Thermal regime metrics and quantifying their uncertainty for North American streams

Understanding and characterizing thermal regimes is gaining popularity, but there has been little assessment of the sources and magnitudes of uncertainty among different thermal metrics. Understanding how the quantity of data influence estimates of metrics and the characterization of thermal regime is critical to resource management. We examine the influence of record length on the uncertainty of estimation for commonly used thermal metrics including mean annual maximum and minimum, timing of the annual maximum and minimum, mean annual temperature range, mean weekly maximum temperature, July maximum, minimum, and range. We selected 19 sites from U.S. Geological Survey hydrometric station network to represent stations with both small and large drainage areas across the ecoregions of the contiguous United States with at least 20 years of daily stream temperature data. We also selected 54 sites from Water Survey of Canada's hydrometric network with at least 7 years of sub‐daily data for the province of Ontario. Randomizing a progressively increasing set of years used to calculate estimates of each metric provided the percentile confidence bands that were compared with various thresholds of acceptable certainty. Bootstrap confidence bands quickly decreased in width with increasing record length and approached an acceptable level at an average of 12 years for daily data metrics. Metrics calculated using the sub‐daily data required approximately 3 years of data. The timing of annual minimum and maximum temperatures required the greatest amount of data to reduce bias to an acceptable level.     

Building Hydrologic Foundations for Applications of ELOHA: How Long A Record Should You Have?

The Ecological Limits of Hydrologic Alteration framework for making regional assessments of environmental flows requires a ‘hydrologic foundation’ of flow data for current and undeveloped conditions. This raises the question how long a record is needed for an adequate hydrologic foundation? The answer depends on the variance in the flow record and on how much uncertainty is tolerable in metrics developed from the flow data. Copyright © 2017 John Wiley & Sons, Ltd.     

Predicting the natural flow regime: models for assessing hydrological alteration in streams

Understanding the extent to which natural streamflow characteristics have been altered is an important consideration for ecological assessments of streams. Assessing hydrologic condition requires that we quantify the attributes of the flow regime that would be expected in the absence of anthropogenic modifications. The objective of this study was to evaluate whether selected streamflow characteristics could be predicted at regional and national scales using geospatial data. Long‐term, gaged river basins distributed throughout the contiguous US that had streamflow characteristics representing least disturbed or near pristine conditions were identified. Thirteen metrics of the magnitude, frequency, duration, timing and rate of change of streamflow were calculated using a 20–50 year period of record for each site. We used random forests (RF), a robust statistical modelling approach, to develop models that predicted the value for each streamflow metric using natural watershed characteristics. We compared the performance (i.e. bias and precision) of national‐ and regional‐scale predictive models to that of models based on landscape classifications, including major river basins, ecoregions and hydrologic landscape regions (HLR). For all hydrologic metrics, landscape stratification models produced estimates that were less biased and more precise than a null model that accounted for no natural variability. Predictive models at the national and regional scale performed equally well, and substantially improved predictions of all hydrologic metrics relative to landscape stratification models. Prediction error rates ranged from 15 to 40%, but were ≤25% for most metrics. We selected three gaged, non‐reference sites to illustrate how predictive models could be used to assess hydrologic condition. These examples show how the models accurately estimate pre‐disturbance conditions and are sensitive to changes in streamflow variability associated with long‐term land‐use change. We also demonstrate how the models can be applied to predict expected natural flow characteristics at ungaged sites. Copyright © 2009 John Wiley & Sons, Ltd.     

HYDROLOGIC MODIFICATION FROM HYDROELECTRIC POWER OPERATIONS IN A MOUNTAIN BASIN

The natural flow paradigm suggests that components of the natural streamflow regime and variability should be managed to maintain important ecosystem functions and services. Mountain rivers can exhibit extreme flow variability and provide critical aquatic habitat and ecosystem services but can be severely impacted by hydroelectric power (HEP) development and operations that will likely increase in the future. The hydrologic modification from HEP operations in the Upper San Joaquin River Basin, California, was evaluated across 15 river and stream locations throughout the basin. Flow modifications in Bear Creek, an otherwise unimpacted high‐elevation subbasin, were evaluated in detail using a number of hydrologic metrics, including Indicators of Hydrologic Alteration (IHA), environmental flow components of IHA, flow duration curves, ecodeficit and comparisons using equivalence testing. The uncertainty of the metrics based on confidence intervals was also evaluated for unimpaired (upstream) and existing (impaired or downstream) conditions. Results showed that metrics for median values changed considerably for most locations under impaired conditions, but the direction and extent of change varied depending on the location and flow metric. Metrics for variability (coefficients of dispersion) changed even more. Most metrics showed that flow modifications in Bear Creek were substantial, including decreases in high flows and increases in most low‐flow metrics. However, some flow variability metrics increased because of large flood flows during several years overwhelming and bypassing the dam/diversion structure. Uncertainty in metrics varied considerably throughout the basin but generally increased for impaired conditions. Uncertainty should be explicitly considered when evaluating hydrologic modification from HEP in mountain watersheds. A number of metrics should be used depending on objectives and spatial scale, including a subset of key IHA metrics across multiple sites and other methods to provide detailed information on flow modification in conjunction with other environmental flow assessment techniques at key locations. Copyright © 2011 John Wiley & Sons, Ltd.     

