Xinyu Gao

Hi Olaf, Thank you so much for these further explanations! I think my result fits better to the first case you talked above. When I checked the aspects of the two neighbors of cell1, they do kind of directing against each other. Using the D8 algorithm for this special case makes sense, but I just did not notice the description for this particular case in the paper, so that is why I kept feeling confused. Now I am clearer about the specific techniques in DEMON and the question in my mind is clarified....

Hi Olaf, Thank you very much for your response! I think fig. 1 works well to explain why the diagonal neighbor can receive flow, but I am still confused about the result in the figure I attached there. As water flows from the northeast to the southwest, the distribution of flow makes sense until for cell1, which seems to me that it assigns its flow directly to the diagonal neighbor cell2. The aspect value of cell1 was calculated as 215.15, and it is expected that flow in cell1 should be assigned...

Hi Volker, This is a follow up question about the DEMON algorithm. I tried to use just one cell as the source by signing the weight in a way as what you said. However, I still got the result showing that the diagonal neighbor cell receive water from its upslope neighbor. Like the case for cells 1, 2, 3 as marked in the attachment, cell 3 receives flow from cell 2, and cell 2 from cell 1 (the general flow direction is to the southwest). For my understanding of DEMON algorithm, should not the flow...

Hi Volker, Thank you very much! This is very helpful to me. Sincerely, Xinyu

Hi Volker, Thank you for the clarification! This confirms my thinking about how the DEMON algorithm works. I guess that the tool in SAGA just calculate the flow accumulation from all cells in the DEM, but it would be great help if I can see the source code of the tool. I know the SAGA software is open source, but where can I get the code for a specific tool (Flow accumulation (flow tracing))? Sincerely, Xinyu

Hello, I have a very specific question about the algorithm implemented in the tool for flow accumulation (flow tracing)--->hydrology--->terrain analysis. Does anybody know if the neighbor cells on the diagonal positions are also counted when distribute flow from the center cell using the DEMON algorithm? I know from the Costa Cabral (1994)'s paper that those neighbors on the diagonal positions are not considered, but when I looked at the result from the SAGA tool, it seems that those neighbors are...