CARTO’s Grants For Good Program is excited to share the great work being done by our grantee, Farmland Monitoring Project. It is our pleasure to introduce Adam Calo, PhD student at UC Berkeley specializing in agricultural policy in California, and Farmland Monitoring Project team member. In this guest post Adam tells us about his project and the mapping process with CARTO.
After interviews with farmers in the California Central Coast, I decided to focus my research on the theme of land access for small scale and beginning farmers. I was originally fascinated by the ecology of agriculture, but after talking to tenant farmers in California I recognized the need to deal with social and political constraints on farmer’s autonomy. Chief among those is the issue of a lack of secure land access.
I wanted to see how promising and more accessible information technologies might interact with this issue of farmland access. Could participatory mapping tools match farmers to available land? And could open mapping platforms bring to the fore key information about land tenure in a critical agricultural region?
These questions brought forth the Farmland Monitoring Project (FMP) and CARTO quickly filled a central role in the project’s main web application. Agriculture’s administrative dynamic relies on land ownership parcels, zoning laws, legal boundaries that indicate, restrict, or incentivize agriculture. In terms of data, these can be thought of administrative boundaries, water district zones, and ownership parcels. A useful mapping tool for farmland access should both represent official data and allow for farmers, landowners and other actors in the food system to submit their own assessments about available land and existing farms. Here’s how we set out to do this, using CARTO as a central piece.
First we go out into the wild and locate public datasets like parcel data from the county assessor. We clean them using ArcGIS as best as possible then import them into CARTO. Removing urban parcel (very tiny) was really helpful in getting the datasets to an appropriate size with an ability to render quickly. There we use the SQL and Postgres functionality embedded in CARTO to make subsets of data and join data layers where appropriate. In doing so, we were able to make a contiguous parcel map of three adjacent counties with ownership information attached.
We then use CARTO.js to embed the data in our custom application. At this point, my skills as an amateur map scientist have been adequate, but to host a custom application a PhD student in the UC Berkeley School of Information, Seongtaek Lim, lent his support. He decided to use Django to support the application, so that the tool could be managed by someone with my level of expertise through built in controls in the admin panel. The main customizable feature needed was to allow individuals, like farmers, landowners, or agricultural professionals to submit their own queries upon the data we have aggregated.
Next, we need to allow for citizen monitoring of farmland access. This meant the ability to capture near-real time submissions from mobile devices. For this, we turned to Ona.io, a mobile data collection platform, and a bit of our own tools to automatically pass new submissions to CARTO layers.
The creation of a technical intervention seems a little backward. But, there is a lot of good evidence out there that when a group monitors a resource (like suitable farmland) the same group is likely to take political action to safeguard that resource. My brief exploration into the world of farmland investment has shown that powerful speculators have been readily monitoring agricultural land for some time. Perhaps the FMP can open the door for a more open farmland monitoring process.
The FMP has two main features: the ‘map’ application is a tool for running custom queries about agricultural land in a select California region. The mapbook page is a series of map stories that explore land access more broadly. There are plenty of maps there and more to come!
Happy data mapping!