Role of discharge and temperature variation in determining invertebrate community structure in a regulated river

This paper quantifies patterns of discharge and temperature variation in the regulated river Lyon and the adjacent, unregulated river Lochay (Scotland) and assesses the importance of these patterns for benthic invertebrate community structure. Invertebrates were sampled at sites in each catchment in autumn, winter and spring during the 2002–2003 hydrological year. Metrics were used to characterize the discharge and temperature regimes in the period immediately preceding invertebrate sample collection. Metric values were then used in a canonical correspondence analysis (CCA) of the invertebrate sample data, in order to assess the significance of individual metrics and the overall importance of flow and temperature variability for community structure. The variance in the invertebrate data explained by this CCA was compared to that from a CCA using a range of environmental data from the sites (stream‐bed algal cover, channel hydraulic, sedimentary and water quality characteristics). This comparison allowed assessment of the relative importance of environmental variables versus hydrologic and thermal regimes. Invertebrate communities in the Lyon were relatively poor and uneven, with Ephemeroptera, Plecoptera and Coleoptera poorly represented. Distinct site and seasonal clusters were evident in the CCA ordination biplots, with Lyon and Lochay sites separated in dimensions represented by geometric mean sediment size, water temperature and algal cover. The cumulative variance values from ordinations using the discharge and temperature metrics were consistently highest, suggesting that differences in invertebrate communities showed a stronger relation to patterns of discharge and temperature variability than to the broader suite of environmental conditions. Although there were marked thermal differences between sites, temperature metrics appeared no more important than discharge metrics in explaining differences in invertebrate community structure. A number of the temperature and discharge metrics appeared similarly important, suggesting that no one aspect of the hydrothermal regime was any more important than others in helping to understand differences in invertebrate community between the study sites. Copyright © 2007 John Wiley & Sons, Ltd.     

Application of wavelet analysis for monitoring the hydrologic effects of dam operation: Glen Canyon Dam and the Colorado River at Lees Ferry,Arizona

Wavelet analysis is a powerful tool with which to analyse the hydrologic effects of dam construction and operation on river systems. Using continuous records of instantaneous discharge from the Lees Ferry gauging station and records of daily mean discharge from upstream tributaries, we conducted wavelet analyses of the hydrologic structure of the Colorado River in Grand Canyon. The wavelet power spectrum (WPS) of daily mean discharge provided a highly compressed and integrative picture of the post‐dam elimination of pronounced annual and sub‐annual flow features. The WPS of the continuous record showed the influence of diurnal and weekly power generation cycles, shifts in discharge management, and the 1996 experimental flood in the post‐dam period. Normalization of the WPS by local wavelet spectra revealed the fine structure of modulation in discharge scale and amplitude and provides an extremely efficient tool with which to assess the relationships among hydrologic cycles and ecological and geomorphic systems. We extended our analysis to sections of the Snake River and showed how wavelet analysis can be used as a data mining technique. The wavelet approach is an especially promising tool with which to assess dam operation in less well‐studied regions and to evaluate management attempts to reconstruct desired flow characteristics. Copyright © 2005 John Wiley & Sons, Ltd.     

