article / 11 May 2023

Panthera: Monitoring Wildlife with Open-Source Data Management Tools 

In our first Conservation Tech Showcase case study, you'll learn about the innovative PantheraIDS, an open-source data management platform and analytical engine that processes, standardizes, and manages conservation-related biological monitoring data.

This case study showcases Panthera's innovative technology, PantheraIDS, a state-of-the-art, open-source data management platform and analytical engine that processes, standardizes, and manages conservation-related biological monitoring data, such as camera trapping data and telemetry data, in an efficient and streamlined manner.

About Panthera

Founded in 2006, Panthera is a nonprofit organization that works to conserve all 40 species of wild cats and their vital ecosystems. Operating in over 50 countries, the organization develops innovative strategies based on scientific research, global partnerships, and the expertise of its team of leading biologists, law enforcement experts, and wild cat advocates. 

Panthera collaborates with local and international NGOs, scientific institutions, and governments to create targeted conservation strategies for the world's most threatened small cats. 

About PantheraIDS

In 2017, Panthera initiated the development of PantheraIDS (Integrated Data System), a state-of-the-art, open-source data management platform and analytical engine to process, standardize, and manage conservation-related biological monitoring data. 

Photo by Sage C

The platform was in response to an urgent need to process camera trapping data, telemetry data, and observation data in an efficient, standardized, and streamlined manner — leveraging advanced data processing techniques and analytical approaches in a user-friendly format — to democratize complex methodologies, and enable rapid conservation action informed by robust conservation science.

PantheraIDS is built within an R shiny framework, and therefore is well positioned to take advantage of the R statistical computing language to conduct advanced analyses using native R. 

This is a particularly powerful combination, since R is the primary coding language used in the conservation field, particularly for advanced analytics and publication-quality graphical content. In addition, the R community is vibrant and highly active, with over 18,000 packages deployed on CRAN (Comprehensive R Archive Network). The majority of these packages are immediately implementable within PantheraIDS without any modification. 

With most analytical approaches needed by the conservation community already built as R packages, PantheraIDS is uniquely poised to take advantage of this need by offering a platform for seamless integration. Importantly, because PantheraIDS provides the user with an easy-to-use interface, the software acts as an enabler to conservation practitioners, allowing non-technical users to implement complex processes and powerful analytics at the click of a button.

Photo by Sage C

Conservation practitioners worldwide often face the need to determine where a certain species lives and moves, and how many of them exist. To meet this conservation requirement, PantheraIDS offers a set of analytical tools that aid in answering these queries. 

Specifically, the occupancy analysis and spatial-capture recapture analysis tools offered by PantheraIDS help to respond to the "where" and "how many" questions, respectively. PantheraIDS utilizes both frequentist and Bayesian statistical frameworks, employing cutting-edge approaches to address these questions. This approach allows non-technical individuals to process a wide range of datasets, from camera trapping images to tracking collar data, and analyze those datasets within a single integrated system. Thus, PantheraIDS generates robust estimates of animal density and species distribution without the need to learn complex statistical or programming languages or switch between multiple applications.

PantheraIDS not only utilizes the R programming language, but also natively incorporates the Python programming language. Additionally, it features advanced computer vision and machine learning technology that allows conservation practitioners to easily deploy convolutional neural networks for a variety of tasks.  

Photo by Sage C

These tasks include automatically classifying camera trap images to species level, counting the number of individuals within camera trap images, detecting a range of animal behaviors within each camera trap image, and matching individually identifiable animals using a combination of unique coat pattern-matching and spatial-temporal inference. The use of these advanced technologies provides conservation practitioners with a powerful tool for monitoring and managing wildlife populations.

Positive Conservation Outcomes Through PantheraIDS

The use of PantheraIDS has already resulted in positive conservation outcomes, both directly and indirectly. For example, PantheraIDS is utilized throughout Panthera's Southern African projects to manage data collected across a vast camera trapping network, which is the world's largest leopard (Panthera pardus) biological monitoring network. These data are being used at the national level in South Africa to inform an adaptive, nationwide management plan for leopards. As a result, species management has significantly improved, and population growth rates in key areas of South Africa have increased. 

Similarly, PantheraIDS is used in West and Central Africa to manage and improve the management of leopards and other wild cat species. For jaguars (Panthera onca) in Mesoamerica and South America, PantheraIDS is relied upon to manage long-term datasets to aid in the delineation of new national preserves or further bolstering existing protected areas. 

