Search

Estimates of forest stocking density per hectare (NHa) are important in characterising ecological conditions and assessing changes in forest dynamics after disturbances due to pyrogenic, anthropogenic and biotic factors. We use Unmanned Aircraft Systems (UAS) LiDAR with mean point density of 1485 points m−2 across 39 flight sites to develop a bottom-up approach for individual tree and crown delineation (ITCD). The ITCD algorithm was evaluated across mixed species eucalypt forests (MSEF) using 2790 field measured stem locations across a broad range of dominant eucalypt species with randomly leaning trunks and highly irregular intertwined canopy structure. Two top performing ITCD algorithms in benchmarking studies resulted in poor performance when optimised to our plot data (mean Fscore: 0.61 and 0.62), which emphasises the challenge posed for ITCD in the structurally complex conditions of MSEF. To address this, our novel bottom-up ITCD algorithm uses kernel densities to stratify the vegetation profile and differentiate understorey from the rest of the vegetation. For vegetation above understorey, the ITCD algorithm adopted a novel watershed clustering procedure on point density measures within horizontal slices. A Principal Component Analysis (PCA) procedure was then applied to merge the slice-specific clusters into trunks, branches, and canopy clumps, before a voxel connectivity procedure clustered these biomass segments into overstorey trees. The segmentation process only requires two parameters to be calibrated to site-specific conditions across 39 MSEF sites using a Shuffled Complex Evolution (SCE) optimiser. Across the 39 field sites, the ITCD algorithm had mean Fscore of 0.91, True Positive (TP) trees represented 85% of measured trees and predicted plot-level stocking (NP) averaged 94% of actual stocking (NOb). As a representation of plot-level basal area (BA), TP trees represented 87% of BA, omitted trees represented slightly smaller trees and made up 8% of BA, and a further 5% of BA had commission error. Spatial maps of NHa using 0.5 m grid-cells showed that omitted trees were more prevalent in high density forest stands, and that 63% of grid-cells had a perfect estimate of NHa, whereas a further 31% of the grid-cells overestimate or underestimate one tree within the search window. The parsimonious modelling framework allows for the two calibrated site-specific parameters to be predicted (R2: 0.87 and 0.66) using structural characteristics of vegetation clusters within sites. Using predictions of these two site-specific parameters across all sites results in mean FScore of 0.86 and mean TP of 0.77, under circumstances where no ground observations were required for calibration. This approach generalises the algorithm across new UAS LiDAR data without undertaking time-consuming ground measurements within tall eucalypt forests with complex vegetation structure.

Field studies that investigate sediment transport between debris‐flow‐producing headwaters and rivers are uncommon, particularly in forested settings, where debris flows are infrequent and opportunities for collecting data are limited. This study quantifies the volume and composition of sediment deposited in the arterial channel network of a 14‐km2 catchment (Washington Creek) that connects small, burned and debris‐flow‐producing headwaters (

Between 2003 and 2013, drought, large wildfires, and record-breaking rainfall contributed to debris flows in southeast Australia that appear to be unprecedented in spatial extent and density in historical records. Here, we used a debris-flow inventory from this period of dry and wet extremes to examine the processes and climatic controls underlying the regionwide debris-flow response. Results reveal shallow landslides and surface runoff as two distinct initiation mechanisms, linked to different geologic settings and contrasting hydroclimatic conditions. Landslide-generated debris flows occurred in sandy soils, independent of past fires, and were tightly controlled by extreme rainfall causing saturation and mass failure during La Nina periods. In contrast, runoff-generated debris flows occurred in clay-rich soils from short and intense rainstorms after wildfires in dry conditions, often associated with El Nino. Thus, it appears that both ends of the wet and dry climate extremes produce the same general geomorphic response, debris flows, but in different areas and by different initiation processes. Debris-flow activity is therefore at a maximum when amplitude and frequency of climate oscillations are large. Debris flows in southeast Australia are likely to become more frequent and widespread as wildfire activity and rainfall intensity are predicted to increase.

Forest growth dynamics affect streamflow with changes in actual evapotranspiration ( AET ) during forest regeneration. Using a “top-down” model approach, we present a seasonal forest hydrology model that estimates streamflow in forested catchments containing sparse soil and climate data. Across a 1423 km 2 remote study area, a rainfall interpolation procedure using available rainfall data and information on terrain was integrated into a new seasonal streamflow model, called ABCF, which uses catchment sapwood area ( SA ) as the emergent property that equilibrates with potential evapotranspiration ( PET ), and a soil storage threshold that reduces AET below PET when soil water is limiting. We produce seasonal estimates of streamflow with Nash Sutcliffe efficiencies of 0.85, 0.87, and 0.91 for three major catchments within the study area. A fundamental feature of the “top-down” model approach is the use of LiDAR data and forest inventory data to model forest structural properties that relate strongly with SA . Building on our previous work with this modelling framework, our representation of eco-hydrological properties of the forest has been refined with a more accurate procedure for estimating stand mean sapwood thickness, and hence SA , and a remotely sensed tree stocking density ( N ) of old-growth forests to correct the temporal evolution of N as a means to improve SA estimates. Regional consistency of model parameters shows that the “top-down” modelling framework may be used to estimate streamflow in ungauged catchments using a forest growth model. The seasonal model generalised for both water-limited and water-unlimited forest conditions has significant potential for application in water supply planning and drought security.

Most of the water that ends up in Melbourne’s water supply catchments originates from wet Eucalyptus forests that are dominated by Eucalyptus regnans, the tallest known angiosperm on earth. Studies had shown that catchments that are dominated by these forests can experience a significant long-term (>100 years) reduction in streamflow after a stand-replacing fire, which was attributed to higher water-use of the dense overstory regrowth. However, despite several lines of evidence, the direction, extent and duration of post-fire hydrological behaviour vary significantly between catchments and between fire events. Here we propose that this variability is caused by initial stocking density and species composition after the fire, and the climatic conditions that prevail during forest regeneration that affect tree growth and mortality rates. In order to test the hypothesis, we formulated an ecohydrological model that simulates hydrology, growth and forest dynamics of E. regnans and Acacia dealbata, which are known to compete for resources during the initial stages of vegetation recovery. The new model shows high skill in predicting long-term streamflow when compared to observations using multiple sources of data. Simulation analysis shows that the direction, extent and duration of post-fire hydrological behaviour are sensitive to initial stocking density and to the relative abundance of species that regenerate after the fire, which influence the rate of self-thinning during stand development. Furthermore, simulation results show that the observed long-term reduction in streamflow is less likely to occur when the forest would have been less dense before the fire, which theoretically could only occur when a high proportion of the short-lived A. Dealbata regenerated after the previous fire. This highlights the importance of including mechanisms that control the effect of species composition on forest dynamics when modelling the effect of possible future climatic scenarios on water yield.

This chapter contains fill-in-the-blank questions in hydrology and water management. Questions are consist of a missing word (or pair of words) in a sentence. The reader is asked to provide the missing words. Fill-in-theblank questions demand a high level of thought about the discipline area and what is missing from the sentence.

This chapter contains essays questions and projects about hydrology and water management which can be enjoyable to do and can add greatly to student's depth of understanding.

This chapter contains image-based questions on hydrology and water management. Images were created to illustrate the answer. In other cases, literature was used to find a relevant maps or illustrations.

This chapter contains true or false questions in hydrology and watershed management. Readers chose true or false values in response to a statement.

