Color as Data for Climate Match Map
One map, instant results for commercial nurseries
We were at first a bit daunted when Sustainable Conservation and the PlantRight team, including the California Invasive Plant Council, approached us about building a global map to allow a person to select any location in the U.S. and show where else in the globe had similar climate conditions.
The query is not that complex, but running any query on global data is pretty costly, especially if you’re trying to do it in a browser.
And the players are significant: Lowe’s, Home Depot, and Orchard Supply have all signed on as PlantRight retail nursery partners in recent years. Growers of ornamental plants who ship their botanical merchandise across California, and across the nation, can now provide their retail and wholesale customers with even more environmentally-friendly plants.
The work is important: This map is a key part of an online plant risk evaluation tool to help national and regional growers of horticultural plants know before they grow if a plant poses an invasive risk in any regional marketplace. Considering that half of known invasive plants are of horticultural origin, this ability to predict invasive risk represents a huge win for the industry and environment, as it will prevent new invasive garden plants from taking root, escaping into the wild, and wreaking havoc on native ecosystems.
Our specific remit, part of the larger PlantRight system, was to take any state or USDA ecoregion (or multiples of same) as the input and then return, globally, every area that had similar (a) precipitation, (b) United Nations ecozones, and (c) USDA Plant Hardiness zones. You can try out the final map for yourself here.
When we started the project, our executive director Dan Rademacher had recently come to GreenInfo from Stamen Design, where he’d worked on two iterations of Climate Central’s Surging Seas sea level rise map. Stamen’s tech director Seth Fitzsimmons devised a system for that to encode multiple flood levels in both meters and feet into the red, green, and blue (RGB) channels of an invisible map layer and then use those to filter other elements of the map. The folks at Mapzen, home to GreenInfo board member Michal Migurski, have done some great work where they’re encoding elevation into RGB to allow dynamic rendering of topographic maps.
So we had it on our minds to pack the data into images. Geospatial Analyst Maianna Voge and Senior Web Developer Greg Allensworth worked together to pull the data out of ArcMap, producing raster layers with the proper values for each data layer in the RGB channels of GeoTIFFs (that we sliced into tiles), as well as text match tables that listed all the RGB codes for each state or ecoregion a user can choose on the map.
Greg stitched those two ends of the system together using the HTML canvas layer for the instant filtering in the final map.
Here’s an animation of the green “PRE Combined” layer followed by the three data layers that comprise the final composite layer:
Once the PlantRight map was up and running, Greg also abstracted the code and pulled it into a Leaflet plugin hosted on Github:
We used it again on the ProtectedLands.net map, which just launched last week to showcase the newest version of the Protected Areas Database of the United States. There are more than 100,000 protected areas in the database, so searching them one by one is a challenge. But thanks to the categorical color filtering we learned to do for the PlantRight map, it’s really easy to make and share a view like this of only open access lands in Colorado:
It’s always satisfying when new techniques we learn for one project immediately pay dividends on another.
Another way that one node in the GreenInfo Network connects to another, and everyone benefits.
