An overview of remote monitoring methods in biodiversity conservation

Dissolved organic carbon estimation in lakes: Improving machine learning with data augmentation on fusion of multi-sensor remote sensing observations

This paper presents a novel approach for estimating Dissolved Organic Carbon (DOC) concentrations in lakes considering both carbon sources and sink operators. Despite the critical role of DOC, the combined application of machine learning, as a robust predictor, and remote sensing technology, which reduces costly and time-intensive in-situ sampling, has been underexplored in DOC research. Focusing on lakes over the states of New York, Vermont and Maine (United States, U.S.), this study integrates in-situ DOC measurements with surface reflectance bands obtained from Landsat satellites between 2000 and 2020. Using these bands as inputs of the Random Forest (RF) predictive model, the introduced methodology aims to explore the ability of remote sensing data for large-scale DOC simulation. Initial results indicate low accuracy metrics and significant under-estimation due to the imbalance distribution of DOC samples. Statistical analysis showed that the mean DOC concentration was 5.37±3.37 mg/L (mean±one standard deviation), with peak up to 25 mg/L. A highly skewed distribution of chemical components towards the lower ranges can lead to model misrepresentation of extreme and hazardous events, as they are clouded by unimportant events due to significantly lower occurrence rates. To address this issue, the Synthetic Minority Over-sampling Technique (SMOTE) was applied as a key innovation, generating synthetic samples that enhance RF accuracy and reduce the associated errors. Fusion of in-situ and remote sensing data, combined with machine learning and data augmentation, significantly enhances DOC estimation accuracy, especially in high concentration ranges which are critical for environmental health. With prediction metrics of RMSE = 1.75, MAE = 1.09, and R2 = 0.74, RF-SMOTE significantly improve the metrics obtained from stand-alone RF, particularly in estimating high DOC concentrations. Considering the product spatial resolution of 30 m, the model's output provides potential revenue for global application in lake monitoring, even in remote regions where direct sampling is limited. This novel fusion of remote sensing, machine learning and data augmentation offers valuable insights for water quality management and understanding carbon cycling in aquatic ecosystems.

Pond Water eDNA Reflects Broad Consistency with Surrounding Terrestrial Plant Ecosystems

This study evaluates the potential of using pond water eDNA to reflect the surrounding terrestrial plant communities, aiming to develop a sustainable, large-scale, and long-term monitoring method for plant diversity in forest ecosystems. Water samples were collected four times from two ponds with different vegetation types during the late spring to autumn seasons in Japan. eDNA was extracted from dissolved particles fractionated by sequential filtration through pore sizes of 200 µm, 5 µm, and 0.45 µm, followed by high-throughput amplicon sequencing targeting the plant rbcL gene. By comparing field surveys with the eDNA data, we identified 79% and 63% of plant families and genera, respectively, suggesting that pond water eDNA may reflect the surrounding terrestrial plant ecosystem. Additionally, different trends were observed in the seasonal variation of plant taxa and their composition detected in eDNA, based on particle size. This study highlights the potential of pond water eDNA to provide valuable insights into forest plant richness and seasonal dynamics, offering a novel approach for ecological monitoring.

Genetic diversity of Anadara tuberculosa in two localities of the Colombian Pacific Coast

Piangua, Anadara tuberculosa, is an economically important mollusk for the human population living on the Colombian Pacific Coast. In the last years, the demand and exploitation of this mollusk have increased, putting it at risk to the point of being endangered. This research aimed to identify the genetic diversity and population structure of piangua in two localities on the Pacific Coast of Colombia. We assembled a chromosome-level genome using PacBio-Hifi and Arima sequencing. We obtained 274 scaffolds with an N50 of 45.42 Mbp, a total size of 953 Mbp, and a completeness of 91% based on BUSCO scores. The transposable elements accounted for 30.29% of the genome, and 24,317 genes were annotated. Genome-guided variant calling for 89 samples using DArT sequencing data delivered 4,825 bi-allelic SNPs, which supported genetic diversity and population structure analyses. Data showed that the piangua populations in the two localities were under expansion events more than 100k years ago. However, results also showed a reduction in genetic diversity, as evidenced by the loss of heterozygosity, which may be caused by high levels of inbreeding, probably due to a recent overexploitation. Furthermore, although we evidenced gene flow between the two localities, there is also a subtle geographical population structure between the two localities and among mangroves in one of the localities. This is the first study in Colombia that provides relevant genetic information on piangua to lay the foundations for conservation strategies.

