Paired Catchment Study Research Articles

Attempts to determine the effects of forest cover on stream flow by direct hydrological measurements in Los Negros, Bolivia

Underlying many payments for watershed services (PWS) schemes in tropical montane forest contexts is the assumption of a direct positive relationship between forest cover and dry season stream flow. We developed a low cost research program to assess the forest cover-stream flow relationship in the Los Negros watershed in eastern Bolivia. We asked three questions: (1) can watersheds that are similar enough to undertake paired catchment studies be identified using only simple parameters such as size, aspect and geographic proximity; (2) can a functioning locally based hydrological monitoring system be set up for less than $10 000 by training local farmers to collect hydrological data, and (3) can such data be used to improve the functioning of a PWS initiative? A land use map of the upper Los Negros valley was created from a 2005 Landsat image and a digital elevation model used to calculate physical and hydrological properties of 10 sub-watersheds. Farmers measured stream flow rates in these sub-watersheds from 2005 to 2008 and maintained 10 automatic rain gauges. We found no relationship between forest cover and stream flow. This may indicate that no such relationship exists, but could also reflect the short period of the study, the low quality of the data, and the fact that the sub-watersheds had relatively similar forest coverage (54–76%). We conclude that (1) watersheds can be identified as “similar-enough-for-analysis” using the criteria of size, aspect and proximity without undertaking further research, (2) a useful hydrological monitoring system can be developed for <$10,000 and (3) although our local farmers did not collect sufficiently high quality data to fully explore the forest/water relationship in Los Negros, with improvements in methodologies, low cost, locally based monitoring has the potential to be an important component of future PWS initiatives. We recommend that stream discharge should be calculated directly; only the most locally relevant hydrological criteria, rather than scientifically complete criteria, should be monitored; locally based monitoring must be institutionalized to reduce staff turnover and hydrological monitoring must be embedded within a context that makes it socially acceptable.

Implications of Uncertainty in a Pre-treatment Dataset when Estimating Treatment Effects in Paired Catchment Studies: Phosphorus Loads from Forest Clear-cuts

Estimates of increased nutrient export caused by forest clear-cuttings are mostly based on long-term paired catchment studies, where the treatment effect is calculated using pre-treatment dataset regression and post-treatment records of nutrient loads. In these studies uncertainty in the regression between the pre-treatment loads from the control and from the treatment catchments is typically neglected, even though it affects determination of the magnitude and duration of the treatment effect. This uncertainty is described in terms of variance in regression coefficients and residuals. The aim was to study how uncertainty in a pre-treatment dataset is propagated to estimates of the treatment effects using two sets of paired catchment data in eastern Finland, where the total phosphorus loads following forest clear-cutting and site preparation were investigated. The results showed that neglecting the uncertainty in the pre-treatment data did not change the conclusions when the treatment effect was strong enough. However, when the treatment effect was small, neglecting the uncertainty in the pre-treatment dataset easily leads to over-interpretation of the results. The uncertainty in a pre-treatment dataset should be taken into account in the paired catchment studies in order to avoid bias in the estimation of management effects on the loads.

Biogeochemical legacy of prescribed fire in a giant sequoia–mixed conifer forest: A 16‐year record of watershed balances

The effects of prescription burning on watershed balances of major ions in mixed conifer forest were examined in a 16‐year paired catchment study in Sequoia National Park, California. The objective was to determine whether fire‐related changes in watershed balances persist as long as estimated low‐end natural fire‐return intervals (≤10 years), and whether cumulative net export caused by fire could deplete nutrient stocks between successive fires. Inputs (wet + dry deposition) and outputs (stream export) of N, S, Cl−, HCO3−, Ca2+, Mg2+, Na+, K+, H+, and SiO2 were measured for 7 years preceding, and 9 years following, a prescribed burn of one of the catchments. After fire, runoff coefficients increased by 7% (in dry years) to 35% (in wet years). Inorganic N was elevated in stream water for 3 years after fire. Increased export of water, SO42−, Cl−, SiO2, and base cations continued through the end of the study. Pools and processes attributed to fire led to the cumulative loss, per hectare, of 1.2 kg N, 16 kg S, 25 kg Cl−, 130 kg Ca2+, 19 kg Mg2+, 71 kg Na+, 29 kg K+ and 192 kg Si, above that predicted by prefire regression equations relating export in the paired catchments. This additional export equaled &lt;1% of the N, up to one‐third of the Ca and Mg, and up to three‐fourths of the K, contained in the forest floor prior to combustion. Changes in watershed balances indicated that low‐end natural fire‐return intervals may prevent complete reaccumulation of several elements between fires.

