This is part 2 which will be pretty simple. This is more of the “get organized” part. In part 1 I was able to generate a watershed boundary from LIDAR Elevation Data. After it was generated I went back and checked the watershed line and really only found one thing that looked weird. Well – more than one thing BUT this was fixable. If you zoomed into one part of the watershed you had this bit of weirdness (notice the triangle).
The DEM had been flattened where the buildings are and this one area had a lot of large buildings. Generally these DEMs aren’t perfect. It’s a small area though. I’m going to leave it in the interest of automating and not fixing every small part of this. I say all of this to go “Check your data as you collect it. It may not be as right as you want it”.
I decided to dump my watershed data into a Geopackage out of GRASS and suck it up into QGIS to check things out.
30 years ago this was all Arc/INFO and we were using the Coverage model. Which meant “1 editor and multiple viewers”. It also meant if you had a project area of any substantial size you chopped it up by grid to get more people working. In a past life we used the USGS 7.5 minute grid system to inventory everything.
One more screen shot:
This area only falls into two 7.5 minute quads (black line). It falls in four 3.75 minute quads (yes that was a thing and that’s in Orange). Geopackage is a one editor multiple viewer set up. So do I chop it up into 4 project areas? 2 project areas? Do I replicate a process from a long time ago? Short answer No.
If I do some quick calculations in QGIS this watershed is about 4500 acres or 7 square miles. It’s not terribly large so I’m not inclined to split this up. For now I’m going to leave this in a geopackage with an eye on PostGIS if this starts growing more than I planned. PostGIS does up the complication BUT it also gives you some functionality as you have one database that will allow multiple editors.
So what do I need to get this moving:
- Impervious surfaces (which are roads, parking lots, buildings etc)
- Imagery (current and maybe historical)
- DEM (which I have)
- Any other data I can grab which may end up being Soils and Wetland data.
- Streams
I spent about an hour last week generating the watershed data. This episode was about 30 minutes of “thinking and planning”.
Part 3 we’re going to hit the ground running and see if we can come up with some numbers very quickly.