Forested catchments are critical to water supply in major cities. Many of these catchments face the threat of postwildfire erosion, which can contaminate reservoir water. The aim of this paper is to determine the probability and duration of disruptions to treatability due to runoff-generated debris flows in the first year after a wildfire, before substantial vegetation recovery takes place. We combine models of reservoir hydrodynamics, postfire erosion, and stochastic rainfall to determine probability and magnitude of sediment concentration at the reservoir water offtake. Central to the paper is our technique for linking model components into a risk framework that gives probabilities to the number of days that the turbidity threshold for treatment is exceeded. The model is applied to the Upper Yarra reservoir, which is the linchpin of the water supply system for Melbourne in SE Australia. However, the framework is applicable to other unfiltered water supply systems where suspended sediment is a risk to treatability. Results show that postwildfire erosion poses a substantial threat, with a relatively high probability (annual exceedance probability = 0.1–0.3) of water being untreatable for >1 year following a high-severity wildfire. Important factors that influence the risk include postwildfire runoff potential, reservoir temperature, and the amount of clay-sized grains in eroding headwaters. Assumptions about spatial-temporal rainfall attributes, reservoir hydrodynamics, and the catchment erosion potential are all important sources of error in our estimate of risk. Our approach to risk quantification will help support planning, risk management, and strategic investment to mitigate impacts.

Climate drives the coevolution of vegetation and the soil that supports it. Wildfire dramatically affects many key eco‐hydro‐geomorphic processes, but its potential role in coevolution of soil‐forest systems has been largely overlooked. The steep landscapes of southeastern Australia provide an excellent natural laboratory to study the role of fire in the coevolution of soil and forests, as they are characterized by temperate forest types, fire frequencies, and soil depths that vary systematically with aridity. The aims of this study were (i) to test the hypothesis that in Southeastern Australia, fire‐related processes are critical to explain the variations in coevolved soil‐forest system states across an aridity gradient and (ii) to identify the key processes and (iii) feedbacks involved. To achieve these aims, we developed a numerical model that simulates the coevolution of soil‐forest systems which employ eco‐hydro‐geomorphic processes that are typical of the flammable forests of southeastern Australia. A stepwise model evaluation, using measurements and published data, confirms the robustness of the model to simulate eco‐hydro‐geomorphic processes across the aridity gradient. Simulations that included fire replicated patterns of observed soil depth and forest cover across an aridity gradient, supporting our hypothesis. The contribution of fire to coevolution increased in magnitude with aridity, mainly due to the higher fire frequency and lower post‐fire infiltration capacity, increasing the rates of fire‐related surface runoff and erosion. Our results show that critical feedbacks between soil depth, vegetation, and fire frequency dictate the trajectory and pace of the coevolution of flammable temperate forests and soils.

Fire as a hydrologic agent has been most frequently examined in terms of erosion and water quality, with studies on the ecohydrology expressed as evapotranspiration/streamflow often focussing on short term perturbation that relaxes with vegetation recovery. Far more dramatic ecohydrologic impacts are possible if repeated fire disturbance leads to species change. Such a scenario occurs in some forests in south-eastern Australia, a region that is among the most flammable in global terms due to the confluence of climatic and stand productivity factors. The most vulnerable of these forests are the “ash” type – mainly Eucalyptus regnans and E. delegatensis. The E.regnans ecology has evolved with long fire intervals as medium/hot fire kill the trees, which then regenerate as single aged strands. However there have been several large short interval fire events in mountain forests (eg. 1926-1939, 2003-2006-2009-2019) in the past decades that overlap in area. E.regnans, and the other ash-type species, require 15-20 years to develop seed. If re-burnt, the stands cannot naturally regenerate. Frequently acacia and other understorey species colonise the sites, resulting in a dramatic change in forest structure and biomass.The implications of this change are significant, with potentially high magnitude changes in ecohydrologic functioning. Further, these areas are the principal water supply catchments the city of Melbourne (> 4 M pop.) and a number of other towns. The impact of high frequency fire that is predicted to increase under climate change therefore has the potential to change ecology, hydrology and essential ecosystem services, in this case, water supply.An extensive field experimentation and modelling program set out to (a) investigate the climatic conditions under which these wet forests burn and the sensitivity of these drivers to predicted climate change; and (b) evaluate the eco-hydrologic impact of a species change from E.regnans to acacia species over an age sequence of 80 years.Results revealed there is an envelope of dry surface soil and maximum vapour pressure deficit (VPD) within which there is a 50% chance of uncontrolled fire. The most damaging fires occurred when VPD was within the upper 0.01% of values and available surface soil water below 55%. Modelling suggests this conjunction of drivers will increase significantly in the future.Stand structure, particularly sapwood area, diverged between the eucalypts and acacias at age 10-20 years, with the difference increasing until acacia death at age 80. This structural parameter scales with ET, with acacias exhibiting a marked decline over time relative to E. regnans. This ET change is principally driven by sapwood area. These differences increase as the stands age, resulting in A.dealbata using around 30% of an E.regnans stand at age 80. This represents a fundamental change in eco-hydrology, and suggests a system pushed to a state of disequilibrium. The stand structural attributes over the age sequence indicate a large change in carbon stocks, resulting in significant alteration of both carbon and water cycles under this disturbance. The results have significant implications for water supply, forest ecosystem services, and system feedbacks of flammability-fire-ecohydrology.

Most of the drinking water provided to the Melbourne metropolitan area originates from catchments that are covered by tall-wet eucalyptus forests, which have an estimated mean fire return interval of 80-150 yr. When stand-replacing fires occur in these forests, they result in a significant reduction in streamflow for an extended period of time due to the recovery strategy of the dominant tree species (Eucalyptus regnans and Eucalyptus delegatensis) and the water use of the dense regrowth. Current hydrological models that express this phenomenon are based on empirical data and lack the mechanistic explanation that links changes in streamflow with forest stand dynamics after disturbance. Here, for the first time, we present a simple theoretical framework that shows that this post-fire reduction in streamflow could be explained by the self-thinning behaviour of forest regrowth, which is driven by competition for water and light during recovery after a stand-replacing fire. First, we show that the trend in streamflow following a stand-replacing fire can be replicated simply by using the generic self-thinning line (which represents the maximum carrying capacity of a forest stand for a given mean tree diameter) of the dominant tree species. We then go one step further and show that the magnitude of streamflow reduction and the time it takes for streamflow to recover to pre-fire conditions, are sensitive to both the recovery success and the environmental conditions that control the maximum vegetation carrying capacity across the catchments. By using a simple stand growth and mortality model, we link the competition for water and light and the self-thinning behaviour of the forest to evapotranspiration and streamflow trajectories. This theory provides a simple alternative approach that can be used to improve models that predict streamflow from forested catchments after stand-replacing fires.