Maximizing water yield with indigenous non-forest vegetation: a New Zealand perspective

Provision of clean freshwater is an essential ecosystem service that is under increasing pressure worldwide from a variety of conflicting demands. Water yields differ in relation to land-cover type. Successful resource management therefore requires accurate information on yields from alternative vegetation types to adequately address concerns regarding water production. Of particular importance are upper watersheds/catchments, regardless of where water is extracted. Research in New Zealand has shown that, when in good condition, indigenous tall tussock grasslands can maximize water yield relative to other vegetation cover types. A long-term hydrological paired-catchment study revealed reductions (up to 41% after 22 years) in water yielded annually from an afforested catchment relative to adjacent indigenous grassland. Furthermore, a stable isotope assessment showed that water from fog may substantially contribute to yield in upland tussock grasslands. The tall tussock life-form and its leaf anatomy and physiology, which minimize transpiration loss, appear to be the differentiating factors. Thus, maintaining dominance of such cover is important for water production, especially in upland catchments. Ecological analogues and integrated land-use planning are discussed in the context of this essential ecosystem service. Water management programs in other countries are reviewed and that of South Africa is commended as a model.

The effect of converting a native broad-leaved forest to a coniferous plantation forest on annual water yield: A paired-catchment study in northern Japan

This study examined the change of water yield when a broad-leaved forest is converted to a coniferous plantation forest after clear-cutting in northern Japan where monthly precipitation is evenly distributed. Annual water yield after clear-cut was greater than observed in a broad-leaved forest and then decreased with the growth of a coniferous plantation forest. This reduction ceased after 30 years from the planting of coniferous seedlings. Water yield of a coniferous forest at this stage was smaller than observed in a broad-leaved forest. Results from monthly water yield examinations suggested that the difference in annual water yield was primarily caused by the difference in winter interception loss between broad-leaved deciduous and coniferous evergreen forests. The above results suggest that converting coniferous forests into broad-leaved forests can be used to increase water yield in these regions.

The Brigalow Catchment Study: II. Clearing brigalow (Acacia harpophylla) for cropping or pasture increases runoff

The Brigalow Catchment Study (BCS) was established to determine the impact on hydrology when brigalow land is cleared for cropping and grazing. The paired catchment study was commenced in 1965 using catchments of approximately 15 ha, with natural vegetation dominated by brigalow scrub (Acacia harpophylla). Three contiguous catchments were selected near Theodore in central Queensland to represent the extensive brigalow bioregion of central and southern Queensland and northern New South Wales (~40 Mha). The hydrology of the 3 catchments was characterised during a 17-year calibration period (1965–81). The catchments were considered hydrologically similar, with sufficient data available for an empirical comparison between catchments. In 1982, two of the catchments were cleared, with one developed for cropping and the other sown to improved pasture. The third catchment was used as an uncleared control. Hydrologic characteristics were then compared for the following 21 years. In their virgin state, the catchments behaved similarly, with average annual runoff being 5% of annual rainfall. Once cleared, total runoff from the cropping catchment increased to 11% of annual rainfall and total runoff from the pasture catchment increased to 9% of annual rainfall; however, timing of the individual runoff events varied between land uses. In order to confirm that changes in hydrology were a function of land use and not just seasonal variability or sampling error, several analytic techniques were used: a simple comparison of runoff totals, comparison of events, comparison of probability of exceedance for daily runoff, and comparison of predicted and observed runoff using a water balance modelling approach.

