Assessment of flow and sediment transport in rivers and on floodplains ecosystem based on hydraulic modeling
Lowland river-floodplain systems are typical interactive ecosystems characterized by low altitude differences, high frequency of flooding, low flow velocity and the exchange of energy and substrate between channel and floodplain. Stream power is recognized as a force of formation and development of the river morphology and channel-floodplain interaction. The rising flood risk of Europe in future global change scenarios indicates an urgent need of the study of in-stream and floodplain flow/sediment processes in this area. The aims of the dissertation are: 1) test the reliability of the mobile OTT Qliner with acoustic Doppler technology (ADQ) and the methodology for velocity and discharge measurement; 2) verification of the new formulas of the vertical flow profile with easy-to-determine parameters for a better understanding of the in-stream flow structure and improvement of the data accessibility; and 3) the evaluation of flow and sediment processes in the channel and on the floodplain of the lowland river-floodplain system in the German Upper Stör catchment by setting up 1-D hydraulic HEC-RAS models for every sub-catchment. In this dissertation 366 measurements at 174 cross sections in 8 catchments of different sizes located in Northern Germany, Central Germany and South eastern China were collected with the ADQ, FlowSens and ADC from Sep 2011 to Jun 2012. The comparisons between data from different equipment were made to test of the reliability of ADQ. With vertical velocity profiles collected from Upper Stör catchment (Northern Germany), Kinzig catchment (Central Germany) and Changjiang catchment (Southeastern China), a new structure of the prediction formulas involving mean vertical velocity (u/ū) and dimensionless relative water depth (y/H) was proposed and verified. Three field campaigns were carried out in Sep 2011, Jan 2012 and Jun 2012 in the ten selected river sections in Upper Stör catchment, in order to set up the HEC-RAS models for each sub-basins. The discharge, velocity and geometry data of the river section and every cross section were collected during the field campaign, while the topography of the floodplain was extracted from 1m digital elevation model (DEM). Four gauge stations located in the Upper Stör catchment provided long-term discharge series (2000-2010) for the HEC-RAS model calibration and validation. The sediment series and the discharge series in the sub-basins without gauge station were generated by the SWAT model, which has been well calibrated and validated by a previous researcher. The mobile OTT Qliner with acoustic Doppler technology (ADQ) provides a highly efficient and accurate way to collect flow discharge and velocity data. The results of flow average, profile, layer and point values of ADQ compare very well with electromagnetic or ultrasonic devices. Inner setting tests revealed that the measurement is more sensitive to cell size than to time interval setting. The cell size depth ratio between 0.1 and 0.2 meters produced the highest reliability. Longer time interval is recommended during measurement in shallow and slow flowing rivers. The applications of the new structure of the prediction formulas show that the substitution of u* and y with the u/ū and y/H are reliable and applicable, and all three resulting curves provide satisfactory results. The parabolic curve describes the vertical flow velocity distribution in the best quality with an error of 7%, while the errors of the logarithmic and power fitting are 10% and 11%. In water depth direction, the predicted results of the middle part of the vertical profiles tend to be more reliable and precise than near the water surface area. The error in the river bed region is the highest. Higher catchment slope resulted in larger coefficients and constants in logarithmic and power fitting. The introduction of the seasonal roughness modification to the HEC-RAS flow models of the sub-catchments of Upper Stör profoundly improves the model performance. The analysis of channelized and floodplain stream discharge and power show that under 10-years-peak discharge conditions only 1-10% of flow power was generated by floodplain flow, while 40-75% volume of water is located on the floodplain. Unit stream power is proportional to the increase of stream discharge, while the increase rate of unit in-channel power was 3 times higher than that of unit flow power on the floodplain. The variation of the increasing rate of the flow power is dominated by the local roughness height, while the flow power distributed on the floodplain mainly depends on the local slope of the catchments. Analysis of the HEC-RAS sediment output indicates that the Upper Stör catchment is dominated by the deposition process during the study period (2000-2010), with a deposition depth of 2.85 cm in the river sections. The floodplain deposition accounts for only 1 % of the total sedimentation amount, and the rest 99% sediment deposits on the river bed. The land use/cover condition results in the different sedimentation amount in different catchments, while the granularity of the channelized sedimentation is dominated by the altitude of the river section, and the granularity of the floodplain sedimentation positively correlates with the stream power of the flood. The results yielded by the combination of HEC-RAS and SWAT model are comparable to the traditional radioactive dating or sediment trapping method in similar and nearby catchments. The dissertation proved the accuracy of the new equipment and the measurement methodology, verified the applicability of the easy-to-determine formulas of the vertical flow velocity profiles, and revealed the basic knowledge of the lateral distribution of the flow, energy and sediment during the flood event in Upper Stör lowland river-floodplain system. The findings and the achievements of the dissertation provide a baseline and guideline for the further studies in lowland catchments, especially the river/floodplain interactive processes and nutrient and contamination related studies.