Carbon and water fluxes are often assumed to be coupled as a result of stomatal regulation during dry conditions. However, recent observations evidenced increased transpiration rates during isolated heatwaves across a range of eucalypt species under experimental and natural conditions, with inconsistent effects on photosynthesis (ranging from increases to stark declines). To improve the empirical basis for understanding carbon and water fluxes in forests under hotter and drier climates, we measured the water use of dominant trees and ecosystem‐scale carbon and water exchange in a temperate eucalypt forest over three summer seasons. The forest maintained photosynthesis within 16% of baseline rates during hot and dry conditions, despite ~70% reductions in canopy conductance during a 5‐day heatwave. While carbon and water fluxes both decreased by 16% on exceptionally dry days, gross primary productivity only decreased by 5% during the hottest days and increased by 2% during the heatwave. However, evapotranspiration increased by 43% (hottest days) and 74% (heatwave), leading to ~40% variation in traditional water use efficiency (water use efficiency = gross primary productivity/evapotranspiration) across conditions and approximately two‐fold differences between traditional and underlying or intrinsic water use efficiency on the same days. Furthermore, the forest became a net source of carbon following a 137% increase in ecosystem respiration during the heatwave, highlighting that the potential for temperate eucalypt forests to act as net carbon sinks under hotter and drier climates will depend not only on the responses of photosynthesis to higher temperatures and changes in water availability, but also on the concomitant responses of ecosystem respiration.

Post-wildfire runoff and erosion are major concerns in fire-prone landscapes around the world, but these hydro-geomorphic responses have been found to be highly variable and difficult to predict. Some variations have been observed to be associated with landscape aridity, which in turn can influence soil hydraulic properties. However, to date there has been no attempt to systematically evaluate the apparent relations between aridity and post-wildfire runoff. In this study, five sites in a wildfire burnt area were instrumented with rainfall-runoff plots across an aridity index (AI) gradient. Surface runoff and effective rainfall were measured over 10 months to allow investigation of short- (peak runoff) and longer-term (runoff ratio) runoff characteristics over the recovery period. The results show a systematic and strong relation between aridity and post-wildfire runoff. The average runoff ratio at the driest AI site (33.6%) was two orders of magnitude higher than at the wettest AI site (0.3%). Peak runoff also increased with AI, with up to a thousand-fold difference observed during one event between the driest and wettest sites. The relation between AI, peak 15-min runoff (Q15) and peak 15-min rainfall intensity (I15) (both in mm h-1) could be quantified by the equation: Q15 = 0.1086I15 × AI 2.691 (0.65

Forest species composition may change following a disturbance. This change can affect long term water yield from forested catchments when the replaced and replacement species have different evapotranspiration rates. Following strip-thinning experiments that removed 50% of the overstorey basal area in several Eucalyptus regnans water supply catchments in south eastern Australia, Acacia spp. (Acacia dealbata and Acacia melanoxylon) became the dominant overstorey species in most of the cut strips. More recently, low regeneration of E. regnans following wildfires in 2009 may result in mixed Acacia and E. regnans stands in some catchments. We compared transpiration of E. regnans and Acacia stands in the uncut and cut strips of a catchment that was strip-thinned in early 1980s (Crotty Creek). We also compared transpiration and throughfall in a mixed E. regnans–A. dealbata regrowth stand 20 years after clear-fell logging (Road 8). Sap flow was measured for 13 and 6 months at Crotty Creek and Road 8, respectively. In both studies, mean daily sap flow density of Acacia spp. was lower than of E. regnans. Estimated Leaf Area Index of E. regnans stands was slightly greater than that of Acacia spp. Stomatal conductance (gc), estimated by inverting the Penman–Monteith equation, differed between the species suggesting species-level physiological differences with Acacia being more sensitive to vapour pressure deficit than E. regnans. Throughfall measurements at Road 8 indicated interception was slightly higher in A. dealbata but only enough to offset about 13% of the difference in transpiration. Replacement of E. regnans by Acacia dominated stands may, therefore, decrease catchment evapotranspiration and increase streamflow.

The conventional Land Surface Temperature (LST)–Normalized Difference Vegetation Index (NDVI) trapezoid model has been widely used to retrieve vegetation water stress. However, it has two inherent limitations: (1) its complex and computationally intensive parameterization for multi-temporal observations and (2) deficiency in canopy water content information. We tested the hypothesis that an improved water stress index could be constructed by the representation of canopy water content information to the LST–NDVI trapezoid model. Therefore, this study proposes a new index that combines three indicators associated with vegetation water stress: canopy temperature through LST, canopy water content through Surface Water Content Index (SWCI), and canopy fractional cover through NDVI in one temporally transferrable index. Firstly, a new optical space of SWCI–NDVI was conceptualized based on the linear physical relationship between shortwave infrared (SWIR) and soil moisture. Secondly, the SWCI–NDVI feature space was parameterized, and an index d(SWCI, NDVI) was computed based on the distribution of the observations in the SWCI–NDVI spectral space. Finally, standardized LST (LST/long term mean of LST) was combined to d(SWCI, NDVI) to give a new water stress index, Temperature Vegetation Water Stress Index (TVWSI). The modeled soil moisture from the Australian Water Resource Assessment—Landscape (AWRA-L) and Soil Water Fraction (SWF) from four FLUXNET sites across Victoria and New South Wales were used to evaluate TVWSI. The index TVWSI exhibited a high correlation with AWRA-L soil moisture (R2 of 0.71 with p < 0.001) and the ground-based SWF (R2 of 0.25–0.51 with p < 0.001). TVWSI predicted soil moisture more accurately with RMSE of 21.82 mm (AWRA-L) and 0.02–0.04 (SWF) compared to the RMSE ranging 28.98–36.68 mm (AWRA-L) and 0.03–0.05 (SWF) were obtained for some widely used water stress indices. The TVWSI could also be a useful input parameter for other environmental models.

Climate warming is expected to increase fire frequency in many productive obligate seeder forests, where repeated high-intensity fire can initiate stand conversion to alternative states with contrasting structure. These vegetation–fire interactions may modify the direct effects of climate warming on the microclimatic conditions that control dead fuel moisture content (FMC), which regulates fire activity in these high-productivity systems. However, despite the well-established role of forest canopies in buffering microclimate, the interaction of FMC, alternative forest states and their role in vegetation–fire feedbacks remain poorly understood. We tested the hypothesis that FMC dynamics across alternative states would vary to an extent meaningful for fire and that FMC differences would be attributable to forest structural variability, with important implications for fire-vegetation feedbacks. FMC was monitored at seven alternative state forested sites that were similar in all aspects except forest type and structure, and two proximate open-weather stations across the Central Highlands in Victoria, Australia. We developed two generalised additive mixed models (GAMMs) using daily independent and autoregressive (i.e., lagged) input data to test the importance of site properties, including lidar-derived forest structure, in predicting FMC from open weather. There were distinct differences in fuel availability (days when FMC &lt, 16%, dry enough to sustain fire) leading to positive and negative fire–vegetation feedbacks across alternative forest states. Both the independent (r2 = 0.551) and autoregressive (r2 = 0.936) models ably predicted FMC from open weather. However, substantial improvement between models when lagged inputs were included demonstrates nonindependence of the automated fuel sticks at the daily level and that understanding the effects of temporal buffering in wet forests is critical to estimating FMC. We observed significant random effects (an analogue for forest structure effects) in both models (p &lt, 0.001), which correlated with forest density metrics such as light penetration index (LPI). This study demonstrates the importance of forest structure in estimating FMC and that across alternative forest states, differences in fuel availability drive vegetation–fire feedbacks with important implications for forest flammability.