The Brigalow Catchment Study: I. Overview of a 40-year study of the effects of land clearing in the brigalow bioregion of Australia

This paper describes a long-term, paired-catchment study, its broad findings, and considerations for future resource management of brigalow lands in north-eastern Australia. The Brigalow Catchment Study (BCS) commenced in 1965 with a pre-clearing calibration phase of 17 years to define the hydrology of 3 adjoining catchments (12–17 ha). After 2 catchments were cleared in 1982, 3 land uses (brigalow forest Acacia harpophylla, cropping, and grazed pasture) were monitored for water balance, resource condition and productivity, providing information for scientific understanding and resource management of the major land uses of the brigalow bioregion. In addition, this paper draws upon several project reviews to highlight the value of the BCS as an ‘outdoor laboratory’, its data resource, and to reflect on the study’s scientific rigor to support present and future value. An assessment of the BCS against national and international attributes of best practice for long-term studies showed the study to rate highly in aspects of design, implementation, monitoring, and data management, and moderately in formal publication, strategic management, and networking. The literature shows that Brigalow Catchment Study is the longest paired-catchment study in Australia, and continues to sample the interactions between climate, soils, water, land use, and management. Finally, this paper provides the context for component-specific papers on changes in hydrology, productivity, and salt balance. Results from the study to date include: a doubling of runoff after clearing, a reduction in wheat yield by more than 60% over 20 years, a halving of pasture availability 3 years after clearing, a decline in cattle liveweight gain of 4 kg/ha.year over an 8-year period with a constant stocking rate, and the leaching of 60% of the root-zone (0–1.5 m) chloride after clearing for cropping. Unanticipated applications of the data from the study include: (i) a crucial set of soil samples for calibration of the RothC soil carbon model used to estimate Australia’s soil carbon emissions; and (ii) estimates of deep drainage as a basis for salinity risk assessment in the region.

Afforestation, seasalt episodes and acidification – A paired catchment study in western Norway

As acid deposition has declined during the past 15–20 years in western Norway, afforestation and episodic seasalt deposition have become factors of increasing importance in explaining the mobilization of toxic aluminum (Al n+ ) to rivers and lakes. We conducted a paired catchment at four sites in western Norway across a gradient in acid deposition to evaluate the importance of afforestation and seasalt episodes. Streamwater was sampled intensively before, during and after seasalt episodes over a three-year period. A seasalt episode in January 2003 caused considerable impact on the streamwater chemistry. pH dropped and concentrations of Al n+ increased due to cation exchange of Na + ions for H + and Al n+ in the soil. The response was larger in streams draining the catchments which receive high acid deposition and in those afforested with spruce as compared with adjacent catchments in native birch. The results indicate that acid pulses induced by episodic inputs of seasalts are exacerbated by land use change from native birch to planted spruce.

SUSPENDED SEDIMENT DYNAMICS IN SMALL FOREST STREAMS OF THE PACIFIC NORTHWEST

This paper reviews suspended sediment sources and transport in small forest streams in the Pacific Northwest region of North America, particularly in relation to riparian management. Mass movements, roading and yarding practices, and burning can increase the supply of suspended sediment. Sediment yields recovered to pre-harvest levels within one to six years in several paired catchment studies. However, delayed mass movements related to roads and harvesting may produce elevated suspended sediment yield one or more decades after logging. There is mixed evidence for the role of streamside tree throw in riparian buffers in supplying sediment to streams. Harvesting within the riparian zone may not increase suspended sediment yield if near stream soils are not disturbed. Key knowledge gaps relate to the relative roles of increased transport capacity versus sediment supply, the dynamics of fine sediment penetration into bed sediments, and the effects of forest harvesting on suspended sediment at different scales. Future research should involve nested catchments to examine suspended sediment response to forest practices at multiple spatial scales, in combination with process-based field studies.

A review of paired catchment studies for determining changes in water yield resulting from alterations in vegetation

Paired catchment studies have been widely used as a means of determining the magnitude of water yield changes resulting from changes in vegetation. This review focuses on the use of paired catchment studies for determining the changes in water yield at various time scales resulting from permanent changes in vegetation. The review considers long term annual changes, adjustment time scales, the seasonal pattern of flows and changes in both annual and seasonal flow duration curves. The paired catchment studies reported in the literature have been divided into four broad categories: afforestation experiments, deforestation experiments, regrowth experiments and forest conversion experiments. Comparisons between paired catchment results and a mean annual water balance model are presented and show good agreement between the two methodologies. The results highlight the potential underestimation of water yield changes if regrowth experiments are used to predict the likely impact of permanent alterations to a catchment's vegetation. An analysis of annual water yield changes from afforestation, deforestation and regrowth experiments demonstrates that the time taken to reach a new equilibrium under permanent land use change varies considerably. Deforestation experiments reach a new equilibrium more quickly than afforestation experiments. The review of papers reporting seasonal changes in water yield highlights the proportionally larger impact on low flows. Flow duration curve comparison provides a potential means of gaining a greater understanding of the impact of vegetation on the distribution of daily flows.