A Quantitative Framework to Derive Robust Characterization of Hydrological Gradients

If ecological management of river ecosystems is to keep pace with increasing pressure to abstract, divert and dam, we must develop general flow–ecology relationships to predict the impacts of these hydrologic alterations. Regional flow gradient analyses are a promising tool to quickly reveal these functional relationships, but there are considerable uncertainties in this method because of variability in the historical extent of flow data across different rivers, combined with multiple indices characterizing the ecological attributes of flow regimes. In response, we outline an objective framework for analysing spatial hydrologic gradients that addresses three major sources of uncertainty: robust estimation of flow indices, the potential for temporal trends to confound spatial variation in flow regimes and the statistical robustness to detect underlying hydrological gradients. The utility of our framework was examined in relation to flow regimes across multiple braided river catchments in Canterbury, New Zealand. We found that a subset of flow indices could be robustly estimated using only 10 years of flow data, although indices that captured flow ‘timing’ required longer time series. Temporal trends were unlikely to confound conclusions from a spatial hydrologic gradient analysis, and there were three statistically supported hydrologic gradients related to flow magnitude, flow variability and low flow events. The widespread application of robust spatial flow gradient analyses has the potential to further our understanding of how altered flow regimes affect the ecology of freshwater and riparian ecosystems, thereby providing the evidence base to inform river management. Copyright © 2016 John Wiley & Sons, Ltd.     

Changes in the flow regimes associated with climate change and human activities in the Yangtze River

Hydrologic regimes are increasingly altered under the impacts of climate change and human activities. Streamflow data from 1960 to 2014 were analysed to investigate changes in the flow regimes in the Yangtze River using multiple hydrologic metrics and the Budyko framework. The long‐term data were separated into two periods: the preimpact period (1960–2002) and the postimpact period (2003–2014), according to the year the Three Gorges Dam began operation. The results showed that both indicators of hydrologic alteration and ecoflow metrics were clearly altered. The highly changed indicators included flow in February, annual minimum 1‐, 3‐, 7‐, 30‐, and 90‐day flows, base flow index, date of annual minimum flow, and low pulse duration. The integrated degree of hydrologic alteration ranged from 41% to 61%, indicating a moderate alteration of the flow regimes in the Yangtze River. The regulation of the Three Gorges Dam increased low flow and weakened peak flow, which resulted in autumn ecodeficit and winter ecosurplus increasing dramatically since the 2000s. The ecoflow metrics were more sensitive to precipitation than to potential evapotranspiration. The joint effects of human activities and climate change varied among the river reaches in the different decades. The streamflow was mainly affected by human activities in the upper reach during the 1970s–1990s, with a contribution ratio ranging from 63% to 77%. Climate change shifted to a major contributor in the middle and lower reaches since the 1980s as well as in the upper reach in 2000–2014, accounting for 50–82% of the streamflow changes. These different responses were primarily caused by the variations of precipitation and intensive human activities, particularly the rapid growth of reservoirs and other large projects since the 1970s in the upper Yangtze River. These results provide interesting insights into the spatio‐temporal hydrologic alteration across the Yangtze River.     

CLASSIFICATION OF FLOW REGIMES FOR ENVIRONMENTAL FLOW ASSESSMENT IN REGULATED RIVERS: THE HUAI RIVER BASIN,CHINA

Flow regime characteristics (magnitude, frequency, duration, seasonal timing and rates of change) play a primary role in regulating the biodiversity and ecological processes in rivers. River classification provides the foundation for comparing the hydrologic regimes of rivers and development of hydro‐ecological relationships to inform environmental flow management and river restoration. This paper presents a classification of natural flow regimes and hydrologic changes due to dams and floodgates in the Huai River Basin, China, in preparation for an environmental flow assessment. The monthly natural flow regime of 45 stations in the upper and middle Huai River Basin were simulated for the period 1963–2000, based on the hydrological model SWAT (Soil and Water Assessment Tool). Six classes of flow patterns (low or high discharge, stable or variable, perennial or intermittent, predictable or unpredictable) were identified based on 80 hydrologic metrics, analysed by hierarchical clustering algorithms. The ecologically relevant climatic and geographic characteristics of these flow classes were tested for concordance with, and to strengthen, the hydro‐ecological classification. The regulation of natural flow patterns by dams and floodgates changed flows at some locations within each flow class and caused some gauges to shift into another class. The research reported here is expected to provide a foundation for development of hydro‐ecological relationships and environmental flow methods for wider use in China, as well as setting a new scientific direction for integrated river basin management in the Huai River Basin. Copyright © 2011 John Wiley & Sons, Ltd.     