In this study we examine the performance of four models that simulate radiation under forest canopies. The models are different in terms of how transmittance of radiation is conceptualised and the data sources used to obtain parameters. Two models (PAI MS and PAI NC models) use plant area index ( PAI ) to represent transmission. A third model (PL model) represents transmission as a function of the path length ( L ) of a directional beam as it crosses the canopy. The fourth model (LPI model) is based on a light penetration index obtained directly from lidar. The PAI MS and PL models were calibrated using 6 months of radiation data at 4 independent sites. The LPI and PAI NC models are uncalibrated. Performance was assessed using sub-daily and daily radiation measurements during summer (December–March) at 10 sites in forests ranging from open dry forests ( PAI = 1.6) to tall temperate forests ( PAI = 4.6). Mean annual precipitation ranged from 760 to 1750 mm across the domain. The PL model with a calibrated extinction coefficient ( k 2 = 0.033) was the most accurate model within sites (R-square = 0.32–0.89 for sub-daily radiation) and across all sites (R-square = 0.82 for daily and sub-daily radiation). The LPI model performed well at most sites, but displayed some systematic bias in dense forests resulting in lower performance (R-square = 0.61 across all sites) while PAI NC was negatively biased and a poor predictor across all sites (overall R-square = 0). The PAI MS model, with a calibrated extinction coefficient ( k 1 = 0.48), produced good results, but inspection of sub-daily radiation patterns show that the model tends to underestimate (overestimate) sub-canopy radiation during times of high (low) radiation. This is because transmittance is highest when the sun elevation is high (i.e. short path length), resulting in sub-canopy radiation that varies non-linearly with the intensity of above canopy radiation. Models that explicitly include path length in the transmission term therefore provide more flexibility for capturing sub-daily, seasonal and latitudinal variation due to sun position and vegetation structure. Other advantages of the path length approach include i) simple parameterisation using tree height as input and ii) explicit representation of diffuse and direct radiation, which can be important when accounting for terrain-effects on energy balance at the forest floor. We therefore recommend path length based modelling approaches for investigating hydrological processes below the canopy in temperate Eucalyptus forests.

After wildfire a hitherto unexplained erosion process that some authors have called ‘miniature debris flows on hillslopes’ and that leave behind levee-lined rills has been observed in some regions of the world. Despite the unusual proposition of debris flow on planar hillslopes, the process has not received much attention. The objectives of this study were to (1) accumulate observational evidence of Hillslope Debris Flows (HDF) as we have defined the process, to (2) understand their initiation process by conducting runoff experiments on hillslopes, to (3) propose a conceptual model of HDF, and to (4) contrast and classify HDF relative to other erosion and transport processes in the post-wildfire hillslope domain. HDF have been observed at relatively steep slope gradients (0.4–0.8), on a variety of geologies, and after fire of at least moderate severity and consist of a lobe of gravel- to cobble-sized material 0.2–1 m wide that is pushed by runoff damming up behind it. During initiation, runoff moved individual particles that accumulated a small distance downslope until the accumulation of grains failed and formed the granular lobe of the HDF. HDF are a threshold process, and runoff rates of 0.5 L s− 1 2 L s− 1 were required for their initiation during the experiments. The conceptual model highlights HDF as a geomorphic process distinct from channel debris flows, because they occur on planar, unconfined hillslopes rather than confined channels. HDF can erode very coarse non-cohesive surface soil, which distinguishes them from rill erosion that have suspended and bedload transport. On a matrix of slope and grain size, HDF are enveloped between purely gravity-driven dry ravel, and mostly runoff driven bedload transport in rills.

To better understand water and energy cycles in forests over years to decades, measurements of spatial and long-term temporal variability in evapotranspiration (Ea) are needed. In mountainous terrain, plot-level measurements are important to achieving this. Forest inventory data including tree density and size measurements, often collected repeatedly over decades, sample the variability occurring within the geographic and topographic range of specific forest types. Using simple allometric relationships, tree stocking and size data can be used to estimate variables including sapwood area index (SAI), which may be strongly correlated with annual Ea. This study analysed plot-level variability in SAI and its relationship with overstorey and understorey transpiration, interception and evaporation over a 670 m elevation gradient, in non-water-limited, even-aged stands of Eucalyptus regnans F. Muell. to determine how well spatial variation in annual Ea from forests can be mapped using SAI. Over the 3 year study, mean sap velocity in five E. regnans stands was uncorrelated with overstorey sapwood area index (SAI) or elevation: annual transpiration was predicted well by SAI (R2 0.98). Overstorey and total annual interception were positively correlated with SAI (R2 0.90 and 0.75). Ea from the understorey was strongly correlated with vapour pressure deficit (VPD) and net radiation (Rn) measured just above the understorey, but relationships between understorey Ea and VPD and Rn differed between understorey types and understorey annual Ea was not correlated with SAI. Annual total Ea was also strongly correlated with SAI: the relationship being similar to two previous studies in the same region, despite differences in stand age and species. Thus, spatial variation in annual Ea can be reliably mapped using measurements of SAI.

Forest carbon and water fluxes are often assumed to be coupled as a result of stomatal regulation during dry conditions. However, recent observations have indicated increased transpiration rates during isolated heat waves across a range of eucalypt species under experimental and natural conditions, with inconsistent effects on photosynthesis (ranging from an increase to a near total decline). To improve the empirical basis for understanding carbon and water fluxes in forests under hotter and drier climates, we measured the water use of dominant trees, and the ecosystem-scale carbon and water exchange in a mature temperate eucalypt forest over three summer seasons. The forest maintained photosynthesis within 16% of peak photosynthesis rates during all conditions, despite up to 70% reductions in canopy conductance during a 5-day heatwave. While carbon and water fluxes both decreased by 16% on exceptionally dry summer days, GPP was sustained at the cost of up to 74% increased water loss on the hottest days and during the heatwave. This led to ∼40% variation in ecosystem water use efficiency over the three summers, and ∼two-fold differences depending on the way water use efficiency is calculated. Furthermore, the forest became a net source of carbon following a 137% increase in ecosystem respiration during the heat wave, highlighting that the potential for temperate eucalypt forests to remain net carbon sinks under future climates will depend not only on their potential to maintain photosynthesis during higher temperatures, but also on responses of ecosystem respiration to changes in climate.Key PointsGPP of temperate eucalypts was sustained at the cost of increased water use during hot periods, but both fluxes decreased during dry periods.WUE estimates for the same period differed up to two-fold depending on the way it was calculated.Doubling of ecosystem respiration turned the forest from a net sink into a net source of carbon during a longer heatwave.

Fire can create, strengthen or destroy soil water repellency, with potential implications for soil infiltration, surface runoff and erosion. Laboratory studies suggest fire-induced changes to water repellency relate to soil temperatures during the burn. However, relations between temperature and repellency are rarely tested in the field where spatial variations in fuel type, soil type and soil moisture may lead to more complex responses to fire. Furthermore, few studies link point-scale water repellency measurements to hydro-geomorphic effects at larger spatial scales. Therefore, the purposes of this study were to (1) measure soil temperatures during prescribed burns, (2) investigate the in-situ effects of soil heating on soil water repellency and (3) investigate the subsequent effects of soil water repellency on infiltration, runoff and erosion. Heat-sensitive liquids and thermocouples measured soil temperatures at points within three prescribed burns. Soil water repellency and infiltration were measured at the same burnt points and at adjacent unburnt points 2–6 weeks post-burn. Rainfall simulations quantified plot-scale infiltration, surface runoff and erosion at one site. Peak temperatures at the surface were highly variable (averaging 238, 129 and 327 °C at each burn) while sub-surface temperatures were lower (averaging 75, less than 79 and 108 °C at each burn). Heating durations were short with surface temperatures greater than 400 °C lasting on average for 2 s and temperatures greater than 200 °C lasting on average for 6 s. Despite pre-existing water repellency in unburnt areas, water repellency was strengthened in burnt areas at two sites. High temperatures were associated with more repellency, even when the temperatures exceeded laboratory-defined thresholds for the destruction of water repellency. Short heating durations may explain why the laboratory-defined temperature thresholds were not applicable in the field. Point-scale steady-state infiltration rates were significantly lower for burnt compared with unburnt areas, reflecting the greater water repellency. However, other factors (e.g. macropore flow, soil sealing and reduced vegetation cover) are also likely to have caused unexpectedly high or low infiltration rates for some points and higher infiltration rates at the plot-scale than expected. The relative importance of water repellency to infiltration, surface runoff and erosion appeared to vary depending on the spatial scale of measurement. These issues of scale together with the apparent spatial heterogeneity of prescribed burnt landscapes warrant the use of connectivity modelling as a way to link point and plot scale measures to impacts at larger spatial scales.

