Search

Alpine glaciers and perennial snow fields are important hydrologic elements in many mountain environments providing runoff during the late summer and during periods of drought. Because relatively long records of glacier mass-balance data are absent from many glacierized catchments, it remains unclear to what extent shrinking perennial snow and glaciers have affected runoff trends from these watersheds. Here we employ a hydrograph separation technique that uses a double mass curve in an attempt to isolate changes in runoff due to glacier retreat and disappearance of perennial snow. The method is tested using hydrometric data from 20 glacierized and 16 non-glacierized catchments in the Columbia Basin of Canada. The resulting estimates on cryosphere storage contribution to streamflow were well correlated to other regional estimates based on measurements as well as empirical and mechanistic models. Annual cryosphere runoff changed from +19 to -55% during the period 1975-2012, with an average decline of 26%. For August runoff, these changes ranged from +17 to -66%, with an average decrease of 24%. Reduction of cryosphere contributions to annual and late summer flows are expected to continue in coming decades as glaciers and the perennial snow patches shrink. Our method to isolate changes in late summer cryospheric storage contributions can be used as a first order estimate on changes in glacier contributions to flow and may help researchers and water managers target watersheds for further analysis.

Flow regime refers to five elements of streamflow including flow magnitude, frequency, timing, duration and change rate. In spite of wide recognition of its critical significance in aquatic functions and ecosystem integrity, its responses to forest or land use change are rarely and quantitatively assessed. This paper used the Meijiang watershed (6983.2 km2), situated in the upper reach of the Poyang Lake basin, as an example to first demonstrate how flow regimes were altered by deforestation and then show if the altered flow regimes were possibly recovered by consequent reforestation. Two breakpoints (year 1968 and year 1985) with significant annual streamflow changes were detected, and they were then used to define three distinct periods including the reference or control period (1957 to 1967), deforestation (1968 to 1984) and reforestation (1985–2006). The paired year approach was then applied to quantitatively analyse the responses of flow regimes to forest cover changes. Both high flows (daily flows ≧ Q5%) and low flows (daily flows ≦ Q95%) were assessed. For high flows, the deforestation significantly increased the averaged magnitudes by 10.4%, increased the return periods (5–10 year) by 23.4%, advanced the averaged timings by 10.7 h and extended the averaged durations over the thresholds by 4 days. In contrast, reforestation delayed the averaged timings by 10.5 h, reduced the averaged duration by 5 days and decreased the averaged magnitude by 17.9%. Regarding low flows, the deforestation decreased the flow magnitudes by 30.1% with 30.5 day advancing in the average timings. To our surprise, however, low flows were not significantly changed by reforestation. All above results clearly demonstrate that flow regimes of both high and low flows were significantly altered by deforestation, and recovery of those alterations through reforestation may take much longer time than expected probably because of severe soil erosion and resultant loss of soil infiltration capacity after deforestation. Copyright © 2015 John Wiley & Sons, Ltd.

Two large neighbouring watersheds, the Bowron (3420 km2) and Willow (2860 km2) situated in the central interior of British Columbia, Canada, were used to compare their hydrological responses to forest harvesting in snow-dominant environment. Both watersheds had experienced significant, comparative forest harvesting level. The long-term hydrometric and timber harvesting data (>50 years of records) were analysed using time series analysis to examine the hydrological impacts of forest harvesting. The hydrological variables including mean, peak and low flows over annual and seasonal scales (spring snowmelt, summer rain and winter base flow) were tested separately. Results showed that forest harvesting in the Willow watershed significantly increased annual and spring mean flows as well as annual and spring peak flows, whereas it caused an insignificant change on those hydrological variables in the Bowron watershed. The contrasted differences in hydrological responses are due to the differences in topography, spatial heterogeneity, forest harvesting characteristics and climate between two watersheds. The relative uniform topography and climate in the Willow watershed may promote hydrological synchronization effects, whereas larger variation in elevations, together with forest harvesting that occurred at lower elevations, may cause hydrological de-synchronization effect in the Bowron watershed. The contrasted results demonstrate that the effects of forest harvesting on hydrology in large watersheds are likely watershed specific, and any attempt to generalize hydrological responses to forest harvesting must be carried out with caution. A landscape ecological perspective is critically needed for future forest hydrology studies, particularly for large watersheds. Copyright © 2013 John Wiley & Sons, Ltd.