Unrestricted ramping rates and long‐term trends in the food web metrics of a boreal river

Long‐term monitoring of the food web of a regulated hydropeaking river was conducted to assess if previously documented effects of changing ramping rates (RRs) were maintained with the addition of 6 years of data. Using carbon and nitrogen stable isotope analyses, we hypothesized that: (1) macroinvertebrates and fish inhabiting areas below peaking hydrodams would be higher in δ¹⁵N and lower in δ¹³C due to increased flow velocity and the influence of light respired dissolved inorganic carbon, relative to those sampled from areas with a natural flow regime; (2) the increase in δ¹⁵N of macroinvertebrates would lead a shorter food web length in the regulated river, but δ¹³C and niche width would be similar between the restricted and unrestricted RR periods (i.e., the BACI analysis); and (3) isotopic metrics (e.g., δ¹³C, δ¹⁵N, niche width [SEA_B], and food chain length [Δ¹⁵N]) would correlate with variations in flow characteristics through time. Consistent with previous analysis conducted over a shorter time period, a shift toward higher δ¹⁵N values was observed for both fish and invertebrates, but, contrarily, only invertebrates (not fish) had a lower δ¹³C value downstream of the dam. Over the long term, the before‐after‐control‐impact analysis found no effect of RRs on any of the food web metrics, implying that the change in operation did not affect the river food web. However, analysis of the time series data indicated that flow metrics and trophic metrics were often correlated, including a negative effect of RR (invertebrates) and discharge (fish) on food chain length. This study illustrates the difficulty in detecting changes in food web structure and function under changing flow regime influenced by natural and anthropogenic effects. As such, this study highlights the need for considering large spatial and temporal scales to differentiate between confounding effects of climate, natural variability, and altered flow regimes on food webs in regulated rivers.     

Trends in low flows of German rivers since 1950: Comparability of different low‐flow indicators and their spatial patterns

Climate change, land‐use shifts, reservoir storage, and water withdrawals impact low flows in rivers, creating challenges for ecological integrity and human uses. A systematic investigation of river discharges was carried out for 79 stream gauges in Germany. Available time series between 1950 and 2013 were analysed for trends in annual minimum low flows, discharge deficits, and low‐flow durations. The application of different low‐flow indicators led to similar spatial patterns, although each metric is used for different purposes in water management applications. Statistical tests identified significant discharge trends at more than half of the stations investigated. Low‐flow trends since 1950 tended to be catchment specific, suggesting that climate change has not been the dominant driver. Most of the gauges investigated showed statistically significant increases in low flows. This can be mainly attributed to reservoir management. For rivers showing snow‐ and icemelt‐dominated flow regimes, such trends are probably overlain by climate‐driven changes (increasing amounts of rainfall, earlier snowmelt in spring). In contrast, stations showing statistically significant decreases in low flows were correlated with areas of decreasing mining activity. Hydrologic impacts of climate change are widespread and significant, but the results here suggest that human river management remains the dominant hydrologic driver on many rivers.     

MULTISTAGE ANALYSIS OF HYDROLOGIC ALTERATIONS IN THE YELLOW RIVER,CHINA

Natural river flow regimes provide an array of ecological and social functions by sustaining the health of riverine ecosystems. To identify the hydrologic alterations in the lower Yellow River basin caused by natural factors and human activities, we developed multistage hydrologic analysis to investigate the temporal variability of the river's flow regimes. We used a cumulative departure curve and Mann–Whitney–Pettitt nonparametric tests to determine possible change points based on hydrologic data from 1950 to 2006. We then used the range of variability approach to characterize and to quantify the temporal variability of the hydrologic regimes that were associated with perturbations such as dam operation, flow diversions or intensive conversion of land use within the watershed. In the case study, three stages in hydrologic alterations of the flow regime were found: a stage without human impacts, a stage with excessive human impacts and a reservoir‐regulation stage. Our results indicated that (i) after 1997, dam operation efficiently achieved flood control using sediment regulation activities; (ii) although effective in flood control, the Xiaolangdi Reservoir could not handle situations with extremely low flow, such as during droughts; and (iii) under the arid climate of the Yellow River basin, water consumption by agriculture was the main cause of water shortages. The current study shows that multistage hydrologic analysis can greatly assist regional water resources management and the restoration of riparian eco‐environmental systems affected by dam construction under a changing environment. Copyright © 2012 John Wiley & Sons, Ltd.     