Effective hydraulic conductivity (Ke) for Hortonian overland flow modeling has been defined as a function of rainfall intensity and runon infiltration assuming a distribution of saturated hydraulic conductivities (Ks). But surface boundary condition during infiltration and its interactions with the distribution of Ks are not well represented in models. As a result, the mean value of the Ks distribution ((Formula presented.)), which is the central parameter for Ke, varies between scales. Here we quantify this discrepancy with a large infiltration data set comprising four different methods and scales from fire-affected hillslopes in SE Australia using a relatively simple yet widely used conceptual model of Ke. Ponded disk (0.002 m2) and ring infiltrometers (0.07 m2) were used at the small scales and rainfall simulations (3 m2) and small catchments (ca 3000 m2) at the larger scales. We compared (Formula presented.) between methods measured at the same time and place. Disk and ring infiltrometer measurements had on average 4.8 times higher values of (Formula presented.) than rainfall simulations and catchment-scale estimates. Furthermore, the distribution of Ks was not clearly log-normal and scale-independent, as supposed in the conceptual model. In our interpretation, water repellency and preferential flow paths increase the variance of the measured distribution of Ks and bias ponding toward areas of very low Ks during rainfall simulations and small catchment runoff events while areas with high preferential flow capacity remain water supply-limited more than the conceptual model of Ke predicts. The study highlights problems in the current theory of scaling runoff generation.

This study represents a uniquely high-resolution observation of postwildfire runoff and erosion from dry forested uplands of SE Australia. We monitored runoff and sediment load, and temporal changes in soil surface properties from two (0.2–0.3 ha) dry forested catchments burned during the 2009 Black Saturday wildfire. Event-based surface runoff to rainfall ratios approached 0.45 during the first year postwildfire, compared to reported values

Debris flows are an important erosion process in wildfire-prone landscapes. Predicting their frequency and magnitude can therefore be critical for quantifying risk to infrastructure, people and water resources. However, the factors contributing to the frequency and magnitude of events remain poorly understood, particularly in regions outside western USA. Against this background, the objectives of this study were to i) quantify sediment yields from post-fire debris flows in southeast Australian highlands and ii) model the effects of landscape attributes on debris flow susceptibility. Sediment yields from post-fire debris flows (113–294 t ha− 1) are 2–3 orders of magnitude higher than annual background erosion rates from undisturbed forests. Debris flow volumes ranged from 539 to 33,040 m3 with hillslope contributions of 18–62%. The distribution of erosion and deposition above the fan were related to a stream power index, which could be used to model changes in yield along the drainage network. Debris flow susceptibility was quantified with a logistic regression and an inventory of 315 debris flow fans deposited in the first year after two large wildfires (total burned area = 2919 km2). The differenced normalised burn ratio (dNBR or burn severity), local slope, radiative index of dryness (AI) and rainfall intensity (from rainfall radar) were significant predictors in a susceptibility model, which produced excellent results in terms identifying channels that were eroded by debris flows (Area Under Curve, AUC = 0.91). Burn severity was the strongest predictor in the model (AUC = 0.87 when dNBR is used as single predictor) suggesting that fire regimes are an important control on sediment delivery from these forests. The analysis showed a positive effect of AI on debris flow probability in landscapes where differences in moisture regimes due to climate are associated with large variation in soil hydraulic properties. Overall, the results from this study based in the southeast Australian highlands provide a novel basis upon which to model sediment delivery from post-fire debris flows. The modelling approach has wider relevance to post-fire debris flow prediction both from risk management and landscape evolution perspectives.

Most of the surface water for natural environmental and human water uses in southeast Australia is sourced from forested catchments located in the higher rainfall areas. Water yield of these catchments is mainly affected by climatic conditions, but it is also greatly affected by vegetation cover change. Bushfires are a major natural disturbance in forested catchments and potentially modify the water yield of the catchments through changes to evapotranspiration (ET), interception and soil moisture storage. This paper quantifies the impacts of bushfire and climate variability on streamflow from three southeast Australian catchments where Ash Wednesday bushfires occurred in February 1983. The hydrological models used here include AWRA-L, Xinanjiang and GR4J. The three models are first calibrated against streamflow data from the pre-bushfire period and they are used to simulate runoff for the post-bushfire period with the calibrated parameters. The difference between the observed and model simulated runoff for the post-bushfire period provides an estimate of the impact of bushfire on streamflow. The hydrological modelling results for the three catchments indicate that there is a substantial increase in streamflow in the first 15 yr after the 1983 bushfires. The increase in streamflow is attributed to initial decreases in ET and interception resulting from the fires, followed by logging activity. After 15 yr, streamflow dynamics are more heavily influenced by climate effects, although some impact from fire and logging regeneration may still occur. It is shown that hydrological models provide reasonable consistent estimates of forest disturbance and climate impacts on streamflow for the three catchments. The results might be used by forest managers to understand the relationship between forest disturbance and climate variability impacts on water yield in the context of climate change.

Mean sapwood thickness, measured in fifteen 73 year old Eucalyptus regnans and E. delegatensis stands, correlated strongly with forest overstorey stocking density (R2 0.72). This curvilinear relationship was used with routine forest stocking density and basal area measurements to estimate sapwood area of the forest overstorey at various times in 15 research catchments in undisturbed and disturbed forests located in the Great Dividing Range, Victoria, Australia. Up to 45 years of annual precipitation and streamflow data available from the 15 catchments were used to examine relationships between mean annual loss (evapotranspiration estimated as mean annual precipitation minus mean annual streamflow), and sapwood area. Catchment mean sapwood area correlated strongly (R2 0.88) with catchment mean annual loss. Variation in sapwood area accounted for 68% more variation in mean annual streamflow than precipitation alone (R2 0.90 compared with R2 0.22). Changes in sapwood area accounted for 96% of the changes in mean annual loss observed after forest thinning or clear-cutting and regeneration. We conclude that forest inventory data can be used reliably to predict spatial and temporal variation in catchment annual losses and streamflow in response to natural and imposed disturbances in even-aged forests. Consequently, recent advances in mapping of sapwood area using airborne light detection and ranging will enable high resolution spatial and temporal mapping of mean annual loss and mean annual streamflow over large areas of forested catchment. This will be particularly beneficial in management of water resources from forested catchments subject to disturbance but lacking reliable long-term (years to decades) streamflow records.

