Volunteer provides GIS technical assistance to South African organization

Background

The Protected Area Management (PAM) initiative is a partnership between the National Geographic Society, Esri, and GISCorps. This initiative seeks to facilitate and support the effective use of Esri’s ArcGIS Solutions for Conservation, which has been made available at no cost to protected areas throughout Africa through a partnership between Esri and the National Geographic Society. ArcGIS Solutions for Conservation provides protected area managers with a secure suite of configurable technology for better managing and protecting parks and sensitive conservation areas.

Established in 1993, Selati Game Reserve (SGR) is a nearly 28,000-hectare private conservation area situated in Limpopo Province, South Africa, near the western boundary of Kruger National Park. SGR was founded by a collective of landowners with a shared vision to restore and preserve the region’s rich biodiversity. The reserve is renowned for its diverse habitats, including savannah grasslands and indigenous flora, and is home to a variety of wildlife, including the Big Five and the sable antelope. Selati is also recognized for its commitment to sustainable resource management and conservation efforts.

Emily Mills, a GISCorps volunteer from Washington DC, was recruited for this project to assist SGR with their GIS needs. Emily submitted the following report at the conclusion of the mission.

Selati is fully fenced, so wildlife management is essential to protect animals from poaching and ensure populations are healthy within the reserve boundaries. The Selati Research Team (SRT) uses a variety of technologies to monitor its priority wildlife species, including GPS radio collars, VHF telemetry, and drones to track and locate individual animals. They also leverage ArcGIS Survey123, Field Maps, and QuickCapture for mobile data collection.

The Mission

The mission started with a few virtual discovery sessions to meet the SRT, get familiar with their ArcGIS Online and data ecosystems, and learn about their key data workflows and tools. Then I spent a week in August 2025 at Selati alongside the team to experience their day-to-day work, understand the realities and challenges on-the-ground, and identify opportunities to optimize their use of ArcGIS tools for wildlife management.

The Work

Mobile Applications

The SRT uses Survey123 to collect data for a variety of field operations, including rhino dehorning, a strategy to deter poachers from killing rhinos for their horns. It takes a large, well-coordinated team to locate and dart the rhino, dehorn, fit it with a GPS collar, and collect biometric data. This process needs to happen quickly to minimize the time the rhino is under anesthesia and ensure its safety during the procedure.

One of the first tasks I worked on was optimizing the rhino dehorning survey so that the SRT can collect data and fit the GPS collar efficiently. This included automating questions where possible and reducing the need for separate maps or reference information. For example, we used the iPad location services and a geopoint type question to determine grid-cell location automatically. We also auto-generated recommended GPS collar settings based on the input measurement of the animal’s leg circumference. It was very rewarding to accompany the team for dehorning operations and observe how the improved survey increased the efficiency of data collection.

The SRT also uses Field Maps and QuickCapture for collecting data on game drives. A key challenge at Selati is spotty or non-existent cell coverage in some areas of the reserve, which affects the SRT’s ability to view and use maps. To address this, I created a custom basemap with key reference layers published as vector tile layers optimized for offline use. 

We set this custom basemap as the default basemap for the Selati ArcGIS Online organization and made it discoverable across all mobile applications (Survey123, Field Maps and QuickCapture). Now, volunteers and guides can still reference the basemap while collecting data and navigating around the reserve, even if they are in areas without cell coverage.

ArcGIS Online Administration

In the past, the SRT has run low on ArcGIS Online credits due to the large amount of data collected and the associated storage required. To avoid this in the future, I generated ArcGIS Online administrative reports to identify items and members that were candidates for backup or deletion to save on data storage and credit usage. I created an ArcGIS Online Python Notebook to manage groups and members, and I scheduled a Notebook Task so that the Notebook runs at a set interval in the future. 

Since SRT has been using GPS collars and other tracking devices for a few years, there is a lot of historical data taking up space and using credits. So, I developed a Python batch script that can be run at scheduled intervals on their local computer to regularly download backups of important data layers. To accompany this, I developed a workflow in ArcGIS Online and ArcGIS Pro that the SRT can follow to backup and delete old records out of the feature layers to free up space.

GPS Collar Data Visulaization

To take full advantage of the real-time GPS collar data, the SRT was interested in developing a data visualization of inter – and intraspecies interactions of collared individuals. The species of interest included lion, cheetah, hyaena, vulture, and buffalo. These interactions can indicate predation events, mating events, or conflicts between carnivore species, e.g. scavenging or stealing prey.

To achieve this, I first developed an ArcGIS Online Python Notebook to automate detection of interactions—GPS points that occur within specified distance and time thresholds. The notebook was automated to run daily overnight and write out interactions of interest to a new feature layer. The feature layer was visualized in a webmap along with the GPS collar layers for each species of interest. 

Next, I used ArcGIS Experience Builder to create an interactive web application that shows the species interactions map. It allows the SRT to filter the GPS collar data by species, individual, and date/time, and displays an animation of the animals’ movements. The goal is for the SRT team to open the app every morning to see if any interesting interactions occurred in the last 24 hours that warrant a field visit.

Conclusion

I loved my time spent volunteering for this GISCorps mission, and it was an amazing and rewarding experience. The site visit was invaluable to see the variety of tools and data workflows of the SRT in practice. It helped me gain a better appreciation for some of the key challenges on-the-ground – such as time/resource constraints and limited bandwidth/cell coverage – and generate creative solutions to address them. Not to mention some incredible wildlife encounters that I will never forget!

Project completed.