Relation between vegetation and sandbar dynamics in the Colorado River

Abstract

The Colorado River with a total length of 2330 km is located in the Southwest of North America, originating from the Rocky Mountains and flowing into the Gulf of California. In 1963 the Glen Canyon Dam was built to generate hydropower. The construction of the dam caused several significant changes in the river system. One of these changes is the expansion of the riparian vegetation. Because of its magnitude and importance as a natural water carrier, the Colorado River has been studied extensively for decades. One can argue that the sandbars in the Colorado River are well mapped and its dynamics are well understood. Additionally, the species of riparian vegetation, the location of growth as well as the dimension of vegetation changes are extensively mapped. However, the influence of the aforementioned riparian vegetation expansion on the sandbar dynamics is poorly understood. It is crucial to shed light on this correlation as the dynamics of the system indicate the course of action and determine the possible activities in the area. The main research objective is to gain better understanding on the correlation between vegetation expansion and sandbar dynamics. Two different methods are used to explain the phenomenon; an analysis of data concerning vegetation and geo-morphology is performed and also a detailed hydraulic and geo-morphodynamic model is created. Seven locations in the Colorado River are selected for the data analysis in the period between 2002 and 2013. Of each location different features are visualized and linked including the geo-morphology of the sandbar and the adjacent channel, the geometry and hydraulics of the channel and finally the vegetation coverage on the sandbars. As a result of this data analysis a clear relation was found between flood cycles and changes in sandbar volume. Moreover, a clear relation between change in sandbar volume and changes in vegetation coverage was observed. Lastly, a clear relation between the location of vegetation and the bed level changes was found. Finally, several methods in order to improve the analytical results are mentioned. The sandbar selected for modelling purposes is a reattachment bar in the Colorado River, overgrown with a significant amount of vegetation which has expanded remarkably over the years. The flow module of the DELFT3D suite is used. The Baptist model assumptions are used for the representation of vegetation. The parameters determined in the Baptist model are derived from field measurements on arrowweed. The flow features that are affected by the expansion of the riparian vegetation are, among others, the characteristic of the primary and secondary eddy, the deflection on the main stream, the location on the reattachment point and the velocity of the flow in the vegetated areas. Inarguably, the above mentioned changes heavily alter the geomorphology of the system. These changes in geomorphology, on its turn, are likely to influence vegetation succession and even the hydraulics. This lastly mentioned feedback, however, has not been thoroughly investigated in this research. Finally, several methods in order to improve the satisfying flow pattern and in order to reduce the discrepancy of the bed level changes are mentioned. Overall it can be stated that vegetation certainly contributes to the stability of sandbars and that it clearly contributes to their growth. Hydraulic forces, which are strongly related to the width of the channel, determine the magnitude of sediment transport fluxes. Sandbars which develop in narrow channels stay low in elevation and are dynamic; this generally obstructs the natural succession of vegetation. Sandbars which develop in wide channels become high in elevation and become stable; this promotes vegetation growth. If vegetation expands on low, dynamic parts of the sandbars the growth of vegetation is likely to promote stabilization and vertical accretion, creating a feedback towards a higher elevation and more stable sandbar. When this occurs the expected long term result is a river system with a significant part of the sandbars becoming high in elevation, less dynamic and vegetated. However continued monitoring of the system is necessary to determine what the exact vegetation extent over time will be. The iconic feature of the Colorado River during the pre-flood period, namely the open dynamic sandbar system, is currently replaced by a more vegetated river system with less open sand. On the one hand, vegetation provides riparian habitat for a variety of species. At the same time it decreases the open sand area available for recreational activities. The final decision upon the fate of vegetation should be taken after weighting the relative importance between scenarios and future planning. A suggestion to improve the overall study is to take into account other aspects that determine the ability of vegetation to expand such as the availability of groundwater, the daily exposure to sunlight, the fertility of the soil and the activity of the local species. If the effect of these parameters gets integrated in the interdependency between sandbar and vegetation dynamics then the result will give a solid picture of the natural system and the current beneficial results of this research will be further strengthened.