© 2015 Elsevier B.V. The objective of this study was to estimate the recovery trajectory of evapotranspiration (Et) and streamflow (Q) in resprouting forested catchments following wildfire. Recovery dynamics were assessed in mixed species eucalypt forests in south-eastern Australia which recover from disturbance largely via vegetative resprouting, and to a lesser degree, via seedling recruitment. Changes in Et were evaluated in two ways. Firstly, we developed semi-empirical models of post-fire Et following moderate and high severity wildfire. These models were based on datasets of plot-scale Et, measured within five years post-fire, and published literature on post-fire changes in vegetation structure. Secondly, we analysed long-term Q records (25years) from a mixed species catchment, including a 1-5year period following a predominately moderate severity wildfire. We found that the overall length of recovery time for Et and Q following wildfire was 8-12years, which is much less than for eucalypt forests recovering via seedlings only. This emphasises the importance of functional responses to fire in forest ecosystems as a key driver of the hydrologic resilience of catchments, with resprouting forest types conferring relatively rapid recovery following disturbance. We also found that the recovery trajectory of post-fire Et was dependent on fire severity. Increased Et and consequent declines in Q occurred following moderate severity fire. In contrast, there was no evidence of increased Et following high severity fire. Based on patterns of long-term Q and rainfall observed in a small mixed species catchment, declines in Q due to increased Et following moderate severity wildfire were of similar magnitude to Q declines driven by a drought that coincided with the fire. We conclude that the coincidence of wildfire with drought exacerbates reductions in Q under moderate severity fire, resulting in greater Q declines. This is due to the enhanced rates of Et, primarily driven by regenerating seedlings and higher rates of transpiration from surviving trees.

Summary Data from two Australian paired-catchment projects were used to assess the effect of length of the calibration period on the quality of calibration achieved. The data were divided into a “calibration” and a “verification” set. The Nash–Sutcliffe (N–S) coefficient of efficiency was used to assess the quality of prediction of the verification set as a function of the length of the calibration set. The results showed a rapid initial increase in quality of calibration with increasing calibration length. This then “plateaued”. With simple linear regression models, reasonable calibration (N–S > 0.7) was achieved in 60 days and good calibration (N–S > 0.8) in 100 days. More complex models achieved good calibration after 300 days of data. In general, there were no increases in the N–S value achieved after 3 years – the main advantage of longer calibrations appeared to be lower mean errors. Similar results were obtained with daily, monthly, quarterly, or annual subdivisions of flows. All residuals suffered from autocorrelation and non-normality; the former was removed by an autoregressive technique, but the latter appears implicit in the technique. Simulation of the use of an n-fold data examination technique to monitor the development of calibration as data flowed in substantially reproduced this result. This appears to be a good strategy for hydrologists monitoring development of calibration in a continuing project. Paired-catchment experimentation is a robust experimental technique but would benefit from application of a set of protocols prescribing techniques.

The radiative index of dryness (or aridity index) is a non-dimensional measure of the long-term balance between rainfall and net radiation. Quantifying aridity requires spatially distributed information on net radiation and rainfall. The variability in net radiation in complex terrain can be modelled at high spatial resolution by combining point data with equations that incorporate the effects of elevation, surface geometry and atmospheric attenuation of incoming radiation. At large spatial scales and over long time periods, however, the combination of seasonality, year to year variations and spatial variability in climate result in complex spatial-temporal patterns of incoming radiation, which are more effectively captured in satellite-based measurements. This study uses a high resolution model of shortwave radiation as a tool for downscaling satellite-derived data on incoming radiation. The aim was to incorporate topographic effects on net radiation in complex terrain while retaining information on regional and seasonal trends captured in satellite data. The method relies on satellite-based measures of incoming radiation from the Australian Bureau of Meteorology (BoM) to provide the spatial coverage and long-term data that represent the average incoming radiation across the state of Victoria in southeast Australia. These long-term data were coupled with a topographic downscaling algorithm to produce estimates of net radiation and aridity at the resolution of a 20 m digital elevation model. Results show that annual precipitation (and cloud fraction) gradients drive the variability in aridity at large scales (10-100 km) while topography (e.g. slope aspect and slope angle) are the main drivers at small scales (e.g. 1 km). The aridity index varied between 0.24 and 10.95 across the state of Victoria. The effect of aridity on vegetation was apparent at local scales through systematic variations in tree-height along rainfall gradients and across aspects with different levels of exposure to solar radiation.

Summary Simple rainfall–runoff models used in the assessment of land-use change and to support forest water management are subjected to a selection process which scrutinises their veracity and integrity. Veracity, the ability of a model to provide meaningful information is assessed using performance criteria, incorporating: a popular mean square error (MSE) approach; empirical distribution functions and information criteria. Integrity, a model’s plausibility reflected in its ability to extract information from data, is assessed using a Bayesian approach. A delayed rejection, adaptive Metropolis algorithm is used with a generalised likelihood to calibrate the models. Predictive uncertainty is assessed using a split sample procedure which uses high runoff data for calibration and drier data for validation. A simple multiplicative latent variable is used to accommodate input uncertainty in rainfall data, enabling a distinction to be made between uncertainty associated with data, parameters and the models themselves. The study demonstrates: the focus provided by setting model evaluation in a philosophical context; the benefits of using a more meaningful range of performance criteria than MSE-based approaches and the insights into integrity provided by Bayesian analyses. A hyperbolic tangent model is selected as the best of five candidates for its superior veracity and integrity under Australian conditions. Models with extensive application in South Africa, Australia and USA are rejected. Challenges to applying this approach in water management are identified in the pragmatic nature of the sector, its capacity constraints and a tendency of researchers to place confidence in accepted methods at the expense of rigour.

Summary Wildfires can reduce infiltration capacity of hillslopes by causing (i) extreme soil drying, (ii) increased water repellency and (iii) reduced soil structure. High severity wildfire often results in a non-repellent layer of loose ash and burned soil overlying a water repellent soil matrix. In these conditions the hydraulic parameters vary across discrete layers in the soil profile, making the infiltration process difficult to measure and model. The difficulty is often exacerbated by the discrepancy between actual infiltration processes and the assumptions that underlie commonly used infiltration models, most of which stem from controlled laboratory experiments or agricultural environments, where soils are homogeneous and less variable in space and time than forest soils. This study uses a simple two-layered infiltration model consisting of surface storage (H), macropore flow (Kmac) and matrix flow (Kmat) in order to identify and analyze spatial–temporal infiltration patterns in forest soils recovering from the 2009 Black Saturday wildfires in Victoria, southeast Australia. Infiltration experiments on intact soil cores showed that the soil profile contained a region of strong water repellency that was slow to take on water and inactive in the infiltration process, thus restricting flow through the matrix. The flow resistance due to water repellent soil was represented by the minimum critical surface tension (CSTmin) within the top 10 cm of the soil profile. Under field conditions in small headwaters, the CSTmin remained in a water repellent domain throughout a 3-year recovery period, but the strength of water repellency diminished exponentially during wet conditions, resulting in some weather induced temporal variation in steady-state infiltration capacity (Kp). An increasing trend in macropore availability during recovery was the main source of temporal variability in Kp during the study period, indicating (in accordance with previous studies) that macropore flow dominates infiltration processes in these forest soils. Storage in ash and burned surface soil after wildfire was initially high (∼4 mm), then declined exponentially with time since fire. Overall the study showed that the two layered soil can be represented and parameterized by partitioning the infiltration process into surface storage and flow through a partially saturated and restrictive soil layer. Ash, water repellency and macropore flow are key characteristics of burned forest soils in general, and the proposed model may therefore be a useful tool for characterizing fire impact and recovery in other systems.

