Search

A long-term flood record from the Buffels River, the largest ephemeral river of NW South Africa (9250 km2), was reconstructed based on interpretation of palaeoflood, documentary and instrumental rainfall data. Palaeoflood data were obtained at three study reaches, with preserved sedimentary evidence indicating at least 25 large floods during the last 700 yr. Geochronological control for the palaeoflood record was provided by radiocarbon and optically stimulated luminescence (OSL) dating. Annual resolution was obtained since the 19th century using the overlapping documentary and instrumental records. Large floods coincided in the past within three main hydroclimatic settings: (1) periods of regular large flood occurrence (1 large flood/~30 yr) under wetter and cooler prevailing climatic conditions (AD 1600–1800), (2) decreasing occurrence of large floods (1 large flood/~100 yr) during warmer conditions (e.g., AD 1425–1600 and after 1925), and (3) periods of high frequency of large floods (~ 4–5 large floods in 20–30 yr) coinciding with wetter conditions of decadal duration, namely at AD 1390–1425, 1800–1825 and 1915–1925. These decadal-scale periods of the highest flood frequency seem to correspond in time with changes in atmospheric circulation patterns, as inferred when comparing their onset and distribution with temperature proxies in southern Africa.

In this study we propose a multi-source data approach for quantifying long-term flooding and aquifer recharge in ungauged ephemeral rivers. The methodology is applied to the Buffels River, at 9000 km2 the largest ephemeral river in Namaqualand (NW South Africa), a region with scarce stream flow records limiting research investigating hydrological response to global change. Daily discharge and annual flood series (1965–2006) were estimated from a distributed rainfall-runoff hydrological model (TETIS) using rainfall gauge records located within the catchment. The model was calibrated and validated with data collected during a two year monitoring programme (2005–2006) at two stream flow stations, one each in the upper and lower reaches of the catchment. In addition to the modelled flow records, non-systematic flood data were reconstructed using both sedimentary and documentary evidence. The palaeoflood record identified at least 25 large floods during the last 700 yr; with the largest events reaching a minimum discharge of 255 m3 s−1 (450 yr return period) in the upper basin, and 510 m3 s−1 (100 yr return period) in the lower catchment. Since 1925 AD, the flood hydrology of the Buffels River has been characterised by a decrease in the magnitude and frequency of extreme events, with palaeoflood discharges five times greater than the largest modelled floods during the period 1965–2006. Large floods generated the highest hydrograph volumes, however their contribution to aquifer recharge is limited as this depends on other factors such as flood duration and storage capacity of the unsaturated zone prior to the flood. Floods reaching flows associated with 5–10 yr return periods (120–140 m3 s−1) and flowing for 12 days are able to fully saturate the Spektakel aquifer in the lower Buffels River basin.

This paper summarises innovative research into the assessment of longterm groundwater recharge from flood events in dryland environments of the Kuiseb (Namibia) and the Buffels (South Africa) rivers. The integrated water resource management (IWRM) policies and institutions affecting the exploitation of groundwater resources in each of these developing countries are compared. The relatively large alluvial aquifer of the Kuiseb River (∼240 Mm3) is recharged from irregular floods originating in the upper catchment. Reported abstraction of 4.6 Mm3 per year is primarily consumed in the town of Walvis Bay, although the groundwater decay (pumping and natural losses along the period 1983–2005) was estimated in 14.8 Mm3 per year. Recharge is variable, occurring in 11 out of 13 years in the middle Kuiseb River, but only in 11 out of 28 years in the middle-lower reaches. In contrast, the Buffels River has relatively minor alluvial aquifers (∼11 Mm3) and recharge sources derive from both lateral subsurface flow and floodwater infiltration, the latter limited to a recharge maximum of 1.3 Mm3 during floods occurring once every four years. Current abstractions to supply the adjacent rural population and a few small-scale, irrigated commercial farms are 0.15 Mm3 yr−1, well within the long-term sustainable yield estimated to be 0.7 Mm3 yr−1. Since independence in 1990, Namibia’s water resource management approach has focussed on ephemeral river basin management of which the Kuiseb BasinManagement Committee (KBMC) is a model. Here, some water points are managed independently by rural communities through committees while the national bulk water supplier provides for Walvis BayMunicipality from the lower aquifers. This provides a sense of local ownership through local participation between government, NGOs and CBOs (community-based organisations) in the planning and implementation of IWRM. Despite the potential for water resource development in the lower Buffels River, the scope for im

