Propagating uncertainty from catchment experiments to estimates of streamflow reduction by invasive alien plants in southwestern South Africa

Long-term catchment experiments from South Africa have demonstrated that afforestation of grasslands and shrublands significantly reduces surface-water runoff. These results have guided the country's forestry policy and the implementation of a national Invasive Alien Plant (IAP) control programme for the past few decades. Unfortunately, woody IAP densities continue to increase, compounding existing threats to water security from population growth and climatic change. Decision makers need defensible estimates of the impacts of afforestation or invasions on runoff to weigh up alternative land use options, or guide investment of limited resources into ecosystem restoration through IAP clearing versus engineering-based water-augmentation schemes. Existing attempts to extrapolate the impacts observed in catchment afforestation experiments to broad-scale IAP impacts give no indication of uncertainty. Globally, the uncertainty inherent in the results from paired-catchment experiments is seldom propagated into subsequent analyses making use of these data. We present a fully reproducible Bayesian model that propagates uncertainty from input data to final estimates of changes in streamflow when extrapolating from catchment experiments to broader landscapes. We apply our model to South Africa's catchment experiment data, estimating streamflow losses to plantations and analogous plant invasions in the catchments of southwestern South Africa, including uncertainty. We estimate that regional streamflow is reduced by 304 million m³ or 4.14% annually as a result of IAPs, with an upper estimate of 408 million m³ (5.54%) and a lower estimate of 267 million m³ (3.63%). Our model quantifies uncertainty associated with all parameters and their contribution to overall uncertainty, helping guide future research needs. Acknowledging and quantifying inherent uncertainty enables more defensible decisions regarding water resource management.     

A soft hydrological monitoring approach for comparing runoff on a network of small poorly gauged catchments

Catchments in many parts of the world are either ungauged or poorly gauged, and the dominant processes governing their streamflow response are still poorly understood. The analysis of runoff coefficients provides essential insight into catchment response, particularly if both range of catchments and a range of events are compared. This paper investigates how well the hydrological runoff of 11 small, poorly gauged catchments with ephemeral streams (0·1‐0·6 km²) can be compared using estimated runoff with the associated uncertainty. Data of rainfall and water depth at a catchment's outlet were recorded using automatic logging equipment during 2008‐2009. The hydrological regime is intermittent and the annual precipitation ranged between 569 and 727 mm. Discharge was estimated using Manning's equation and channel cross‐section measurements. Innovative work has been performed under controlled experimental conditions to estimate Manning's coefficient values for the different cover types observed in studied streams: non‐aquatic vegetations (giant reed, bramble and thistle), grass and coarse granular deposits. The results show that estimates derived using roughness coefficients differ from those previously established for larger streams with aquatic vegetation. Catchment runoff was compared at both the event and the annual scale. The results indicate significant variability between the catchment's responses. This variability allows for classification in spite of all the uncertainty associated with runoff estimation. This study highlights the potential of using a network of poorly gauged catch ments. From almost no catchment understanding the proposed methodology allows to compare poorly gauged catchments and highlights similarity/dissimilarity between catchment responses. Copyright © 2011 John Wiley & Sons, Ltd.     

A joint analysis of river runoff and meteorological forcing in the Karakoram,upper Indus Basin

Satellite‐geodetic altimetry investigations in the Karakoram have indicated slight mass gain or loss of the glaciers during the early part of 21st century. Equivalent discharge in the upper Indus Basin due to these mass changes has been estimated at 5 to 10% of mean annual flow. However, satellite altimetry and geodetic glacier mass estimates in the extreme topography of the Karakoram have not yet been counter‐validated by hydrological analysis. Therefore, we present a first cross validation of three to five decades of river flow data from the three major watersheds in the Karakoram, with matching series of monthly precipitation, temperature, and evaporation provided by six atmospheric reanalysis products for 1979–2014. The analyses suggest that in most cases river flows have been increasing steadily from the end of the 1960s and 1970s to the middle of the 1990s and have stabilized or are in decline since then. Hunza watershed in Karakoram West shows consistently declining flows over the first half of the analysis period and stable flows during the second half for most of the summer melting season, suggesting mass accumulation. Rising river flows in the Shyok and Shigar watersheds, followed by stabilizing or slightly declining flows from 1995 onward, can be explained by consistently increasing precipitation during the first half of the analysis period, and successive stabilization or minor decline thereof. Flow data do not necessarily suggest considerable loss or gain of glacial mass in the Karakoram during the late 90s and early 2000s as suggested by satellite‐based altimetry studies.     