CLASSIFICATION OF NATURAL FLOW REGIMES IN THE EBRO BASIN (SPAIN) BY USING A WIDE RANGE OF HYDROLOGIC PARAMETERS

This paper presents a classification of different natural flow regimes found in Ebro basin, one of the largest in the Mediterranean region. Determination of flow regimes was based on multivariate analyses using long‐term discharge series of unaltered flow data. Mean monthly discharges of the 30 ‘best’ flow series and a total of 52 flow series containing unaltered flow data were selected to represent baseline flow conditions for tributaries throughout the basin. Metrics representing magnitude, duration and frequency components of flow were used to identify hydrologic differences across the basin. A total of six natural flow regimes were identified in the Ebro Basin, using a Ward cluster method. The flow patterns identified and their spatial distribution largely corresponded with climatic zones previously reported for the Ebro Basin, with regime types ranging from pluvio‐oceanic in the western part of the basin to Mediterranean in the eastern region. Geologic characteristics of the catchment and altitude of headwaters were also found to play an important role in defining flow regime type. A 19‐hydrologic variable subset was used to explain main hydrologic differences among groups (such as magnitude and frequency of extreme flow conditions or magnitude and variance of average flow conditions). However, stepwise discriminant analysis was not able to identify consistent subsets of hydrologic variables that adequately identified the six natural flow regime types in this basin. Canonical discriminant analysis was useful to understand class separation and for the interpretation of results. Copyright © 2012 John Wiley & Sons, Ltd.     

A CLIMATE‐BASED INDEX OF HISTORIC SPAWNING‐STREAM HYDROLOGY

Reproductive success of stream‐spawning Oncorhynchus fishes (Pacific salmon, rainbow trout, cutthroat trout and their allies) may be greatly affected by stream discharge or its covariate, stream temperature, during the spawning season. Because such data for the physical environment may not have been routinely collected as part of previous investigations of these fishes, identification of simple but robust indices of historic, seasonal stream discharge and temperature, using long‐term climate data sets, would be important, especially to investigations of historic population dynamics. This study examined statistical associations among several climate variables and the spawning‐season (approximately June) discharges and temperatures of Clear Creek, a Yellowstone Lake tributary used by spawning Yellowstone cutthroat trout, Oncorhynchus clarkii bouvieri (YCT), from the lake. Correlation analysis showed that total water‐year degree‐days (calculated on the basis of mean daily air temperature > 0°C) at Lake Village, on the lake's north shore, was a robust index (both negative and positive, respectively) of consecutive, total semi‐month metrics of creek discharge and temperature during the YCT spawning season. This study (and subsequent use of the Lake Village degree days metric as an environmental variable in a dynamic, age‐structured model of the lacustrine–adfluvial YCT population of Clear Creek) showed how exploratory analyses of the fragmentary but long‐term and regionally unique data sets for Clear Creek discharge and temperature revealed a simple but robust index of climate variation important to understanding the historic dynamics of Clear Creek's YCT population, which is a key spawning stock of Yellowstone Lake. In addition, the extensive statistical associations among the climate variables, along with the temporal trends in two key variables, broadly showed how climate varied across the Yellowstone Lake region during the past several decades. Those observations have implications for the historic, seasonal hydrology of all Yellowstone Lake tributaries used by spawning YCT. Copyright © 2011 John Wiley & Sons, Ltd.     

A functional flows approach to selecting ecologically relevant flow metrics for environmental flow applications

The science and practice of environmental flows have advanced significantly over the last several decades. Most environmental flow approaches require quantifying the relationships between hydrologic change and biologic response, but this can be challenging to determine and implement due to high data requirements, limited transferability, and the abundance of hydrologic metrics available for evaluation. We suggest that a functional flows approach, focusing on elements of the natural flow regime known to sustain important ecosystem processes, offers a pathway for linking understanding of ecosystem processes with discrete, quantifiable measures of the flow regime for a broad range of native taxa and assemblages. Functional flow components can be identified as distinct aspects of the annual hydrograph that support key biophysical processes, such as wet season flood flows or spring recession flows, and then quantified by flow metrics, such as 5% exceedance flow or daily percent decrease in flow, respectively. By selecting a discrete set of flow metrics that measure key functional flow components, the spatial and temporal complexity of flow regimes can be managed in a holistic manner supportive of multiple ecological processes and native aquatic species requirements. We provide an overview of the functional flows approach to selecting a defined set of flow metrics and illustrate its application in two seasonally variable stream systems. We further discuss how a functional flows approach can be utilized as a conceptual model both within and outside of existing environmental flow frameworks to guide consideration of ecological processes when designing prescribed flow regimes.     