Measurement of post-fire hillslope erosion to evaluate and model rehabilitation treatment effectiveness and recovery

The increasing size and severity of wildfires in the western United States has caused a corresponding increase in post-fire emergency erosion control activities. Hillslope treatments, such as broadcast seeding, mulching and installed barriers, are applied to reduce runoff and erosion, as well as downslope sedimentation. However, there are few data to determine if these post-fire treatments are practical and effective. Direct measurement of hillslope erosion, particularly in the remote settings where wildfires occur, is time consuming and costly. Rainfall simulation, sediment fences and paired catchment studies have been adapted for measuring post-fire erosion in the mountainous forest regions of the western USA. The use of paired catchments to measure hillslope erosion and evaluate treatment effectiveness is illustrated by an ongoing experiment of six contour-felled log erosion barrier research sites. Deciding which type of treatments to use, as well as the locations and timing of application, requires treatment cost and effectiveness to be weighed against potential damage from unmitigated erosion. To assist in this process, a web-based Erosion Risk Management Tool has been developed that incorporates variability in rainfall, burn severity and soil properties, as well as treatment options to provide probabilistic erosion estimates for 4 years after a fire.

Effects of atmospheric N deposition on coarse organic matter in a headwater stream

Input, storage, export potential, and system-level processing of coarse organic matter were investigated in the intermittent streams that drain the Bear Brook Watershed in Maine (BBWM). BBWM is a paired catchment study investigating ecosystem effects of atmospheric N and S deposition. We predicted that the increased N loading to the treatment catchment would elevate input of organic matter, result in higher levels of coarse organic matter biomass, and increase litter processing rates in the treatment stream relative to the reference stream. We found that the streams draining BBWM did not have statistically different coarse organic matter input, biomass, or processing rates and we found only modest differences in export potential. System-level processing rates for maple (Acer spp.) litter were similar to rates previously quantified using litterbag methods. However, system-level processing rates for American beech (Fagus grandifolia) litter were an order of magnitude faster than rates measured with litterbags. This difference was likely due to movements of these leaves from riffle/runs and pools into debris dams, rather than differences in measurements of leaf tissue processing rates between methods. Organic matter dynamics of the intermittent streams at BBWM were similar to other forested, headwater streams. Our results indicate that the long-term N manipulation experiment at BBWM has not altered input, storage or processing of coarse organic matter in the treatment stream. Physical characteristics of these stream ecosystems appear to regulate organic matter dynamics rather than differences in nutrient chemistry.

Summer flows in experimental catchments with different forest covers, Chile

Runoff and peak flows in four experimental catchments with different land uses are analyzed for summer periods. The catchments have a rainy temperate climate with annual precipitations between 2000 and 2500 mm, 70% of which is concentrated in the winter period between May and August. The final harvest of the forest plantation in one of these catchments generated increases in summer runoff. Also, differences between the maximum instantaneous discharge and the flow at the beginning of the storm then almost duplicated those registered in rainfall events of similar magnitude when the catchment was fully forested. Runoff analysis in this catchment is difficult because the two post-harvesting summer periods are much wetter than the two pre-harvesting ones but a double mass analysis shows the effect of harvesting clearly. In a paired catchment study, low cover in one of the two neighbour catchments explains higher direct runoff and base flows although lower maximum instantaneous specific discharge occurred in the less vegetated but larger catchment. Low vegetation cover explains increases in summer flows, although the size, topography, rainfall conditions, road density, extent of affected area and runoff generation processes play an important role in the hydrological effects of different land uses.