Estimating runoff from a glacierized catchment using natural tracers in the semi‐arid Andes cordillera

This paper presents a methodology for hydrograph separation in mountain watersheds, which aims at identifying flow sources among ungauged headwater sub‐catchments through a combination of observed streamflow and data on natural tracers including isotope and dissolved solids. Daily summer and bi‐daily spring season water samples obtained at the outlet of the Juncal River Basin in the Andes of Central Chile were analysed for all major ions as well as stable water isotopes, δ¹⁸O and δD. Additionally, various samples from rain, snow, surface streams and exfiltrating subsurface water (springs) were sampled throughout the catchment. A principal component analysis was performed in order to address cross‐correlation in the tracer dataset, reduce the dimensionality of the problem and uncover patterns of variability. Potential sources were identified in a two‐component U‐space that explains 94% of the observed tracer variability at the catchment outlet. Hydrograph separation was performed through an Informative‐Bayesian model. Our results indicate that the Juncal Norte Glacier headwater sub‐catchment contributed at least 50% of summer flows at the Juncal River Basin outlet during the 2011–2012 water year (a hydrologically dry period in the Region), even though it accounts for only 27% of the basin area. Our study confirms the value of combining solute and isotope information for estimating source contributions in complex hydrologic systems, and provides insights regarding experimental design in high‐elevation semi‐arid catchments. The findings of this study can be useful for evaluating modelling studies of the hydrological consequences of the rapid decrease in glacier cover observed in this region, by providing insights into the origin of river water in basins with little hydrometeorological information. Copyright © 2016 John Wiley & Sons, Ltd.     

Base flow‐driven shifts in tropical stream temperature regimes across a mean annual rainfall gradient

Climate change is expected to affect air temperature and watershed hydrology, but the degree to which these concurrent changes affect stream temperature is not well documented in the tropics. How stream temperature varies over time under changing hydrologic conditions is difficult to isolate from seasonal changes in air temperature. Groundwater and bank storage contributions to stream flow (i.e., base flow [BF]) buffer water temperatures against seasonal and daily fluctuations in solar radiation and air temperature, whereas rainfall‐driven runoff produces flooding events that also influence stream temperature. We used a space‐for‐time substitution to examine how shifts in BF and runoff alter thermal regimes in streams by analyzing hydrological and temperature data collected from similar elevations (400–510 m above sea level) across a 3,500‐mm mean annual rainfall gradient on Hawai'i Island. Sub‐daily water temperature and stream flow gathered for 3 years were analyzed for daily, monthly, and seasonal trends and compared with air temperature measured at multiple elevations. Results indicate that decreases in median BF increased mean, maximum, and minimum water temperatures as well as daily temperature range. Monthly and daily trends in stream temperature among watersheds were more pronounced than air temperature, driven by differences in groundwater inputs and runoff. Stream temperature was strongly negatively correlated to BF during the dry season but not during the wet season due to frequent wet season runoff events contributing to total flow. In addition to projected increases in global air temperature, climate driven shifts in rainfall and runoff are likely to affect stream flow and groundwater recharge, with concurrent influences on BF resulting in shifts in water temperature that are likely to affect aquatic ecosystems.     

Influence of melt‐freeze‐cycles on the radionuclide transport in homogeneous laboratory snowpack

Radionuclides released to the environment and deposited with or onto snow can be stored over long time periods if ambient temperature stays low, particularly in glaciated areas or high alpine sites. The radionuclides will be accumulated in the snowpack during the winter unless meltwater runoff at the snow base occurs. They will be released to surface waters within short time during snowmelt in spring. In two experiments under controlled melting conditions of snow in the laboratory, radionuclide migration and runoff during melt‐freeze‐cycles were examined. The distribution of Cs‐134 and Sr‐85 tracers in homogeneous snow columns and their fractionation and potential preferential elution in the first meltwater portions were determined. Transport was associated with the percolation of meltwater at ambient temperatures above 0 °C after the snowpack became ripe. Mean migration velocities in the pack were examined for both nuclides to about 0.5 cm hr^?1 after one diurnal melt‐freeze‐cycle at ambient temperatures of ?2 to 4 °C. Meltwater fluxes were calculated with a median of 1.68 cm hr^?1. Highly contaminated portions of meltwater with concentration factors between 5 and 10 against initial bulk concentrations in the snowpack were released as ionic pulse with the first meltwater. Neither for caesium nor strontium preferential elution was observed. After recurrent simulated day‐night‐cycles (?2 to 4 °C), 80% of both radionuclides was released with the first 20% of snowmelt within 4 days. 50% of Cs‐134 and Sr‐85 were already set free after 24 hr. Snowmelt contained highest specific activities when the melt rate was lowest during the freeze‐cycles due to concentration processes in remaining liquids, enhanced by the melt‐freeze‐cycling. This implies for natural snowpack after significant radionuclide releases, that long‐time accumulation of radionuclides in the snow during frost periods, followed by an onset of steady meltwater runoff at low melt rates, will cause the most pronounced removal of the contaminants from the snow cover. This scenario represents the worst case of impact on water quality and radiation exposure in aquatic environments.     

