Arcpy: Polygons to Centroids (within Polygons, and with all Attributes)

Here’s a fairly simple python script that creates centroid points (constrained to fall within polygons) and attaches all the attributes (plus a link field, “ORIG_ID”) from the polygons to the centroids. Notably, it uses cursors to read raw geometry objects – no geoprocessing tools (except CreateFeatureclass). Also, while Advanced licensees may use the Feature to Point tool for this procedure, the method presented here is completely free, using no extensions or non-basic licensing – I don’t know of another free way to achieve this in ArcGIS.

Wouldn’t it be easier to run a spatial join on the points to attach the attributes? Yes, but spatial join is slow. In my test of 1000 polygons, the cursor method completed in about 2 seconds, while adding a spatial join call extended the run time to more than 10 seconds. This could mean big savings on large datasets.

  1. # import libraries
  2. import arcpy, os
  3. # set input/output parameters
  4. polyFC = arcpy.GetParameterAsText(0)
  5. outCentroids = arcpy.GetParameterAsText(1)
  6. # set overwrite environment
  7. arcpy.env.overwriteOutput = True
  8. # if the output file does not exist, create it. Add “ORIG_ID” field.
  9. if not arcpy.Exists(outCentroids):
  10.     arcpy.CreateFeatureclass_management(os.path.dirname(outCentroids),
  11.                                     os.path.basename(outCentroids),
  12.                                     “POINT”,
  13.                                     polyFC,
  14.                                     “”,
  15.                                     “”,
  16.                                     polyFC)
  17.     arcpy.AddField_management(outCentroids,‘ORIG_ID’‘LONG’)
  18. # create an InsertCursor containing all the fields in ourCentroids, which are all the fields in polyFC plus “ORIG_ID”
  19. cursor = arcpy.da.InsertCursor(outCentroids, ['*'])
  20. # read all features in polyFC
  21. for row in arcpy.da.SearchCursor(polyFC, ["SHAPE@",'*','OID@']):
  22.      # create an array to hold a new, modified row
  23.     rowArray = []
  24.      # read all the fields except “SHAPE@”
  25.     for fieldnum in range(1,len(row)):
  26.         # if this is the second field [i.e. SHAPE], replace it with the centroid coordinates
  27.         if fieldnum == 2:
  28.             rowArray.append(row[0].centroid)
  29.         else:
  30.             rowArray.append(row[fieldnum])
  31.     # write the new row to the cursor
  32.     cursor.insertRow(rowArray)
  33. del row

Looping through the fields in this way (lines 28 – 35) feels cumbersome – if anyone knows a better way to do this, please let me know!

Using ArcObjects in Python

Recently, I found the need for more fine-grained control over a map document than the Arcpy library alone could afford, so I turned to the comtypes library (downloaded here) and associated Snippets file (ArcGIS 10.1 version here), which allows access to VBA-based ArcObjects. The Mark Cederholm presentation I followed is from the 2010 ESRI Dev Summit, so it is possible that this method is somewhat outdated, but it does exactly what I need.

Here is a simple example showing how to add some guides to the layout rulers in a map document, which is not possible solely using Arcpy:

# Import the modules
from comtypes.client import GetModule, CreateObject
from Snippets import GetStandaloneModules, InitStandalone

# First time through, need to import the “StandaloneModules”. Can comment out later.
# GetStandaloneModules()

# Get the Carto module
esriCarto = GetModule(r”C:\Program Files (x86)\ArcGIS\Desktop10.1\com\esriCarto.olb”)

# Create a map document object
mxdObject = CreateObject(esriCarto.MapDocument, interface=esriCarto.IMapDocument)

# Create new mxd file

# Create some new layout guides
for i in range(1,21):

# Change current view to layout

# Save the mxd

# Delete the map document object
del mxdObject

…and the result:


Solar Radiation: free method

This is a follow-up post to Tuesday’s Garden Placement Using Publicly Available LiDAR, which relied on ArcGIS’ Spatial Analyst extension. Huge hat tip to the LAStools Linkedin conversation, and particularly Daniel Grohmann, for mentioning the freely available SAGA GIS tool, Potential Incoming Solar Radiation. I should note that this is truly my first attempt at using SAGA, so apologies in advance if I’m leading you astray in some way.

To recap, we had created a digital elevation model (DEM) from publicly available LiDAR from the City of Prince George, and were ready to calculate solar radiation values for a city block in order to optimally locate a garden plot.

