Tech-Driven Coexistence: Integrating "Sky & Ground" Monitoring and Early Warning to Mitigate Human-Elephant Conflict in Yunnan, China

The Asian elephant remains endangered due to poaching, habitat loss, and growing human-elephant conflict—one of the most pressing conservation challenges globally. In response, our team developed an integrated monitoring and early warning system that combines aerial drones with ground-based infrared cameras. This “sky & ground” network ensures real-time, full-time, and large-scale coverage of elephant activity across Yunnan Province. Drones use thermal imaging and zoom to detect elephants even at night, while 600 infrared cameras enable 24/7 data capture with real-time alerts.

Supported by China’s national and provincial forestry departments, the project trains local youth as monitors and forms community teams. Over 130,000 early warnings have been issued, helping reduce human-elephant conflict and improving safety.

It contributes to GBF Targets A, B, and 1, 4, 20, 21. It strengthens biodiversity protection, enhances community engagement, and offers a scalable model for coexistence.

Visit page

Related content

Radical Restoration: Democratizing Climate Tech for Ecosystem Recovery

In 2024, Distant Imagery Solutions planted 5.5 million mangroves in the UAE using self-engineered wooden drones designed for simplicity, adaptability, and cost-effectiveness. This milestone redefines the potential of community-driven, ecosystem-based restoration.

With projects ready in Brazil, Tonga, Kenya, and Indonesia, Distant Imagery empowers communities to lead. Our licensing and training platform equips locals to build and operate drones, fostering global knowledge exchange.

More than a technological solution, this is a movement of shared empowerment and stewardship. By connecting communities, Distant Imagery creates a global network innovating together to restore ecosystems and build climate resilience.

Also, our AI-powered Monitoring, Reporting, and Verification (MRV) system tracks restoration progress and combats illegal activities, ensuring ecological recovery and long-term protection.

Shorebirds conservation

Reclamation led to a significant reduction in coastal wetlands, resulting in a dramatic decline in waterbirds, highlighting the urgent need to restore habitats. In order to provide an ideal high tide roost site for the most important shorebird staging site in Yellow Sea, we conducted a restoration project in Tiaozini Wetland. A 720 acre seaward fish pond was developed as a demonstration area for the restoration of high tide roost site. The project has successfully created habitats that meets the needs of various migratory waterbirds through a series of scientific measures such as micro terrain modification, water level regulation, vegetation control, and automatic auxiliary monitoring of waterbirds. Since its establishment, TZN720 has attracted a large number of waterbirds, with a single record of 188500 waterbirds in 2024. Our successful experience of this project has been included in the “Biodiversity 100+Cases” released at the 15th Conference of the Parties to the Convention on Biological Diversity (CBD-COP15) NGO Forum.

Using drones for non-invasive monitoring and assessment of crocodilian populations – a generalizable and accessible tool for stakeholders in conservation

This standartised solution combines drone technology and an innovative allometric approach to monitor and assess crocodilian populations who face significant threats, leading to population declines with 50% the 27 crocodilian species are threatened, with 25% critically endangered. Drones equipped with high-resolution cameras capture images of even partially submerged crocodiles, allowing precise, non-invasive length estimates based on head-to-body allometric ratios. This method overcomes challenges like logistical constraints, cost, need for highly trained personnel, observer bias, detectability, wildlife disturbance, and safety risks of traditional surveys. By covering remote areas efficiently, the solution enhances biodiversity mo,nitoring, informs conservation strategies, and enables a wide range of additional information to be collected. This solution is very cheap, requiring minimal training, and accessible to a wide range of users, including indigenous peoples and local communities and conservation practitioners.

Himalayan wolf conservation. Human-carnivore conflict mitigation to support traditional Tibetan mountain communities face global warming

Carnivores are among the most endangered species. The loss of apex predators jeopardizes global biodiversity. The Himalayan wolf (Canis lupus chanco) inhabits ecosystems highly vulnerable to climate change. The survival of the Himalayan wolf depends on human-carnivore coexistence and the systems ability to adapt to climate change impacts on the Himalayan ecosystem.