Process‐based hydrological modelling: The potential of a bottom‐up approach for runoff predictions in ungauged catchments

Conceptual rainfall–runoff models are a valuable tool for predictions in ungauged catchments. However, most of them rely on calibration to determine parameter values. Improving the representation of runoff processes in models is an attractive alternative to calibration. Such an approach requires a straightforward, a priori parameter allocation procedure applicable on a wide range of spatial scales. However, such a procedure has not been developed yet. In this paper, we introduce a process‐based runoff generation module (RGM‐PRO) as a spin‐off of the traditional runoff generation module of the PREVAH hydrological modelling system. RGM‐PRO is able to exploit information from maps of runoff types, which are developed on the basis of field investigations and expert knowledge. It is grid based, and within each grid cell, the process heterogeneity is considered to avoid information loss due to grid resolution. The new module is event based, and initial conditions are assimilated and downscaled from continuous simulations of PREVAH, which are also available for real‐time applications. Four parameter allocation strategies were developed, on the basis of the results of sprinkling experiments on 60‐m² hillslope plots at several grassland locations in Switzerland, and were tested on five catchments on the Swiss Plateau and Prealps. For the same catchments, simulation results obtained with the best parameter allocation strategy were compared with those obtained with different configurations of the traditional runoff generation module of PREVAH, which was also applied as an event‐based module here. These configurations include a version that avoids calibration, one that transfers calibrated parameters, and one that uses regionalised parameter values. RGM‐PRO simulated heavy events in a more realistic way than the uncalibrated traditional runoff generation module of PREVAH, and, in some instances, it even exceeded the performance of the calibrated traditional one. The use of information on the spatial distribution of runoff types additionally proved to be valuable as a regionalisation technique and showed advantages over the other regionalisation approaches, also in terms of robustness and transferability.     

In situ measurements of soil saturated hydraulic conductivity: Assessment of reliability through rainfall–runoff experiments

The saturated hydraulic conductivity, K_s, is a soil property that has a key role in the partitioning of rainfall into surface runoff and infiltration. The commonly used instruments and methods for in situ measurements of K_s have frequently provided conflicting results. Comparison of K_s estimates obtained by three classical devices—namely, the double ring infiltrometer (DRI), the Guelph version of the constant‐head well permeameter (GUELPH‐CHP) and the CSIRO version of the tension permeameter (CSIRO‐TP) is presented. A distinguishing feature in this study is the use of steady deep flow rates, obtained from controlled rainfall–runoff experiments, as benchmark values of K_s at local and field‐plot scales, thereby enabling an assessment of these methods in reliably reproducing repeatable values and in their capability of determining plot‐scale variation of K_s. We find that the DRI grossly overestimates K_s, the GUELPH‐CHP gives conflicting estimates of K_s with substantial overestimation in laboratory experiments and underestimation at the plot scale, whereas the CSIRO‐TP yields average K_s values with significant errors of 24% in the plot scale experiment and 66% in laboratory experiments. Although the DRI would likely yield a better estimate of the nature of variability than the GUELPH‐CHP and CSIRO‐TP, a separate calibration may be warranted to correct for the overestimation of K_s values. The reasons for such discrepancies within and between the measurement methods are not yet fully understood and serve as motivation for future work to better characterize the uncertainty associated with individual measurements of K_s using these methods and the characterization of field scale variability from multiple local measurements.     

Parameter uncertainty of the AWBM model when applied to an ungauged catchment

In this study, a quantitative assessment of uncertainty was made in connection with the calibration of Australian Water Balance Model (AWBM) for both gauged and ungauged catchment cases. For the gauged catchment, five different rainfall data sets, 23 different calibration data lengths and eight different optimization techniques were adopted. For the ungauged catchment case, the optimum parameter sets obtained from the nearest gauged catchment were transposed to the ungauged catchments, and two regional prediction equations were used to estimate runoff. Uncertainties were ascertained by comparing the observed and modelled runoffs by the AWBM on the basis of different combinations of methods, model parameters and input data. The main finding from this study was that the uncertainties in the AWBM modelling outputs could vary from ?1.3% to 70% owing to different input rainfall data, ?5.7% to 11% owing to different calibration data lengths and ?6% to 0.2% owing to different optimization techniques adopted in the calibration of the AWBM. The performance of the AWBM model was found to be dominated mainly by the selection of appropriate rainfall data followed by the selection of an appropriate calibration data length and optimization algorithm. Use of relatively short data length (e.g. 3 to 6 years) in the calibration was found to generate relatively poor results. Effects of different optimization techniques on the calibration were found to be minimal. The uncertainties reported here in relation to the calibration and runoff estimation by the AWBM model are relevant to the selected study catchments, which are likely to differ for other catchments. The methodology presented in this paper can be applied to other catchments in Australia and other countries using AWBM and similar rainfall–runoff models. Copyright © 2014 John Wiley & Sons, Ltd.