The objective of this research was to develop and parameterise a physically justified yet low-parameter model to quantify observed changes in surface runoff ratios with hillslope length. The approach starts with the assumption that a unit of rainfall-excess runoff generated at a point is a fraction β of precipitation P (m) which travels some variable distance down a slope before reinfiltrating, depending on the local rainfall, climate, soils, etc. If this random distance travelled Y is represented by a distribution, then a survival function will describe the probability of this unit of runoff travelling further than some distance x (m). The total amount of per unit width runoff Q (m2) flowing across the lower boundary of a slope of length λ (m) may be considered the sum of all the proportions of the units of rainfall excess runoff integrated from the lower boundary x = 0 to the upper boundary x = λ of the slope. Using these assumptions we derive a model Q(λ) = βPμλ/(μ + λ), (μ > 0, 0 ≤ β ≤ 1, λ ≥ 0) that describes the change in surface runoff with slope length, where μ (m) is the mean of the random variable Y. Dividing both sides of this equation by Pλ yields a simple two-parameter equation for the dimensionless hillslope runoff ratio Qh(λ) = βμ/(μ + λ). The model was parameterised with new rainfall and runoff data collected from three replicates of bounded 2 m wide plots of four different lengths (0.5, 1.0, 2.0 and 4.0 m) for 2 years from a forested SE Australian site, and with 32 slope length–runoff data sets from 12 other published studies undertaken between 1934 and 2010. Using the parameterised model resulted in a Nash and Sutcliffe statistic between observed and predicted runoff ratio (for all data sets combined) of 0.93, compared with –2.1 when the runoff ratio was fixed at the value measured from the shortest plot. Copyright © 2014 John Wiley & Sons, Ltd.

Forested catchments generally present conserved aquatic ecosystems without anthropogenic disturbances, however, forest management operations can degrade these environments, including their water quality. Despite the potential degradation, few studies have analyzed the effects of forest management in subtropical regions, especially in forest plantations with intensive management, such as Eucalyptus plantations in Brazil. The intensive management of those plantations is characterized by fast-growing, short rotation cycles, and high productivity. This study aimed to assess the effects of Eucalyptus plantations harvesting on the concentration and exportation of nutrients and suspended solids in subtropical streams. Results showed that clear-cut harvesting and subsequent forest management operations do not alter most of the concentration of nitrate, potassium, calcium, and magnesium. The concentration of suspended solids increased during the first year after timber harvesting in all studied catchments, however, the increases were statistically significant in only two catchments. In the first year after harvest, it was observed an increment of water yield/precipitation ratio at three catchments, which also increased export of nutrients and suspended solids. Our results showed that harvesting of fast-growing Eucalyptus forest plantations partially affected sediment exports and did not compromise water quality in the studied catchments. However, the catchment land-use design, especially related to road density and land-use composition, showed significant relationship with sediment exportation.

Forest thinning has the potential to be used as a water supply augmentation strategy through the reduction of sapwood and leaf areas that also decreases stand evapotranspiration. This study investigated the long term effects of a range of thinning treatments (patch-cutting, uniform thinning, strip-thinning, understorey removal) on the vegetation structure and water yield of mountain ash ( Eucalyptus regnans ) forests in south eastern Australia. The paired catchment method and Light Detection and Ranging (LiDAR) data were utilised to assess the post-thinning changes in water yield and vegetation structure respectively. The first post-treatment phase started at the end of the treatments in the late 1970s/early 1980s until the suspension of measurements in 1997, the second post-treatment phase started when the measurements resumed in 2007/2008 until a wildfire in February 2008. Up to 36% increase in annual water yield was detected during the first post-treatment phase. Strip-thinning produced the largest cumulative increase in water yield (up to 1813 mm), followed by uniform thinning and patch-cutting. The magnitude of the treatment effect appeared to be amplified during a very wet/dry period, which may be due to post-thinning changes in the runoff generation processes. During the second post-treatment phase, a trend of decreased water yield was detected at several treated catchments. Canopy height profiles (CHPs) and mean effective leaf area index (LAI) that were derived from LiDAR data, along with recent forest inventory data, indicated a recovery of vegetation cover. This may partially explain the decrease in water yield. Meanwhile, the decade-long drought (1997–2008) that coincided with this period of decreased water yield might have exacerbated its magnitude. Thinning may not be a net water gain over the long term if the recent trend of decreasing water yield continues.

Rainfall-runoff modelling is widely used for runoff estimation at the catchment scale. However, its simulation capability is sometimes influenced because of rapid land cover changes occurring in catchments. This paper investigates whether modification of a rainfall-runoff model, Xinanjiang, by the incorporation of dynamic remote sensing data (MODIS leaf area index (LAI) and albedo) can improve runoff estimates for four south-east Australian catchments which experienced severe bushfire impacts. The results show that incorporation of remote sensing LAI and albedo data into the modified Xinanjiang model can improve model performance in three wetter bushfire impacted catchments, compared to the modified model using mean annual vegetation data as model inputs. The improvement is indicated by a slight increase (0.01–0.07) in the Nash–Sutcliffe efficiency of daily runoff and noticeable decrease (3–11%) in volumetric errors. However, use of vegetation dynamics does not improve runoff time series simulation in a dry catchment for which mean annual runoff is only 38 mm/yr. It indicates that incorporation of vegetation dynamic data into Xinanjiang model may show more benefits for catchments located in the wet regions

Fire severity and burn patchiness are frequently cited as important to post-fire surface runoff and erosion, yet few studies quantify their effects. A better understanding of their role is needed to predict post-fire erosion and design prescribed burns. Therefore, this study quantified the effects of fire severity and burn patchiness on surface runoff, erosion and hydrologic connectivity using 116 unbounded runoff samplers. The samplers were installed in recently prescribed burnt dry eucalypt forest in Victoria, Australia. Sediment loads over 16-months were approximately three orders of magnitude higher on burnt compared with unburnt hillslopes while differences in runoff and erosion between the low and high severity hillslopes were relatively small. Unburnt patches were often highly effective at reducing hydrologic connectivity from upslope burnt areas, with sediment loads over 16-months reduced by 1.3%, 98.1% and 99.9% downslope of 1, 5 and 10 m wide unburnt patches respectively. Hydrologic connectivity was limited most effectively by wider unburnt patches (10 m) and during lower magnitude storms. The results suggest overall that post-fire runoff and erosion may be substantially limited by unburnt patches while fire severity is a less important factor (within the context of prescribed burning). Consequently, post-fire erosion models should consider the spatial arrangement of unburnt patches, and unburnt patches (>10 m wide) should be retained within prescribed burns to minimise erosion.

