Keywords
water quality, measurement network, modelling, validation
Start Date
1-7-2008 12:00 AM
Abstract
In many cases, applications of water-quality models have a poor basis of calibration and validation due to limited or lack of input data. Water-quality monitoring often is limited to a few sites within the stream network at which samples are collected and analysed at some intermittent frequency (e. g. monthly). In particular, this limited temporal sample coverage causes problems in modelling water-quality variables such as phosphorous that are mainly transported during short-time events. The resultant loads calculated based upon these data are often subject to large errors, due to this sparse time resolution. In order to address this problem, a meso-scale catchment was equipped with a measurement network for obtaining spatially and temporally higher-resolution data. As test site, the upper catchment of the Gera River (approx. 850 km²) was selected for this study. The catchment features different landscape forms that are typical for middle mountainous regions in middle Europe. The measurement network consisted of six online water-quality stations and 23 additional water-quality sample points distributed in the river network. At these monitoring sites, streamflow, temperature, pH-Value, electrical conductivity, turbidity, suspended solids, dissolved oxygen, nitrate, ammonium, total phosphorous, total organic carbon and other variables were measured or analyzed in the laboratory. In addition, eight weather stations and 24 soil-moisture sites were installed on farm land distributed throughout the catchment. The goal of this study is to improve the existing components describing the hydrological, nitrogen loadings, and overall land-use management using the J-2000-S model with the measured data. These data will also help to implement an erosion-phosphorus component and a stream water-quality component for this particular model.
Field Measurement Network in the Mesoscale Catchment of the Gera River
In many cases, applications of water-quality models have a poor basis of calibration and validation due to limited or lack of input data. Water-quality monitoring often is limited to a few sites within the stream network at which samples are collected and analysed at some intermittent frequency (e. g. monthly). In particular, this limited temporal sample coverage causes problems in modelling water-quality variables such as phosphorous that are mainly transported during short-time events. The resultant loads calculated based upon these data are often subject to large errors, due to this sparse time resolution. In order to address this problem, a meso-scale catchment was equipped with a measurement network for obtaining spatially and temporally higher-resolution data. As test site, the upper catchment of the Gera River (approx. 850 km²) was selected for this study. The catchment features different landscape forms that are typical for middle mountainous regions in middle Europe. The measurement network consisted of six online water-quality stations and 23 additional water-quality sample points distributed in the river network. At these monitoring sites, streamflow, temperature, pH-Value, electrical conductivity, turbidity, suspended solids, dissolved oxygen, nitrate, ammonium, total phosphorous, total organic carbon and other variables were measured or analyzed in the laboratory. In addition, eight weather stations and 24 soil-moisture sites were installed on farm land distributed throughout the catchment. The goal of this study is to improve the existing components describing the hydrological, nitrogen loadings, and overall land-use management using the J-2000-S model with the measured data. These data will also help to implement an erosion-phosphorus component and a stream water-quality component for this particular model.