Synthetic hydrograph methods of Snyder, Williams and Hann, Rodriguez-Iturbe and Valdes, and Soil Conservation Service were applied to eight catchments in Antioquia Colombia. Those basins have addequate rainfall and runoff records and therefore one can compare the errors of model’s results. Next an optimization was run on each model parameters in order to minimize the mean quadratic error. Despite of the substantial reduction in the errors, they remain high. We interpret these results as a consequence of the very irregular space-time distribution of rainfall that it is not adequately represented by the rain gages, and of the nonlinearity of the ranifall-runoff transformation not captured by the linear hydrograph methods., Se hace un análisis de los errores producto de la aplicación directa a ocho cuencas en Antioquia Colombia, de los hidrogramas unitarios sintéticos desarrollados por Snyder, Williams y Hann, Rodriguez-Iturbe y Valdés y por el SCS (Soil Conservaron Service) y se reporta el resultado de la recalibración de las ecuaciones. Las cuencas escogidas en el departamento de Antioquia tienen registros simultáneos de precipitación y caudal. El procedimiento de recalibración consiste en la minimización de la suma de los errores en el caudal y el tiempo al pico. Los errores entendidos como la diferencia entre valores calculados y observados, elevada al cuadrado. Mediante un método de optimización, se estiman los parámetros de cada modelo. Los resultados permiten concluir que los errores obtenidos luego de la recalibración disminuyeron notablemente. Sin embargo, los errores aún son altos, lo que indica que así se obligue a estos modelos a predecir caudales y tiempo al pico con el mínimo error no, son satisfactorios para una gama de diversos eventos, con diferente duración de la lluvia. Interpretamos estos resultados por la irregular distribución espacio temporal de la precipitación y por el carácter no lineal del proceso precipitación-escorrentía.

This study makes part of a broader project to estimate hydrological balances all over the country, at a spatial scale of 5 minute-arc. Several methods are implemented to estimate long-term evaporation over Colombia, such as those by Turc (1945), Morton (1982), Penman (1948), and Holdridge (1978). Point estimations are interpolated to obtain regional averages. Results are tested against observed long term average precipitation and runoff for a wide range of catchment sizes. Turc’s methods turned out to give small error despite being dependat only on precipitation and temperature. Morton's method is based upon the energy balance for which radiation data are needed at a high spatial resolution (not available yet), and therefore results exhibited large errors. Penman's method turns out to be an approximation of the Priestley - Taylor method, based on aerodynamical and energy balance equations. This method exhibited small estimation errors. Holdridge’s method is based on the concept of life zones produced results with large spatial variability and large errors. Morton and Penman’s methods exhibited larger evaporation estimated values than those obtained with Turc or Holdridge's, throughout the country. Despite the quality and lack of existence of long-term hydro-climatological data sets, these methods provide a better understanding of both potential and actual evaporation in Colombia., Este trabajo hace parte de un estudio de Balances Hidrológicos para Colombia, en el cual se han investigado y evaluado las componentes del ciclo hidrológico, a nivel de promedios de largo plazo y a una escala espacial definida por retículas de 5 minutos de arco (aproximadamente 9 km de lado). Se presentan los resultados de los métodos de Turc (1945), Morton (1982), Penman (1948) y Holdridge (1978) para estimar la evaporación multianual en Colombia. Los datos usados se han extractado de múltiples fuentes nacionales e internacionales. La evaluación de los diversos estimadores se ha hecho mediante balances hídricos en cuencas con datos existentes de caudal y precipitación media multianual. El método de Turc es el más sencillo de implementar puesto que sólo tiene como variables independientes la temperatura y la precipitación. Morton y Penman son métodos difíciles de implementar debido a la información requerida, pero por su desarrollo teórico, es una metodología muy interesante. El método de Holdridge esta basado en zonas de vida, que dependen de las condiciones climáticas de las diferentes regiones, muestra gran variabilidad, pero fue uno de los que arrojó mayores errores y tiende a la subestimación. A pesar de las limitaciones en calidad y cantidad de información, la implementación y análisis de estos métodos dan las pautas para mejorar las estimaciones de la evaporación (potencial y real) a largo plazo y para el desarrollo de posibles métodos de cálculo aplicables a las condiciones climáticas de Colombia.