Within Panthera's Small Cat Program, PantheraIDS plays a central role in ensuring standardized and accessible data across the globe to inform robust conservation interventions at scale. Prior to the development of PantheraIDS, conservation practitioners were collecting data in disparate formats, which is still the case for many conservation organizations. The unstandardized approach is a significant barrier to conservation efforts because standardized, high-quality data that are easily available are crucial for designing conservation interventions and creating real impact.

Photo by Sage C

PantheraIDS has played a crucial role in Panthera's conservation mission, with its widespread use across five continents, 35 countries, and by over 30 organizations. This analytical tool facilitated the production of over 40 scientific publications annually, with much of the research relying on the data curated within PantheraIDS. 

Camera trapping surveys are an extremely popular and common biological monitoring technique, and PantheraIDS remains one of the most advanced tools for this data type. Its extensive use is evident in the processing of more than 15.2 million camera trap images, covering all cat landscapes and comprising over 96 million species-level records and 11.2 million individual-level records. These records were gathered using 5,600 cameras and encompassing 470 different species and 4,900 individually identified animals. In addition, PantheraIDS collects over 1.1 billion data points from a vast suite of ancillary data.

Scaling PantheraIDS: Strategies for Expansion and Implementation in Other Regions

Notably, many conservation practitioners do not have access to suitable internet, and so require solutions that enable efficient data processing and management within an offline environment. PantheraIDS represents the only in-production and scalable conservation tool capable of deploying these complex technologies in an offline mode — especially our AI technology. 

Importantly, all tools present within PantheraIDS are applicable in both online and offline modes, across all operating systems using containerization technology (i.e., Docker) or via the PantheraIDS web application hosted on Amazon Web Services. 

With PantheraIDS being open-source (GPL-3 licensed software), Panthera aims to provide the tool to as many conservation organizations as possible. Since PantheraIDS is containerized using Docker, the software is easily deployable and scalable, either as a locally installed instance (i.e., on a laptop or desktop computer), or via cloud computing infrastructures (e.g., container orchestration services, Kubernetes).

Photo by Sage C

PantheraIDS has not only revolutionized the way conservationists manage data, but it has also addressed the challenge of limited access to suitable internet in the field. The tool is the only scalable solution capable of deploying complex technologies in offline mode, including AI technology. 

Furthermore, all tools within PantheraIDS are applicable in both online and offline modes, across all operating systems, through containerization technology or the PantheraIDS web application on Amazon Web Services. PantheraIDS is open-source (GPL-3 licensed software), and its availability to other conservation organizations is a priority for Panthera. The tool's containerization using Docker allows for easy deployment and scalability, either as a locally installed instance or via cloud computing infrastructures, such as container orchestration services, Kubernetes.

What's Next for PantheraIDS?

PantheraIDS still had a shortcoming in its ability to integrate rapidly with other conservation software, including SMART and EarthRanger, due to its direct JDBC connections in R instead of API endpoints. To address this, Panthera's Conservation Technology department and Programming Unit urgently needed to design the PantheraIDS API, requiring additional in-house capacity to develop the API and associated endpoints for PantheraIDS and any future integrations with other applications. 

Panthera proposed to capacitate the Programming Unit with a new developer to work on the API build, and to collaborate with SMART and EarthRanger developers via the Gundi team to integrate PantheraIDS. 

Successful completion of the PantheraIDS API would allow conservation practitioners to leverage their datasets in PantheraIDS directly within SMART and EarthRanger, or vice versa, and enable better identification and rapid response to global conservation challenges. SMART, for example, would benefit from PantheraIDS' statistical features and analytical capacity, creating new analytical opportunities and breaking down computational and statistical barriers evident in software not fully utilizing native R or Python.

How can I engage with PantheraIDS?

As a WILDLABS conservation tech community member, you have the opportunity to leverage PantheraIDS, a state-of-the-art, open-source data management platform, and analytical engine designed to process, standardize, and manage conservation-related biological monitoring data. 

By taking advantage of PantheraIDS's advanced data processing techniques and analytical approaches, you can democratize complex methodologies, and enable rapid conservation action informed by robust conservation science. 

Contact their point of contact at [email protected] to learn more!

Explore the 2023 Conservation Tech Showcase

Our new Conservation Tech Showcase is a series exploring successes and bold ideas in the world of conservation technology, shared through case studies about exciting projects from outstanding organizations around the world.

Learn more about the series and its case studies here on WILDLABS.

Add the first post in this thread.

Want to share your own conservation tech experiences and expertise with our growing global community? Login or register to start posting!