Effect of a whole-catchment N addition on stream detritus processing

The Bear Brook Watershed in Maine (BBWM) is a paired catchment study investigating ecosystem effects of N and S deposition. Because of the decade long (NH4)2SO4 addition, the treatment catchment has higher stream NO3− and enriched foliar N concentrations compared to the reference catchment. We investigated how both stream N and foliar N affect stream detritus processing. Differences in litter processing were assessed by measuring mass loss, tissue softness, and shredder biomass. To examine both stream effects and leaf source, N-enriched and reference litter bags were prepared for 3 leaf species and placed in each stream. Red maple leaves were examined in 1997, 1998, and 1999. Sugar maple and American beech were examined in 1999. In all years, the only stream effects were increased mass loss for sugar maple and higher shredder biomass for red maple in 1998 in the treatment stream. Several leaf source effects were observed. N-enriched leaves of sugar maple, American beech, and red maple in 1998 had significantly higher microbial activity, as indicated by softer tissue, and had higher total mass loss. Further, shredder biomass tended to be highest in N-enriched litter bags. Although significant effects were detected, our results suggest that elevated dissolved N concentrations as a result of N deposition play a minimal role in regulating stream detritus processing at BBWM. Increased foliar N, however, did influence rates of stream detritus processing by increasing microbial activity, and possibly increasing shredder biomass. Our study shows how N deposited on entire catchments can affect litter processing in stream ecosystems.

Parameter uncertainty and the significance of simulated land use change effects

Uncertainty in parameters characterising different land covers leads to uncertainty in model predictions of land use change effects. In this study, a new approach is presented which allows a model to be assessed to see whether it is suitable for investigating land use change scenarios in the sense that different land covers can be significantly distinguished in their effects on model output. It consists of the following steps: (a) The uncertainty in land cover-dependent parameters is quantified. (b) The model of an artificial catchment with representative characteristics and uniform land cover is established. (c) Using this artificial catchment, Monte Carlo simulations are carried out to determine the uncertainty in the model response to different land covers. (d) By comparing the results for two covers, respectively, a dimensionless test statistic, the distinction level, is calculated. The distinction level is a normalised probability that two independent realisations of land covers which are parameterised within their range of natural uncertainty will yield distinct model responses. If the distinction level is greater than or equal to 90%, the land covers are assumed to have a significantly different effect on the model output. An example of the application of the new method is provided using the eco-hydrologic model SWAT-G. The land covers forest, pasture and arable land can be significantly distinguished by their long-term means of surface runoff, groundwater recharge and streamflow. The minimum proportion of the catchment area on which land cover must change in order to obtain significantly distinct model responses depends on the land covers involved and the considered hydrologic variable. In the case of a change between pasture and forest and with regard to average streamflow, this minimum proportion amounts to about 25%, a value that compares well with the results of paired catchment studies.

The long-term control of vegetation and woody debris on channel and flood-plain evolution: insights from a paired catchment study in southeastern Australia

Numerous case studies have demonstrated that alluvial and semi-alluvial rivers in SE Australia have undergone dramatic metamorphosis in historical times. However, very few studies place these changes within a long-term evolutionary context. As a consequence, the magnitude of, and ultimate controls on, the changes to river form and processes are not fully appreciated. In this study, a paired catchment analysis is undertaken between two moderate-sized sand-bed rivers in East Gippsland, Australia. From the Thurra River, direct insight is gained into the predisturbance control exerted by riparian vegetation and wood in a lowland alluvial river. This river is effectively in the same condition today as it was at the time of the arrival of Europeans in Australia. In contrast, the adjacent Cann River, which has been settled by Europeans for 150 years, but was previously very similar to the Thurra River, exhibits stark differences today. Channel morphodynamics observed within the Thurra River, when coupled with historical and geomorphic evidence for the former condition of the Cann River, provide a detailed reference by which the recent changes to the Cann River are measured. Chronostratigraphic evidence from both flood plains places recent channel behavior within an evolutionary context extending well into the Pleistocene. Since European settlement, the study reach of the Cann River has experienced a 360% increase in channel depth, a 240% increase in channel slope, a 700% increase in channel capacity, and up to a 150-fold increase in the rate of lateral channel migration. The contemporary condition of the Cann River channel differs profoundly to that which has prevailed over at least the previous 27 ka. The first-order control on the historical channel metamorphosis is the removal of riparian vegetation and woody debris (WD). Numerous thresholds have been crossed as a result of historical channel changes, particularly the relationship between average length of woody debris pieces and channel width. Consequently, channel recovery will not simply be achieved by reintroducing the pre-existing riparian vegetation and woody debris load. Devising sound management strategies and realistic management targets for the Cann River, or any similarly altered river, requires an understanding of the long-term channel and flood plain evolutionary history, as well as the underlying controls on historical changes.

