nav[aria-label="Primary Navigation"] { padding: 0; & ul { list-style: none; width: 100%; display: flex; flex-direction: row; justify-content: start; align-items: start; gap: 30px; padding: 0; & li { margin: 0; } & ul li { list-style: none; } } }

Nástroj Spatial Analyst Watershed

MatthewPinson

<\/HEAD>

Snažím se identifikovat několik podpovodí v rámci studované oblasti; nicméně nástroj Spatial Analyst Watershed nevygeneroval očekávané výsledky. Omlouvám se, pokud se toto opakuje z předchozích dotazů, ale neviděl jsem žádné existující diskuse, které by dostatečně podrobně řešily mou aktuální situaci.<\/P>

<\/P>

Již jsem použil možnosti Fill sinks a Snap Pour Points k úpravě vstupních datových sad, přesto výstupní výsledky nástroje Watershed vytvářejí pouze velmi malý rastrový obrazec v bezprostřední blízkosti každého pour pointu. Použil jsem 30m DEM i resamplovanou 10m verzi. Všechny datové sady používají stejný geografický souřadnicový systém, takže si nejsem jistý, co dalšího zvážit. Musí existovat nějaký krok v procesu, který mi uniká a který není identifikován nebo vysvětlen v nápovědě nástroje.<\/P>

<\/P>

Přiložil jsem kopii aktuálního výsledku nástroje Watershed. Jakákoli pomoc by byla oceněna.<\/P>

<\/P>

Děkuji.<\/P>

<\/P>

AKTUALIZACE 04\/02\/2015 17:15 ET<\/STRONG><\/P><\/P>Začal jsem proces znovu od začátku pomocí následujících kroků:<\/P><\/P>1. Nový elevacní raster s rozlišením 3m.<\/P>2. Fill sinks.<\/P>3. Resamplovaný elevacní raster na rozlišení 10m (dle návrhu, že 3m může být příliš jemné).<\/P>4. Vytvořen raster směru toku (s vybranou možností Force edge cells to flow out). testováno s vybranou i nevybranou možností Force<\/EM><\/STRONG><\/P>5. Vytvořen raster akumulace toku.<\/P>6. Nastaveny Snap Pour Points pomocí rastru akumulace toku.<\/P>7. Spuštěn nástroj Watershed s použitím Snap Pour Points.<\/P><\/P>Aby bylo jasno, všechny vstupní datové sady používají stejný projected<\/EM><\/STRONG> souřadnicový systém (NAD 1983 HARN State Plane Delaware).<\/P><\/P>Výsledky stále vypadají nespolehlivě. Přiložena je aktualizovaná verze výstupu. Raster akumulace toku je viditelný na pozadí. Černý obrys (dříve vytvořený kolegou) popisuje větší oblast povodí zájmu. Mým cílem je vymezit podpovodí za odběrovými místy popsanými pour pointy. Chybí mi něco? Má specifická projekce význam? Nebo jsou data o elevaci a směru toku problematická, protože je oblast relativně rovná?<\/P><\/BODY><\/HTML>

Najděte více příspěvků označených jako

Water Resources

Accepted answers

All comments

ThomasTaggart

I'd just like to add that I feel it is fine to resample elevation measurements, as long as someone stays away form the nearest neighbor algorithm. With elevation data you always want to choose an algorithm that is appropriate for continuous data (such as bilinear or cubic spline) and never the nearest neighbor.

This problem also surface when projecting the data from a GCS - never ever choose nearest neighbor. I can result in added or missing grid cells that will greatly mess with any hydrology analysis.

See my question and self answer here for more info:

arcgis desktop - What is the source of horizontal and vertical striping in USGS DEMs? - Geographic Information Systems S…

curtvprice

Glad you finally got it working. I suspect the two things that may have tripped you up are:

1) One should not resample or project elevation derivatives. In fact, it's a good plan to use the raster environment settings to help you out once your elevation data are finalized, that is, before the Fill tool is run. Right before the fill, set the extent, snap and cell size environment to match the elevation raster.

2) Interactively edit the gage point locations using the synthetic flow network, and after that, use the Snap Pour tool to make doubly sure they are placed on the locations.

MatthewPinson

Thank you for all of the advice. Although each comment was certainly constructive, no single suggestion offered a successful solution. At this point, I achieved what I believe to be a usable result after clipping the 3m DEM to the preexisting watershed extent and creating the flow direction and accumulation rasters from that clipped DEM. I created the flow direction raster without using the 'force edge cells to flow out' option. After using an edit session to place the proposed pour points in the approximate locations where I think they belong, I then used the Snap Pour Point tool to find the points of highest accumulation within 12m of the points that I placed and used those results in the Watershed tool. That watershed result is shown here.

I think this result is acceptable. If there are any ideas to improve this result, I will certainly appreciate further suggestions. Otherwise, I will convert this to a shapefile and adjust the boundaries as necessary. Thank you.

CAEPMAN

Hi Matthew:

I believe the problem you are having can be resolved by setting the analysis extent for the WATERSHED tool. To do this:

Click on the "Environments..." button at the bottom of the WATERSHED GUI
Expand the "Processing Extent" category
From the "Extent" dropdown select your flow direction grid from the list populated from your table of contents
Finally, select "OK" to return to the WATERSHED GUI.

After setting the analysis extent you can populate the WATERSHED GUI as required and then run the tool.

For whatever reason, the default analysis extent for the WATERSHED tool does not seem to be set the same way as it is for other tools, thereby requiring us to remember to set this each time we use the WATERSHED tool. To overcome this, I have put together a Python script tool that sets the extent to the flow direction grid, and I have built in a few other enhancements as well.

Hope this helps!

SepheFox

Matthew, it looks to me like you snapped your pour points to a vector stream layer, which may not be the same as the calculated streams from a flow accumulation. As Paul is saying, you need the pour points to be on the cells of highest flow accumulation. You can use the flow accumulation to produce a modeled vector stream layer, and then snap your pour points to that.

PaulHuffman

I have that problem when my pour points are a little off the local minimums. But you said you used the Snap Pour Point Tool. In an edit session, try displaying your flow accumulation raster behind your pour points. Play with the symbology classification of your accumulation raster so you can see the stream lines distinct from the real low accumulation values. Move the pour points so they will sit right on top of a pixel in the accumulation with the highest value near where you thought the pour point should be.

Sometimes, however, with real flat terrain, the flow modeling just doesn't work as expected. Especially with coarser DEM data like 30 m cells. You don't get more accurate data by resampling to 10m, you get each 30 m cell chopped into 9 cells with the same value (basically).

DanPatterson_Retired

You said "geographic coordinate system" You are using projected data I assume and that this is just a slip on your part? If not, then the elevations and X,Y coordinates will be in different units and the results will be confusing at best. This may affect how the fill process goes etc. Make sure that you are indeed using projected data in all steps and that the sequence of steps is indeed correct.

ChrisDonohue__GISP

I run into this sometimes and when I do I find it is best to fall back on an overview of the whole process to make sure a step was not skipped. More times than not I realize then that I have skipped a necessary step or did a process out of sequence. So to start, check out the diagram here and see if you did all the steps, and in the correct sequence:

Deriving Runoff Characteristics

ArcGIS Help 10.1

Chris Donohue, GISP

Připojte se k diskusi

Přihlášení členové mohou přispívat, sledovat aktualizace a další. Jste tu noví? Zaregistrujte si bezplatný účet.

Rychlé odkazy

Community Help
Find useful guides, FAQs, and documents to help you navigate and make the most of Esri Community.