Our Solution aims to improve the livelihood of local Tibetan communities and promote human-carnivore coexistence. We achieve this through scientific research in collaboration with locals; forming Community Conservation Groups (CCG); building communal predator-proof livestock corrals (co-funded by the local communities), piloting fox lights, conducting conservation workshops for herders, women, and school children and installing mini-libraries in local schools. This Solution for fostering human-carnivore coexistence in the high Himalayas is part of an ongoing long-term science and conservation project, the Himalayan Wolves Project (www.himalayanwolvesproject.org).

Forest Cloud: A Digital Hub for Global Restoration and Conservation

The Restoration Platform, the core of Forest Cloud, transforms global restoration by simplifying fundraising and ensuring transparency for forest conservation. Since 2018, it has evolved to bring back a trillion trees and conserve 3 trillion existing trees. The open-source, open access, digital solution currently supports scientific restoration initiatives as well as managing and coordinating conservation and restoration efforts globally. It benefits from robust Restoration Standards, proprietary restoration monitoring applications, a robust peer-review system and an ecosystem of ancillary digital solutions – the ‘Forest Cloud’. Proven across 300 restoration initiatives and growing, the Platform unifies restoration organizations (ROs), donors, and scientists. Having demonstrated its success by restoring over 94M trees in 6 years, we now seek to improve its scale and geographic reach and solve the logistical challenges to contribute to a sustainable, thriving future.

Empowering Conservation with AI-Powered Monitoring and Early Warning Systems.

WildGuard AI is an Edge AI-powered sensing system by NOARKTECH designed to protect biodiversity and reduce climate-related risks in forest and farm landscapes. Using IIoT (Industrial internet of things) devices connected via LoRaWAN (Long Range Wide Area Network) and LTE (Long Term Evolution), it detects wildfires, heatwaves, and human-wildlife conflict in real time. It combines bioacoustic AI (e.g., elephant and hornbill calls), temperature, and gas sensors to detect anomalies early.

Paired with PAMS (Protected Area Management & Security System), a dashboard for real-time alerts, analytics, and visualization, the system supports proactive action. Piloted in India’s Western Ghats, it helped reduce crop loss, fire spread, and wildlife threats. WildGuard AI and PAMS enable scalable, affordable, and sustainable nature-based intelligence for climate resilience and conservation.

A Park-Led Conservation Education Model: Ranger Goes to School in Komodo National Park

Ranger Goes to School (RGTS) is a ranger-led, school-based conservation education initiative developed by Komodo National Park to inspire high school students in Labuan Bajo—the gateway to the park—to become future stewards of nature. Founded by senior park ranger Muhammad Ikbal Putera, RGTS addresses the growing disconnect between youth and their environment through experiential learning, storytelling, and the use of accessible technology. Since 2022, the program has reached over 1,000 students across five high schools. RGTS blends classroom and outdoor settings using tools such as PictureThis for plant identification, Canva for creating student presentations, and Kahoot for interactive quizzes. Drones capture stunning aerial footage that is transformed into documentary films to build emotional connections with the park, while social media extends the program’s reach and visibility. These technologies not only enrich the learning experience but also strengthen students’ sense of place and commitment to conservation.

Computer vision for vulture species monitoring across Africa

This AI-powered solution supports Global Biodiversity Framework (GBF) Targets 17 and 20 by leveraging biotechnology to strengthen species monitoring and fostering technology transfer for biodiversity conservation. Using a deep learning model, “You Only Look Once version 11” (YOLOv11), it automates the identification and analysis of critically endangered vultures (Gyps africanus, Gyps coprotheres, Gyps rueppelli, Torgos tracheliotos) in drone and camera trap data. Data from African Parks Network (APN), Southern African Wildlife College (SAWC), Endangered Wildlife Trust (EWT), platforms like iNaturalist, and GBIF will serve to train and validate the model.

