The Effects of Topographic Map Scale and Costs of Land Surveying on Geometric Model and Flood Inundation Mapping

摘要

The quality of topographic datasets plays a key role in deriving terrain model and hydraulic simulation. In developing countries, data-sparse regions and most parts of the world, accessing to high-quality datasets is almost impossible and topographic maps with a specific scale should be used to derive the geometry of river bed and floodplains. For these areas, understanding the effects of map scale on the costs of land surveying/river projects and flood inundation mapping may lead to an appropriate flood modeling and reduce huge amounts of financial resources. This research addresses the effect of using topographic maps with different scales on the hydraulic properties and focuses on using the relationship between topographic map scale, the costs of land surveying and hydraulic properties. The results from the three river reaches (SojasRood and SafaRood rivers in Zanjan and Mazandaran provinces, respectively and Sarbaz river in Sistan-Baluchistan province, Iran) used in this study show that in small-scale rivers (SojasRood and SafaRood rivers), the effect of topographic map's scale on water surface elevation (WSE) isn't meaningful up to 1:4000 (4K) map scale, while in the case of Sarbaz river WSE is approximately scale independent (even up to 1:10000 (10K) map) and there is no considerable discrepancy between low and high-quality maps in predicting this important variable. For example, in SojasRood river the mean absolute error (MAE) in simulation of WSE for low-quality maps (5K and 6K maps) varies between 0.79m and 1.78m, while for high-quality maps (1K, 2K and 3K) it restricts to only 0.25m. Also, in the case of Sarbaz river, for maps with the scale of higher than 5K (high resolution maps) the maximum values of MAE and RMSE statistics limit to 0.13m and 0.16m, respectively. Moreover, findings demonstrate that using the topographic maps with the scales of 2K and 7K instead of high quality maps (1K and 2K maps in small-scale rivers and Sarbaz river, respectively as base maps) lead to the same geometric model and the mean relative error (MRE) in simulating inundation extents is lower than 10%. These outcomes clearly indicate that by accepting some reasonable errors the low quality maps, that are cost-effective and not time-consuming, can be considered as alternative maps for flood simulation in low budget projects. Furthermore, assessing the costs of ground surveying shows that it highly depends on the scale of topographic maps and by using low quality instead of high quality maps the costs of topographic maps' production significantly decreases. In addition, changes in hydraulic properties due to using these maps are not considerable when compared to the significant saved financial resources.