Protect , restore and promote sustainable use of terrestrial ecosystems , sustainably manage forests , combat desertification , and halt and reverse land degradation and halt biodiversity loss
15.1 EDUCATION & RESEARCH
Using remote sensing to monitor mangroves
KAUST researchers are
developing methods to monitor mangroves and land-based ecosystems using remote sensing . Over 2021 and 2022 , the team has been working on a project aiming to determine how much carbon from the air do mangroves actually capture , and how much they might store . Previous studies have shown that mangroves absorb more carbon than any other terrestrial ecosystem , including rainforests . Offsetting carbon in the atmosphere to mitigate the impact of carbon-producing activities is a global priority , and healthy mangrove systems may be one way to contribute to carbon neutrality targets . The project , which is part of the KAUST Circular Carbon Initiative , explores the concept of Nature-Based Solutions , being part of a broader effort to support the Kingdom ’ s conservation and afforestation goals , which include mangroves along the Red Sea and east coast of the country .
Using deep learning to detect land desertification
Researchers from the Computer , Electrical and Mathematical Sciences and Engineering ( CEMSE ) Division , have published a study
on land desertification detection through Generative Adversarial Networks ( GAN ) to address desertification detection challenges . Published in 2021 , the team developed a GAN-based method to analyze and detect desertification changes in multi-temporal Landsat optical images . The developed method outperformed state-of-the-art methods and was able to separate desertification events from other land cover changes like deforestation or areas undergoing seasonal phenomena of wild grasses ' dryness .
Improving wildfire detection
As a result of the changing climate , wildfires have increased both in intensity and severity worldwide . To tackle this problem , a team of KAUST researchers from CEMSE has collaborated with the University of British Columbia , Canada , to
propose a novel wildfire detection solution based on unmanned aerial vehicles assisted Internet of Things ( UAV-IoT ). Published in 2021 , the study suggested that deployment of a massive number of low-cost IoT sensors through forests would allow for early wildfire detection at the sensor level . Since inexpensive sensors do not have the necessary battery or computational power to communicate a fire detection event across a massive IoT network to the fire control center , UAVs can be utilized . The combination of IoT sensors and UAVs are able to detect wildfires faster than satellite imaging and are best suited to high-risk regions such as human settlements and national parks .
Unmanned aerial vehicle ( UAV ) piloting and field work on the mangrove sites at KAUST .
34 | KAUST SDG HIGHLIGHT REPORT