The project tackles challenges like labor-intensive monitoring and data gaps. Its open-source design promotes accessibility, collaboration, and capacity-building across African conservation networks, directly addressing gaps in biodiversity data and monitoring systems.

Connected Conservation Foundation

Protected areas are vital for biodiversity, livelihoods, and climate stability—yet 60% of African parks remain “paper parks” due to limited infrastructure.

Connected Conservation Foundation (CCF) is bridging this gap with Internet of Things (IoT) tools that deliver real-time data to monitor ecosystem health, detect threats early and empower communities.

Since its 2019 pilot at Kenya’s Lewa Wildlife Conservancy, CCF has developed a large-scale sensor ecosystem with Actility, Cisco and EarthRanger. Long Range Wide Area Network (LoRaWAN) networks now cover 10M hectares across Kenya, South Africa and Uganda—including the Masai Mara, Tsavo and Northern Rangelands. Over 1,000 sensors and 131 gateways in 35 conservancies feed into EarthRanger, providing a comprehensive 360-degree view of ecosystem health and threats, supporting teams to track assets, onboard sensors, and respond to dangers. This scalable tech stack drives biodiversity conservation action, protects livelihoods and supports the global 30×30 conservation goal.

Leveraging Remote Sensing for Ecological Management and Conservation at Jiangsu Yancheng Wetland & Rare Birds National Natural Reserve

Jiangsu Yancheng Wetland & Rare Birds National Natural Reserve is home to the critically endangered Grus japonensis and other rare species, dependent on wetland mudflat ecosystems. However, invasive species like Spartina alterniflora threaten biodiversity, causing ecological degradation and diminishing biodiversity. To address this, remote sensing technology is used to monitor vegetation evolution, offering valuable insights for ecological management.

Yangtze River Delta Pilot Site of the UN Ocean Decade’s CoastPredict Program, led by Nanjing Normal University, combines remote sensing technology and conservation efforts to support management in Yancheng. Remote sensing has tracked the evolution of key wetland vegetation (Spartina alterniflora, Suaeda salsa, and Phragmites australis) since 1990, helping to understand environmental impacts, human activities, and species interactions. This initiative aids in developing more effective conservation strategies, benefiting both the reserve and broader global effort.

The Mamba: A Drone-based Robotic Arm to Preserve Endangered Plants in Cliff Environments

Tropical islands are home to unique floras, and host a vast number of endemic plant species. Threats have led to a significant population decrease in Hawaii, where 97% of these endemic plant species are listed as endangered, critically endangered, or extinct. Vertical cliff habitats work as refugia to protect plants from these threats. However, these habitats make conservation work particularly difficult, forcing botanists to use risky and time-consuming methods such as abseiling to access remote plant populations. We developed the Mamba, the first aerial system capable of sampling plants on cliffs. The system has allowed the collection of seeds from many critically endangered species including Hibiscadelphus woodii, as well as the recent discovery of a new species: Schiedea waiahuluensis. The conservation collections made by this tool are thriving in nurseries, demonstrating the impact on the conservation cycle for organisms located in difficult-to-reach environments, and the actions necessary to prevent plant extinction.

Ribbit – a web app for automated identification and classification of anuran species

Ribbit is a citizen science web app that uses few-shot transfer machine learning to record, identify, and classify frog and toad calls, contributing crowdsourced data to the Global Biodiversity Information Facility (GBIF) to address data gaps, especially in the Global South. Identification apps offer significant potential for automatic in situ biodiversity monitoring (Global Partnership on Artificial Intelligence, 2022; Tuia et al., 2022; Nieto-Mora et al., 2023). Our app focuses on anurans, as they are crucial ecosystem indicators (Estes-Zumpf et al., 2022), with over 40% of the species at risk of extinction, and their unique vocalizations are ideal for acoustic identification. Beta testers revealed the app’s potential to contribute data to GBIF while empowering citizen scientists to engage in ecological monitoring. By creating an open-access platform for labeling biodiversity data, Ribbit enables conservation organizations to develop strategies for protecting vulnerable populations and preserving critical ecosystems.