[1] Erodibility describes the inherent resistance of soil to erosion. Hillslope erosion models typically consider erodibility to be constant with depth. This may not be the case after wildfire because erodibility is partly determined by the availability of noncohesive soil and ash at the surface. This study quantifies erodibility of burned soils using methods that explicitly capture variations in soil properties with depth. Flume experiments on intact cores from three sites in western United States showed that erodibility of fire-affected soil was highest at the soil surface and declined exponentially within the top 20 mm of the soil profile, with root density and soil depth accounting for 62% of the variation. Variation in erodibility with depth resulted in transient sediment flux during erosion experiments on bounded field plots. Material that contributed to transient flux was conceptualized as a layer of noncohesive material of variable depth (dnc). This depth was related to shear strength measurements and sampled spatially to obtain the probability distribution of noncohesive material as a function of depth below the surface. After wildfire in southeast Australia, the initial dnc ranged from 7.5 to 9.1 mm, which equated to 97–117 Mg ha−1 of noncohesive material. The depth decreased exponentially with time since wildfire to 0.4 mm (or

Erosion, flash floods and debris flows are hydro-geomorphic processes that intensify due to catchment disturbance by wildland fire. Predictive models of these processes are used by land managers to quantify rehabilitation effectiveness, prioritize resources and evaluate trade-offs between different management strategies. Predictions can be difficult to make, however, because of heterogeneous landscapes, stochastic rainfall, and the transient and variable fire effects. This paper reviews hydro-geomorphic response models for burned areas and explores how modelling approaches and sources of uncertainty change depending on the focus question (or purpose) and the associated spatial-temporal scale of the model domain. The review shows that current models focus primarily on predicting catchment responses during a recovery period (within-burn timescales), a relatively short temporal window during which rainfall is an important source of uncertainty. At longer (between-burn) timescales, the fire regime itself, and not just fire severity, becomes a variable component of the model. At this temporal scale, the catchment processes respond to variations in the frequency and severity with which a landscape is conditioned (or ‘primed’) by fire and rain storms. Conditioning is a stochastic process that is determined by the spatial-temporal overlap of fire disturbance and rain storms. The translation of overlaps to hydro-geomorphic responses is a function of intrinsic catchment attributes (e.g. permeability, slope and catchment area). Capturing the stochastic interplay between fire and rain storms is important when land-management questions shift towards the issues of climate change and landscape-scale interventions such as prescribed burning. The review therefore includes a discussion on fire and rainfall regimes as variables which drive decadal and regional variability in hydro-geomorphic processes.

Summary In February 2009, wildfire affected nine catchments, or approximately 28% of forested catchment area that supplies water to the city of Melbourne, Australia. This has potential to significantly affect the long term water use of these Eucalyptus forests and the consequential water yield because of the ecohydrologic response of some eucalypt species. Approximately 11% of the catchment area was severely burnt by intense fire, where vegetation mortality is higher. Catchment scale models using a physically-based approach were developed for the fire-affected water supply catchments. Different inputs of climate and forest mortality after fire were used to examine the relative contributions of rainfall, fire severity, forest type and forest age on post-fire streamflow. Simulations show the effect of fire on long term streamflow is likely to depend on a number of factors, the relative influence of which changes as rainfall becomes more limiting. Under average rainfall conditions, total reduction in post-fire streamflow after 100 years estimated to be between 1.4% (∼12 GL year −1 ) and 2.8% (∼24 GL year −1 ) are an order of magnitude lower than reductions in total catchment inflow during the period of low rainfall between 1997 and 2009, in which reservoir inflow was reduced by nearly 37%. The main reasons for the lower than expected changes in water yield are that a lower proportion of the catchments were affected by severe fire, and so mortality within the fire area was relatively low, and that the average age of the forest canopy (93 years) is younger than what is generally considered old growth forest. This means that the baseline (no-fire) streamflow used for reference is lower than would be expected with older, mature forest. The greatest post-fire affect on total water yield was predicted for the O’Shannassy catchment. This is due to the average forest age, which is the oldest of any of the catchments, that it has the highest average rainfall (1680 mm year −1 ), and that it contains the largest proportion of ash-type forest severely burnt (38.7%). Under wetter than average conditions, change in post-fire water yield is largely explained by changes in average age of the forest. The rates of ET are largely determined by the conductance and interception of the forest canopy. Under lower than average rainfall conditions, when water becomes limiting, annual rainfall is the best predictor of post-fire change in water yield. Under conditions of low rainfall and low soil water content that are conducive to larger wildfires, any initial increase in post-fire streamflow due to reduced canopy cover may not occur or be detected because a substantial soil water deficit must first be removed before appreciable changes in streamflow will occur. This partly explains the lack of increase in initial post-fire streamflow reported after wildfire compared to an increase in streamflow following forest harvesting experiments during wetter periods.

Large fires can have marked impacts on forest ecology and consequently on hydrology, including catchment water yields. Decisions on levels of fire protection and post-fire forest management require accurate information on the consequences of fire, including forest hydrological responses. In February 2009, wildfires burnt large areas of wet sclerophyll eucalypt forest in catchments supplying water to the city of Melbourne, Australia. Fire intensity was highly spatially variable, ranging from low intensity surface fire to high intensity crown fire. A survey of forest survival in three catchments 10–15 months after the fires revealed a good correspondence between overstorey survival and a satellite-based fire severity rating. In forests dominated by Eucalyptus regnans and Eucalyptus delegatensis, overstorey survival was close to zero in the two highest fire severity classes (crown burn and crown scorch). However, in areas mapped as intermediate fire severity (moderate crown scorch), survival varied markedly both spatially and between forest age classes. In fire tolerant eucalypt species, in mixed stands of Eucalyptus obliqua, Eucalyptus radiata, Eucalyptus cypellocarpa, and Eucalyptus baxteri, and in pure stands of Eucalyptus nitens, survival was >90% at high and intermediate fire severity. For all overstorey eucalypt species, survival in the lowest two fire severity classes (light or no crown scorch) was >75% but in most of these areas the fire temporarily removed the understorey. The fire severity mapping and ground survey indicate large areas of senescing (approximately 250–300 years old) and smaller areas of 70, 83 and 158 year old E. regnans and E. delegatensis forest were killed. Based on past research, this would be expected to result in very dense seedling regeneration, increasing evapotranspiration and decreasing catchment yields over the coming 2–3 decades. However, 2 years after the fire there were substantial differences between the three catchments in the density of seedling regeneration in these fire-killed areas. Where seedling densities are high, water yields are likely to decline. Areas with little or no eucalypt regeneration, however, might provide a persistent water yield increase. In addition, large areas of 70–158 year old E. regnans and E. delegatensis overstorey survived the fire where fire severity was low. Removal of the understorey from these areas and suppression of understorey regrowth by the intact overstorey canopy might result in water yield increases persisting for years to a decade or more.

Across southern Australia, a large proportion of urban water supply is sourced from mountainous catchments forested with native eucalypts. Mixed species eucalypt forest (MSEF) is the most common forest type in this region and occurs on relatively dry, fire prone sites, yet factors controlling forest water use and stream flow in response to topography, disturbance and drought are poorly understood. This study investigated the patterns and drivers of water balance over a 4 year period in a 1.36 km2, MSEF catchment by: quantifying spatial and temporal variability in evapotranspiration (Et) and its components; evaluating the abiotic, structural and physiological factors controlling water use across the catchment; and testing the effects of antecedent soil water conditions on water fluxes after drought. This was done using a ‘bottom up’ measurement approach that included stream flow and Et (sap flow, interception troughs and evaporation dome) and a simple empirical model of Et to track catchment response to drought. Spatial variability was considerable, with 40% lower rates of Et at an up slope plot compared to mid and bottom slope plots. Tree transpiration was the dominant flux annually and was correlated to reference Et (r2 = 0.35–0.80), implying strong limitation by atmospheric demand across the catchment. Annual Et totals were relatively consistent between years (841 ± 34 mm) despite large variation in rainfall (463–1179 mm y−1). Annual stream flow represented a very small proportion of the water budget (