrological models have identified saturated hydraulic conductivity (Ksat) as one of the most sensitive input parameters. However, Ksat is also one of the most difficult landscape properties to measure accurately, casting doubt on the ability of modellers to estimate this parameter a priori for catchment simulations. Several studies have shown that Ksat estimates are greatly influenced by the measurement method used, primarily because of scale effects. In this paper, we evaluate the effect of Ksat measurement method on catchment simulations aimed at predicting water yield from forested catchments. A series of simulations are conducted using the Topog_Dynamic catchment model, with Ksat estimated by means of the constant head well permeameter, small core (6·3 cm×7·3 cm) and large core (22·3 cm×30 cm) methods. These were applied in a deep, permeable forest soil in which macropore flow has been noted to occur. The three measurement methods yielded very different Ksat estimates and these had a large effect on model results. The model predictions based on small core and well permeameter measurements were extremely poor, as these methods did not adequately account for preferential flow through the soil. The large core estimates of Ksat , which were one to three orders of magnitude higher than the values obtained by the other two techniques, produced good predictions of catchment discharge and known spatial patterns of water table depth. Our results highlight the need for caution when applying soil hydraulic measurements to catchment-scale models. Copyright

Large-scale distribution modelling and the utility of detailed ground data

A large-scale distribution function model was used to investigate the effect of differing parameter mapping schemes on the quality of hydrological predictions. Precipitation was mapped over a large forested catchment area (163 km 2 ) using both one-dimensional linear and three-dimensional non-linear interpolation schemes. Lumped stream flow predictions were found to be particularly sensitive to the different precipitation maps, with the three-dimensional map predicting 12% higher mean annual precipitation, resulting in 36% higher modelled stream flow over a three-year period. However, spatial predictions of stream flow appeared worse when derived from the three-dimensional map, which is considered the better of the two precipitation maps. This implies uncertainty in either the models response to precipitation or the precipitation mapping process (the 12% precipitation difference was strongly determined by a single, short term gauge). Leaf area index (LAI) was mapped using both remote sensing and species based methods. The two LAI maps had similar lumped mean values but exhibited significant spatial differences. The resulting lumped predictions of stream flow did not vary. This suggests a linear response of water balance to LAI in the non-water-limited conditions of the study area, and de-emphasizes the importance of quantifying relative spatial variations in LAI. Topographic maps were created for a small experimental subcatchment (15 ha) using both air photographic interpretation and ground survey. The two maps differed markedly and lead to significantly different spatial predictions of runoff generation, but nearly identical predicted hydrographs. Thus, at scales of small basins, accurate topographic mapping is suggested to be of little importance in distribution function modelling because models are unable to make use of complex spatial data. Predictions of water yield can be very sensitive (in the case of precipitation) or insensitive (in the case of small-scale topography) to changes in spatial parameterization. In either case, increased complexity in spatial parameterization does not necessarily result in better, or more certain prediction of hydrological response.

Early hydrological response to intense forest thinning in southwestern Australia

Abstract A small forested catchment in southwest Western Australia was thinned to study the effect on hydrology, wood production and disease escalation. This paper deals primarily with the hydrological aspects. The uniform, intensive thinning treatment reduced crown cover from 60 to 14%, which resulted in an increase in streamflow of approximately 20% of annual rainfall (260 mm for an average year) after 3 years, compared with a streamflow yield of 6% of annual rainfall before thinning. The deep groundwater attained a new equilibrium after 2 years, rising by approximately 2 m in the area adjacent to the swamp, and by 5 m upslope. The ephemeral shallow groundwater system expanded in duration from 2 months per year pre-treatment to approximately 6 months per year after thinning. The major components of streamflow generation present before thinning remained as the major components after thinning. The expansion of the saturated source area and the presence of a shallow groundwater system for extended periods, as a result of an increase in available water from the reduction in interception and evaporation from the overstorey, were considered to be the major causes of increased streamflow.