Search

Long-term river flow data are instrumental in understanding the risks posed by fluvial flooding and drought events. They assist in providing greater understanding of the drivers of hydrological variability and are fundamental to contextualising extreme events and detecting emerging climate change signals. In many parts of the world, these insights are limited given the lack of long-term river flow records, which in turn limits the opportunity for robust planning decisions in the water sector. Here we link datasets that take advantage of data rescue efforts and paleoclimate reconstructions with rainfall-runoff modelling techniques, to develop river flow reconstructions at a monthly timestep. Using Ireland as an exemplar, gridded European temperature and precipitation datasets are used to reconstruct monthly flows from 1766 to present for 30 river catchments. Uncertainties in reconstructions from hydrological model structure and parameters are integrated using the Generalised Likelihood Uncertainty Estimation (GLUE) method. Reconstructions are validated through use of quality assured long-term precipitation series for selected catchments and an assessment of historical drought undertaken. We find that the methodology produces robust reconstructions of monthly flows that allow for detailed assessment of catchment responses to varying climatic conditions on monthly, seasonal and annual timescales. Despite the uncertainty associated with reconstructions, we find that the low flows signatures associated with notable historic drought events are identifiable in our reconstructions. Finally, we highlight that the datasets and methods used can be readily employed to reconstruct river flows in other European contexts. [ABSTRACT FROM AUTHOR]

Without careful land management, agricultural land in karst systems can pollute water courses, with polluted waters travel quickly to and through the sub-surface in karst systems. However, detailed understanding of how rapidly nitrate moves within the karst critical zone (from soils to aquifers) is limited by low resolution sampling for this highly-transmissive system. To understand nitrate behavior and its controlling factors, we deployed sensor technology at five sites to generate autonomously high-resolution time series of discharge and [NO3−N] in a farmed karst catchment in SW China. The [NO3−N] time series exhibited rapid response to rainfall-induced increases in discharge and a large magnitude in [NO3−N], ranging from 0.72 to 16.3 mg/L. The highest mean [NO3−N] and normalized annual fluvial export were observed in one headwater catchment with well-developed conduit structure. Elsewhere this was less-pronounced due to buffering by the karstic aquifer network as the contributing catchment area increased with distance downstream. Clear seasonal variation in NO3−N export occurred in response to source availability, most notable in catchments with valley agriculture: up to 97% of nitrate was exported from the headwater catchment in the wet season, with much of that loss occurring in only a couple of months, compared to 61% of nitrate exported at the larger catchment scale. The sensor time series showed that the aquifers were chronically polluted with nitrate. [ABSTRACT FROM AUTHOR]

The incision of rivers into bedrock is considered as one of the most important processesdriving long-term landscape evolution. River incision is non-linearly dependent on riverdischarge, because river discharge must exceed a certain threshold to be able to transportbedload sediment and expose the riverbed to erosion. Fluvial erosion efficiency is, therefore,strongly controlled by the frequency and magnitude of flood events. In this study wehypothesise that catchment attributes, such as regolith thickness, lithology and vegetationcover, modulate the discharge response after a precipitation event, which has implications forthe fluvial erosion efficiency. We present preliminary results from a study, in which we exploit daily discharge recordsfrom 322 river gauging stations on a large climatic gradient in Chile (18˚ S-55˚ S). Thestudied rivers have catchment areas of 15 km2 to 24200 km2, which differ in topography,lithology, regolith thickness, vegetation cover and snow and glacier cover. First, we studieddischarge variability (i.e. magnitude-frequency distribution of flood events) andthe occurrence of different discharge regimes as result of seasonal variations, byobtaining the best fit of a weighted sum of two inverse gamma functions to the data(Scherler et al., 2017). Each of the distributions is interpreted to correspond to oneflow regime. When the weighted sum of two inverse gamma functions performedbetter than one, and each flow regime lasted at least 3 months, we interpret thedifferent flow regimes to reflect seasonal variation in a catchment. Results reveal largevariations in discharge variability along the climate gradient and strong seasonalvariations around 30˚ S-40˚ S which is mainly driven by climatic forcing and snowmelt. We are currently studying discharge variability by obtaining the flood flashinessparameter, the variability index and hydrologic response time from fitting the streamflowrecession model to the daily discharge records (Deal et al., 2018 after Botter et al., 2007). Wewill focus on the differences between seasons and El Niño, La Niña and normalyears and will compare all parameters with calculated topography and datasets ofclimate, vegetation density, regolith thickness, lithology and snow and glacier cover todisentangle their controls on flood flashiness, variability index and hydrologic responsetime. References: Botter, G., Porporato, A., Rodriguez-Iturbe, I. and Rinaldo, A.: Basin-scale soil moisturedynamics and the probabilistic characterization of carrier hydrologic flows: Slow,leaching-prone components of the hydrologic response, Water Resour. Res., 43(2),1–14. Deal, E., Botter, G. and Braun, J.: Understanding the role of rainfall and hydrology indetermining fluvial erosion efficiency, J. Geophys. Res. Earth Surf., 2018. Scherler, D., DiBiase, R. A., Fisher, G. B. and Avouac, J. P.: Testing monsoonal controlson bedrock river incision in the Himalaya and Eastern Tibet with a stochastic-thresholdstream power model, J. Geophys. Res. Earth Surf., 122(7), 1389–1429, 2017. [ABSTRACT FROM AUTHOR]

Sediment dynamics in river catchments are governed by tectonics, climate and biota. They act directly on the transport process by controlling topography and sediment transport capacity (runoff) and indirectly by controlling the flux and grain size distribution (GSD) of sediments supplied from hillslopes to the fluvial system. Since climate is the main driver for biotic activity, both effects are strongly coupled, and often difficult to disentangle. However, while climate forcing shows a clear trajectory towards increased erosion with increasing humidity, the effect of vegetation is more ambiguous. On short timescales vegetation reduces hillslope erosion and sediment supply from hillslopes by stabilizing soils and trapping sediments. On longer timescales an increased sediment residence time and enhanced biochemical weathering of material in the soils enhances the production fine grained material and hence strongly alters the GSD of material delivered to the fluvial transport system. We state that the influence of vegetation on the grain size distribution of sediments supplied to the river system is a key factor in the biotic control on sediment dynamics, since it exerts a strong control on the transport mechanics and morphodynamics in the river system which in turn control incision and deposition and hence landscape evolution. To investigate vegetation induced shifts in GSD, we collected grain size samples of hillslope and channel sediments in four granitic mountainous river catchments located along the strong bioclimatic gradient of the Chilean Coastal Cordillera, ranging from the arid, non-vegetated Atacama desert in the north (26°S) to the humid and densely vegetated Cordillera Nahuelbuta in the south (38°S). 78 volumetric bulk samples were taken with grain sizes ranging from clay (d = 0.3μm) to boulders (d >300mm). By sampling hillslope and fluvial sediments from the hilltop to the channel outlet we track the effect of vegetation on the spatial variability of GSDs along the sediment transport path.Results show that hillslope sediments in humid areas contain higher fraction of fines compared to the arid areas, while channel sediments in contrast are much coarser. Uniform GSDs in the arid catchments reflect nonselective and thus transport-limited conditions in the river channels. In contrast, in humid catchments GSDs show strong differences between fine hillslope sediment and coarse channel deposits, indicating size selective and supply-limited transport conditions in catchment with higher vegetation coverage. [ABSTRACT FROM AUTHOR]