Pine wood nematode (Bursaphelenchus xylophilus) is an invasive pathogen in South Korea, where it has caused pine wilt disease (PWD) with extremely high mortality of native pine species (Pinus densiflora, Pinus thunbergii, and Pinus koraiensis). Since the disease spreads by its vectors, native pine sawyer beetles (Monochamus alternatus and Monochamus saltuarius), the cost of monitoring the expansion has been rapidly increasing. Furthermore, it is even more costly to eliminate new and isolated infections since unremoved infected trees act as new sources of infection through the preferred oviposition of the beetles on such trees. The methodology of combining unmanned aerial vehicle (UAV) and object detection based on deep learning provides the opportunity to solve such problems, as UAV with RGB camera can provide high spatial resolution aerial image and digital surface model (DSM), which can be used for object detection with excellent results. In this study, we evaluated the performance of this method to detect dead pine trees in PWD-damaged areas. In particular, to ensure low omission error of monitoring, YOLOv3 was employed for object detection as the model design is focused on minimizing the omission error. We also modified the model so that the positions and crown diameter could be estimated. Four detection models were trained using four different combinations between aerial images (R, G, B) and DSM from UAV. Among them, the model from RGB showed the highest performance (recall: 0.9909, precision: 0.8438) and was selected as the optimal model. Our results suggest that our method can contribute to low-cost and effective monitoring of the dead pine trees while maintaining low omission error, which is critical for PWD management.

African swine fever (ASF) is a viral hemorrhagic fever fatal to animals of the Suidae family. It has spread from Africa to Europe and Asia, causing significant damage to wildlife and domesticated pig production. Since the first confirmed case in South Korea in September 2019, the number of infected wild boars has continued to increase, despite quarantine fences and hunting operations. Hence, new strategies are needed for the effective control of ASF. We developed an agent-based model (ABM) to estimate the ASF expansion area and the efficacy of infection control strategies. In addition, we simulated the agents’ (wild boars) behavior and daily movement range based on their ecological and behavioral characteristics, by applying annual hunting scenarios from past three years (2019.09–2022.08). The results of the simulation based on the annual changes in the number of infected agents and the ASF expansion area showed that the higher the hunting intensity, the smaller the expansion area (24,987 km² at 0% vs. 3533 km² at 70%); a hunting intensity exceeding 70% minimally affected the expansion area. A complete removal of agents during the simulation period was shown to be possible. In conclusion, an annual hunting intensity of 70% should be maintained to effectively control ASF.

Forests are essential for sustaining ecosystems, regulating the climate, and providing economic benefits to human society. However, activities such as commercial practices, fuelwood collection, and land use changes have resulted in severe forest degradation and deforestation. Timor-Leste, a small island nation, faces environmental sustainability challenges due to land use changes, limited infrastructure, and agricultural practices. This study proposes a simplified and highly accessible approach to assess deforestation (SHAD) nationally using limited human and non-human resources such as experts, software, and hardware facilities. To assess deforestation in developing countries, we utilize open-source software (Dryad), employ the U-Net deep learning algorithm, and utilize open-source data generated from the Google Earth Engine platform to construct a time-series land cover classification model for Timor-Leste. In addition, we utilize the open-source land cover map as label data and satellite imagery as model training inputs, and our model demonstrates satisfactory performance in classifying time-series land cover. Next, we classify the land cover in Timor-Leste for 2016 and 2021, and verified that the forest classification achieved high accuracy ranging from 0.79 to 0.89. Thereafter, we produced a deforestation map by comparing the two land cover maps. The estimated deforestation rate was 1.9% annually with a primary concentration in the northwestern municipalities of Timor-Leste with dense population and human activities. This study demonstrates the potential of the SHAD approach to assess deforestation nationwide, particularly in countries with limited scientific experts and infrastructure. We anticipate that our study will support the development of management strategies for ecosystem sustainability, climate adaptation, and the conservation of economic benefits in various fields.

Lithobates catesbeianus (American bullfrog), known to be one of the notorious invasive species, was introduced to South Korea and has proliferated in the Korean natural environment for the past 25 years. The ecological impact caused by the species is well known, and several management decisions have been implemented to cull its population. However, the effectiveness of past control decisions is largely unknown. We built a population dynamics model for L. catesbeianus in the Onseok reservoir, South Korea, using STELLA architect software. The population model was based on the demographics and ecological process of the species developing through several life stages, with respective parameters for survivorship and carrying capacity. Control scenarios with varying intensities were simulated to evaluate their effectiveness. The limitations of isolated control methods and the importance of integrated management are shown in our results. The population of the American bullfrog in the reservoir was reduced to a manageable level under intensive control of the tadpole stage, using three sets of double fyke nets and 80% direct removal of juvenile and adult stages. According to our results, integrated, intensive, and continuous control is essential for managing the invasive American bullfrog population. Finally, our modeling approach can assist in determining the control intensity to improve the efficiency of measures against L. catesbeianus.

African swine fever (ASF) is a highly contagious viral disease affecting both domestic and wild boars. Since its first outbreak in South Korea in 2019, substantial efforts have been made to prevent ASF transmission by reducing the wild boar population and eliminating infected carcasses; however, the persistence of ASF transmission has posed challenges to these efforts. To improve ASF management strategies, the limitations of current management strategies must be identified by considering disparities between wild boar habitats and ASF-managed areas with environmental and anthropogenic characteristics of wild boars and their management strategies. Here, ensemble species distribution models were used to estimate wild boar habitats and potential ASF-managed areas, with elevation, distance to urban areas, and Normalized Difference Vegetation Index as important variables. Binary maps of wild boar habitats and potential ASF-managed areas were generated using the maxSSS as the threshold criterion. Disparity areas of ASF management were identified by overlying regions evaluated as wild boar habitats with those not classified as ASF-managed areas. Dense forests near urban regions like Chungcheongbuk-do, Gyeongsangbuk-do, and Gyeongsangnam-do were evaluated as disparity areas having high risk of ASF transmission. These findings hold significant potential for refining ASF management strategies and establishing proactive control measures.

Wildfires are a major ecological disturbance in Southern California and often lead to great destruction along the Wildland-Urban Interface. Live fuel moisture has been used as an important indicator of wildfire risk in measurements of vegetation water content. However, the limited field measurements of live fuel moisture in both time and space have affected the accuracy of wildfire risk estimations. Traditional estimation of live fuel moisture using remote sensing data was based on vegetation indices, indirect proxies of vegetation water content and subject to influence from weather conditions. In this study, we investigated the feasibility of estimating live fuel moisture using vegetation indices, Soil Moisture Active Passive L-band soil moisture data and the modeled vegetation water content using a non-linear model based on VIs and the stem factor associated with remote sensing moisture data products. The stem factor describes the peak amount of water residing in stems of plants and varies by land cover. We also compared the outcomes from regression models and recurrent neural network using the same independent variables. We found the modeled vegetation water content outperformed vegetation indices and the L-band soil moisture observations, suggesting a nonlinear relationship between live fuel moisture and the remotely sensed vegetation signatures. We discuss our results which will improve the predictability of live fuel moisture.