Landscape and flow path-based nutrient loading metrics for evaluation of in-stream water quality in Saginaw Bay,Michigan

Landscape metrics are often used to model nonpoint source pollution from agricultural and urban surface runoff. By considering topography and the spatial arrangement of land cover, landscape metrics can better account for hydrologic connectivity, loading quantity, and vegetated buffer filtering between nutrient loading sources and streams. For this study we develop a surface runoff nutrient loading metric that considers source (i.e. cropland or developed) loading and buffer filtering along hydrologic transport vectors, or flow paths. We use General Additive Modeling to evaluate the relationship between this metric and in-stream nitrogen (N) and phosphorus (P) concentrations in the Saginaw Bay watershed in Michigan, US and compare the relative predictive power between this metric and other landscape metrics that do not consider hydrologic connectivity. The flow path-based cropland loading metric was a stronger predictor of in-stream NO₃ concentrations than alternative metrics. In-stream P concentrations were best predicted by models that included 48-h antecedent precipitation and catchment-wide proportion of developed landcover. Metric maps reveal high nutrient loading areas where only a small proportion of loading reaches streams via surface runoff, highlighting the need to consider nutrient loading via drainage tiles and other subsurface pathways in efforts to quantify nonpoint source loading to surface waters. The flow path-based loading metric is represented spatially as a gridded dataset showing estimates of nutrient loading adjacent to streams, and with higher resolution stream delineation data it could be used by land managers to target locations for precision buffer placement to intercept surface runoff and reduce nutrient loading.     

HYDROGEOMORPHIC CLASSIFICATION OF WASHINGTON STATE RIVERS TO SUPPORT EMERGING ENVIRONMENTAL FLOW MANAGEMENT STRATEGIES

As demand for fresh water increases in tandem with human population growth and a changing climate, the need to understand the ecological tradeoffs of flow regulation gains greater importance. Environmental classification is a first step towards quantifying these tradeoffs by creating the framework necessary for analysing the effects of flow variability on riverine biota. Our study presents a spatially explicit hydrogeomorphic classification of streams and rivers in Washington State, USA and investigates how projected climate change is likely to affect flow regimes in the future. We calculated 99 hydrologic metrics from 15 years of continuous daily discharge data for 64 gauges with negligible upstream impact, which were entered into a Bayesian mixture model to classify flow regimes into seven major classes described by their dominant flow source as follows: groundwater (GW), rainfall (RF), rain‐with‐snow (RS), snow‐and‐rain (SandR), snow‐with‐rain (SR), snowmelt (SM) and ultra‐snowmelt (US). The largest class sizes were represented by the transitional RS and SandR classes (14 and 12 gauges, respectively), which are ubiquitous in temperate, mountainous landscapes found in Washington. We used a recursive partitioning algorithm and random forests to predict flow class based on a suite of environmental and climate variables. Overall classification success was 75%, and the model was used to predict normative flow classes at the reach scale for the entire state. Application of future climate change scenarios to the model inputs indicated shifts of varying magnitude from snow‐dominated to rain‐dominated flow classes. Lastly, a geomorphic classification was developed using a digital elevation model (DEM) and climatic data to assign stream segments as either dominantly able or unable to migrate, which was cross‐tabulated with the flow types to produce a 14‐tier hydrogeomorphic classification. The hydrogeomorphic classification provides a framework upon which empirical flow alteration–ecological response relationships can subsequently be developed using ecological information collected throughout the region. Copyright © 2011 John Wiley & Sons, Ltd.     

Regional flood frequency analysis in the central‐western river basins of Argentina

Knowing the probability of occurrence of a flood event is an important issue for water resources planning. At‐site probability models require a long extension of hydrological data for robust estimation of low‐frequency events. As the mean record length of 25 gauge stations in western river basins of Argentina is 49 years (until 2010), regional models are an interesting tool to determine mountain rivers system dynamics. This study aims to estimate low‐frequency quantiles of annual maximum flow in Argentinean western river basins (28°S–37°S) applying regional frequency analysis based on the L‐moments method. Besides, mean annual maximum flow of 75 gauge stations (22°S–52°S) was analysed. First, an exploratory data analysis was performed; normality, independence, and randomness were accepted in the 27%, 87%, and 91% of cases, respectively. Increasing trends in annual maximum flows in the north‐western and central‐western rivers of Argentina were detected, whereas decreasing trends in annual maximum flow in the Patagonian Andes were identified. Base on at‐site characteristics and at‐site statistics, a homogeneous region of 12 stations with a record period of 568 years was formed. General extreme value was the most appropriate distribution for this homogeneous study region. Estimation accuracy using Monte Carlo simulations was performed. The error bounds were set at 90%, the mean square error was 9.23%, and the relative bias was 1.6%. The regional method performed better than the at‐site estimation.