Impact of forest ditching on nutrient loadings of a small stream—a paired catchment study in Kronoby, W. Finland

The effect of ditching of boreal forest land (W. Finland) on P and N dynamics and export was assessed using a paired catchment approach (response stream+control stream) for a 4-year study period. After the ditching operations, there was a small but significant increase in the NH 4 + export, but no significant changes in the concentrations and loads of total reactive P, total non-reactive P, total organic N and NO 3 −. In both streams, the reactive P fraction largely followed the Fe concentrations. High input to output ratios, and thus, high retention was a typical feature of the inorganic N compounds. Total organic N and the TOC/TON ratio were positively correlated with the total organic carbon.

Effects on catchment water balance from the management of Pinus plantations on the coastal lowlands of south‐east Queensland, Australia

AbstractA paired catchment study was conducted over a 10‐year period on the hydrology of an exotic Pinus plantation in the coastal lowlands of south‐east Queensland, Australia. Each catchment was instrumented with a stream monitoring station, tipping bucket rain gauge, and a network of piezometers to monitor the shallow perched water table. After a 6‐year calibration period a harvest treatment was imposed on one catchment (Review): clearfelling approximately 90% of the catchment area, which contained a mature (44‐year‐old) Pinus elliottii plantation. This subsequently was re‐established with a second rotation plantation of a hybrid of P.elliottii × P.caribaea var. hondurensis. The control catchment (Crayfish) contained a P. elliottii plantation similar to that clearfelled at Review. The post‐harvest period was monitored for a further 4 years.Evapotranspiration was found to be the major output flux, with stream flow only a minor component of the study catchments' water budget. Areas with gleyed podzolic soils were found to remain waterlogged for periods up to 7 months following the seasonally high summer rainfall period, with other soils having water logging periods of only a few months. Little change was observed in waterlogging characteristics following harvesting, in contrast to stream flow which increased for 3 years but with evidence of a decline after 5 years.The perched‐watertable piezometric surface indicated that its lateral drainage differs from that of surface flows; the perched‐watertable drainage fluxes from the study catchments seem minor and off‐site movement of solutes (e.g. nutrients and contaminants) via them would be limited. It was identified that the interaction between the perched and deep aquifers was poorly understood, as well as the relative importance of the soil moisture storage of the aquitard clay layer between them. It was concluded that there is a need to undertake more detailed analysis using modelling, and to obtain additional field data on soil‐layer properties and piezometric levels of the deep aquifer. Copyright © 2002 John Wiley &amp; Sons, Ltd.

On the future of forest hydrology and biogeochemistry

Correspondence to: Jeff McDonnell, Oregon State University, USA. This comment arises from a USA-Japan Seminar on the Hydrology and Biogeochemistry of Forested Catchments, funded by the US National Science Foundation, Japan Society for the Promotion of Science (JSPS), International Geosphere-Biosphere Program (IGBP). The meeting was hosted by Jeff McDonnell and Tadashi Tanaka in February 2000 at the East West Center in Hawaii. This meeting took place about ten years after the last such USA-Japan hydrology seminar (see details in McDonnell et al., this issue). The accompanying special issue is a distillation of papers from that meeting. These papers represent a cross section of work from both countries across the physical-biogeochemical interface. It is just over 30 years since Hewlett and Hibbert’s seminal work on the variable source area was published. As we enter the 21st century, the traditional domain of forest hydrologists is changing: from paired catchment studies examining changes in water balance associated with afforestation or deforestation, to issues of physical controls on stream biogeochemistry, process-based modelling, tracing water flowpath, residence time and mixing modeling. Nevertheless, while the orientation of forest hydrology has changed a great deal in the past 30 years, the basic concepts have changed only slightly. More detailed process representations are being provided for more catchments. Yet, rather than synthesis of common findings across catchments (that might reveal common controlling processes), much of the science is directed at studying the idiosyncrasies of “yet another catchment”. Particular processes such as macropore flow, capillary fringe-induced groundwater ridging, rapid effusion of old water from hillslopes, etc. have occupied many journal pages in recent decades, but conceptual advancement beyond variable source area theory has been limited. The discussions at the workshop began with an examination of the following questions:

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.