Algorithm Research Of Watershed Analysis Without Filling Depression Based On Grid DEM

Digital watershed analysis was an important branch of digital terrain analysis,was the construction of wisdom city,an important part of digital earth,was the main source of water conservancy information.As a subset of digital terrain analysis,digital watershed analysis was also based on DEM data.DEM was a kind of real topography discrete digital expression,the representation of the actual surface patterns.In the organization form of DEM data,regular grid DEM based on its simple organization form,convenient algorithm design and more suitable for computer processing,became the most widely and frequently used terrain analysis data now.DEM data production process due to the resolution,elevation interpolation and rounding the output error made DEM data produces the data which was not tally with the real terrain.It was called spurious features whose form of expression for topographic depression.The traditional way was preprocessing DEM data and removed spurious features.But the traditional way existed some disadvantages.That was there were always many similar dams,lakes,ditches sag area in the real surface patterns.In the traditional algorithm,these area always been treated as spurious features and removed.That makes the results did not tally with the real terrain.So,in order to close to the real surface,this paper presented no filling depression watershed analysis algorithm.This algorithm reserved real depression and analyzes watershed characteristics included in the watershed.This paper studied flow analysis,accumulated of confluence,watershed extract etc.main steps of digital terrain analysis under the condition of no filling depression.The main study and the results could be summarized as the following respects:(1)Identified the real and the spurious depression.By analyzing the characteristics of the real and the spurious depression,from the area,depth,characteristic of the border of depression and the distribution of the depression,the paper analyzed the feasibility of discriminant depression by these characteristic factors and designed a way of identifying the real and the spurious depression and extracted the real depression effectively and achieved the effect that wanted.(2)Extracted the border of depression.The border of depression was an important factor of judging the real and the spurious depression.This paper used the rough set theory and analyzed the ambiguity of the border of depression.By confirming the similar set of the rough set,the paper confirmed a range of the border of depression.And it described the border of depression with point by point tracking method.(3)Analysis on flow direction of the inside of the depression.The real surface morphosis of depressions like lakes and the reservoirs were a part of the watershed.And their internal flow need to be analyzed.The paper used priority-flood,confirmed their internal flow and connected with outer region and made a whole river network.(4)Divided no filling depression watershed.It would have an effect on the subsequent steps if we had put depression influence into the whole watershed analysis.This paper elaborated the main step of accumulated of confluence analysis,extracting river network,dividing watershed.And through the experiment,produced the river network map and watershed map with the depression.