The hydrological effect of forest recovery is receiving renewed interest globally because information on forest carbon–water relationship is critically needed to support carbon management through reforestation and sustainable water management. In Northeastern China, summer (June to August) streamflow accounts for about 50% of total annual streamflow and is vital to water supply and management in the region. Understanding how forest recovery may affect streamflow is important to both reforestation campaign and long-term water sustainability. In this study, we analysed 33 years of summer hydrologic data (1970–2002) from two comparable small-scale watersheds located in the Xiaoxing'anling, Northeastern China. Time series analysis and two graphic methods (double mass curve and flow duration curve) with statistical testing as well as long-term data on forest cover changes and climate were used. Our results show that the significant streamflow reduction as a result of reforestation occurred when forest cover reached 70% or 10 years after planting. After forest cover reached 85%, water reduction became stabilized. The accumulative streamflow reduction in 2002 reached 8·61% of the total accumulative streamflow. Among those water reduced, high flows (from 5 to 25 percentiles) were mostly affected, demonstrating that northeastern forests have an important role in reducing high flows. Implications of these results are discussed in the context of climate change, reforestation and water resource management. Copyright © 2011 John Wiley & Sons, Ltd.

The Integrated Biosphere Simulator is used to evaluate the spatial and temporal patterns of the crucial hydrological variables [run-off and actual evapotranspiration (AET)] of the water balance across China for the period 1951–2006 including a precipitation analysis. Results suggest three major findings. First, simulated run-off captured 85% of the spatial variability and 80% of the temporal variability for 85 hydrological gauges across China. The mean relative errors were within 20% for 66% of the studied stations and within 30% for 86% of the stations. The Nash–Sutcliffe coefficients indicated that the quantity pattern of run-off was also captured acceptably except for some watersheds in southwestern and northwestern China. The possible reasons for underestimation of run-off in the Tibetan plateau include underestimation of precipitation and uncertainties in other meteorological data due to complex topography, and simplified representations of the soil depth attribute and snow processes in the model. Second, simulated AET matched reasonably with estimated values calculated as the residual of precipitation and run-off for watersheds controlled by the hydrological gauges. Finally, trend analysis based on the Mann–Kendall method indicated that significant increasing and decreasing patterns in precipitation appeared in the northwest part of China and the Yellow River region, respectively. Significant increasing and decreasing trends in AET were detected in the Southwest region and the Yangtze River region, respectively. In addition, the Southwest region, northern China (including the Heilongjiang, Liaohe, and Haihe Basins), and the Yellow River Basin showed significant decreasing trends in run-off, and the Zhemin hydrological region showed a significant increasing trend. Copyright © 2009 John Wiley & Sons, Ltd.

Overexploitation of forest resources in China has caused serious concerns over its negative impacts on water resources, biodiversity, soil erosion, wildlife habitat and community stability. One key concern is the impact of forestry practices on hydrological processes, particularly the effect of forest harvest on water quality and quantity. Since the mid 1980s, a series of scientific studies on forest hydrology have been initiated in major types of forest across the country, including Korean pine (Pinus koraiensis), Chinese fir (Cunninghamia lanceolata), oak (Quercus mongolica), larch (Larix gmelinii), faber fir (Abies fabri), Chinese pine (Pinus tabulaeformis), armand pine (Pinus arandi), birch (Betula platyphylla) and some tropical forests. These studies measured rainfall interception, streamflow, evapotranspiration and impacts of forest management (clearcutting and reforestation). This paper reviews key findings from these forest hydrological studies conducted over the past 20 years in China. Forest canopy interception rates varied from 15 to 30% of total rainfall, depending on forest canopy and rainfall characteristics. Stemflow is generally a small percentage (