  1. Download, install, and open SAGA GIS (it’s free)
  2. Import your DEM into SAGA. I used the GDAL: Import Raster module (Modules -> File -> GDAL/OGR -> GDAL: Import Raster). Double-click the layer in the Data workspace window to view the raster and ensure it imported correctly.
  3. Open the Potential Incoming Solar Radiation tool (Modules -> Terrain Analysis -> Lighting -> Potential Incoming Solar Radiation). Here is how I set up my run:
  4. After changing the colour ramp to the default (blue -> red), we are left with a very similar, if not identical, solar radiation map comapred to that produced by Spatial Analyst:

Garden Placement Using Publicly Available LiDAR

Here’s a quick method for finding the sunniest spot on your property in Prince George, BC (and elsewhere if you can find the data). Unfortunately, this method is not free from start to finish (it requires ArcGIS Spatial Analyst), but there are a fair number of free tools and data used.

  1. Get the data. Specifically, you need the raw LiDAR (.las file) for your area, which are not available for download from the City of PG’s Open Data Catalogue. While you’re at it, also ask for the orthophotos (aerial photos) and cadastral data (parcel boundaries, road files, etc.). Get in touch with the City of Prince George’s GIS department to make your data request:
  2. Load your orthophoto into your GIS program (I’m using ArcGIS, but you can use QGIS for the time being [it's free]).
  3. Create a new polygon feature class and draw a polygon to narrow down your area of interest.Image
  4. Download LAStools (it’s free for the tools you need). Bring the LAStools toolbox into your GIS (there are ArcGIS and QGIS versions).
  5. Run the lasclip tool to ignore the millions of LiDAR points that you’re not interested in.
  6. Run las2dem to create an elevation raster of your LiDAR data.Image
  7. Consult your favourite source, like the Farmer’s Almanac, to determine the timing for your growing season, .
  8. Here is where I used a Spatial Analyst tool called Area Solar Radiation, which is not free (in fact, it’s darn expensive). Run the tool, using your latitude (PG is about 53.914), and frost free start and end dates (for PG, June 4 to Sept. 3). You should end up with something like this, which my legend tells me ranges from blue (353 watt hours per square metre) to red (883061 WH/m^2): Image
  9. Now, it’s a matter of overlaying parcel boundaries, finding your property, and seeing what kind of sunlight you can expect. I’ve circled a few good candidates on this block with lots of sun in the backyard, and (surprise!) some of them are existing gardens.ImageImage

Bonus: another cool thing you can do (for free) is load your monochrome elevation model in Blender, extrude the terrain heights, drape the aerial imagery overtop, and create a 3D animation like this one.

PaperJS Voronoi Diagrams


I’ve been tinkering around with Voronoi diagrams, inspired by the example on the PaperJS site. My example (here) uses Raymond Hill’s Voronoi Javascript implementation and PaperJS to derive the diagram from a point distribution. Note: the redraw animation is a bit backwards, as the point locations are determined first, driving the placement of the Voronoi diagram.

Another, more festive example, can be found here, which shows the distribution of lights on our Christmas tree. Falalalala-la-la-la-la.

Refractometer: ABV Calculator


I got a mysterious package in the mail yesterday – by the time I figured out that it was a Christmas present from my brother, it was too late.

Turns out the gift was a refractometer, a device used to measure an index of refraction, or how light gets bent going through another substance. Few people outside laboratories have much use for such a thing, but along the way some science-savvy beer brewer realized that you could measure the amount of dissolved sugar, and thus the amount of potential alcohol, in freshly brewed beer. By comparing the amount of sugar at the beginning and end of fermentation, you can infer the amount of alcohol by volume (ABV) in the finished beer.

Most homebrewers use a hydrometer in a similar manner, but this requires several chances of contamination (each time the wine-thief is filled, often several times) and wastes a good deal of beer along the way (about 100ml for each measurement). Refractometer measurements use a mere couple of drops which are collected only once per measurement.

Unfortunately, once alcohol is present in the liquid (i.e. any time after the initial measurement), it messes with the refraction measurement and a mathematical correction must be applied in order to compare the two samples. Using formulas from Northern Brewer, Primetab, Realbeer, I made an interactive calculator for calculating ABV from initial and final refractometer readings (in Brix units), and you can find it here. Drag the bars up and down to match your initial and final measurements.

Notes: there are tons of other online refractometer calculators, and I’ve seen some that use different formulas (or at least different coefficients), so beware. Also, if you’ve got Matlab, my brother made a library for charting and doing the calculations for yourself (blog post here).