Exploring the uptake of advanced digital technologies in environmental assessment practice - Experiences from Austria and Germany

Environmental assessment (EA) evaluates the environmental impacts of proposed projects, plans or policies to inform decision making. While several studies have highlighted the potential and opportunities of digitalisation for EA, few have explored practitioners' perceptions using a mixed methods approach in order to discover concerns and risks identified by EA of novel technological approaches. In addition, this initial exploratory study examines the perception of benefits and contributions to quality and effectiveness of advanced digital approaches, such as the introduction of artificial intelligence, in EA practice. The research process was based on focus group discussions and exploratory interviews with EA consultants, environmental authorities, researchers, environmental associations and NGOs. Relevant technologies were identified from the existing scientific literature and their applicability, benefits and use were discussed in context of real-world experience made by the practitioner. It became evident that the majority of practitioners in the field of EA in Austria and Germany are not familiar with advanced digital approaches and tools. While other planning disciplines are exploiting the potential of advanced digital tools, EA practitioners still share concerns about data quality, security, legal uncertainties, but also skills and know-how. The study identifies a gap and a need for training and confidence building. It aims to contribute to the promotion of inter- & transdisciplinary exchange involving the wider EA community.

Forest fragmentation causes an isolated population of the golden takin (Budorcas taxicolor bedfordi Thomas, 1911) (Artiodactyla: Bovidae) in the Qinling Mountains (China)

Budorcas taxicolor bedfordi is a rare animal uniquely distributed in the Qinling Mountains (China). Human disturbance and habitat fragmentation have directly affected the survival of B. t. bedfordi. It is urgent to clarify the genetic diversity and genetic structure of the B. t. bedfordi population and implement effective conservation measures. In this study, 20 new polymorphic microsatellite loci were isolated by Illumina sequencing. The genetic diversity and population structure of 124 B. t. bedfordi individuals from three populations (Niubeliang population, Zhouzhi population, and Foping population) were analysed according to these 20 microsatellite loci. Our results indicated that B. t. bedfordi had a low level of genetic variability and that there was inbreeding in the three populations. The population genetic structure analyses showed that the Niubeliang population had a trend of differentiation from other populations. National roads can affect population dispersal, while ecological corridors can promote population gene exchange. None of the three B. t. bedfordi populations experienced bottleneck effects. For conservation management plans, the Zhouzhi population and Foping population should be considered one management unit, and the Niubeliang population should be considered another management unit. We suggest building an ecological corridor to keep the habitat connected and formulating tourism management measures to reduce the influence of human disturbance on B. t. bedfordi.

Advances and applications of machine learning and deep learning in environmental ecology and health

Machine learning (ML) and deep learning (DL) possess excellent advantages in data analysis (e.g., feature extraction, clustering, classification, regression, image recognition and prediction) and risk assessment and management in environmental ecology and health (EEH). Considering the rapid growth and increasing complexity of data in EEH, it is of significance to summarize recent advances and applications of ML and DL in EEH. This review summarized the basic processes and fundamental algorithms of the ML and DL modeling, and indicated the urgent needs of ML and DL in EEH. Recent research hotspots such as environmental ecology and restoration, environmental fate of new pollutants, chemical exposures and risks, chemical hazard identification and control were highlighted. Various applications of ML and DL in EEH demonstrate their versatility and technological revolution, and present some challenges. The perspective of ML and DL in EEH were further outlined to promote the innovative analysis and cultivation of the ML-driven research paradigm.