Predicting Habitat Suitability and Conserving Juniperus spp. Habitat Using SVM and Maximum Entropy Machine Learning Techniques

Deteriorating wintertime habitat conditions for waterfowls in Caizi Lake, China: Drivers and adaptive measures

China has made enormous strides to achieve high-quality development and biodiversity conservation, and the establishment of nature-protected areas is one of the essential initiatives. Caizi Lake involves a natural reserve and two national wetland parks, accommodating winter migratory waterfowl over the middle and lower Yangtze River basin in China. However, the water transfer from the Yangtze River to the Huai River (YR-HR water transfer) has modified the winter hydrological conditions of Caizi Lake, negatively affecting wintertime waterfowl habitats. Hence, conserving wintertime waterfowl habitats necessitates knowledge of the dynamical mechanisms behind the impacts of YR-HR water transfer on wintertime waterfowl habitats and adaptive measures. Here we developed a machine learning model, the normalized difference vegetation index, and on-spot observatory datasets such as the spatial distribution of waterfowl species and underwater topography of Caizi Lake. We found that the rising winter water level of Caizi Lake encroaches on winter waterfowl habitat with extremely high suitability. Meanwhile, rising water levels reduced waterfowl food sources. Thus, rising water levels due to YR-HR water transfer deteriorated waterfowl living conditions over Caizi Lake. Therefore, we proposed adaptive measures to alleviate these negative effects, such as water level regulation, artificial feeding of waterfowls, restoration and reconstruction of contiguous mudflats, grass flats. This study highlights human interferences with waterfowl habitats, necessitating biodiversity conservation at regional scales.

Assessment of wetland ecosystem health in Rarh Region, India through P-S-R (pressure-state-response) model

The current study attempted to assess wetland ecosystem health (EH) in the Murshidabad district's Rarh tract using the P-S-R (Pressure-State-Response) model and machine learning (ML) algorithms and validated it with a field-based validation approach as well as conventional validation approaches. To assess the ecosystem's health, 27 metrics were used to monitor the wetlands' pressure, state, and response. All of the models found that 46.1 % of wetlands in strong EH zones have transformed to 11.41 % in relatively fragile EH zones during the previous thirty years, demonstrating a progressive loss of EH quality throughout larger wetland areas. All of the applied models were deemed to be acceptable based on the results of the model validation process, however, the Random Forest (RF) model performed exceptionally well. The deterioration of EH in the wetlands happened due to the rapid expansion of settlement areas and agricultural land. So, the findings of the study deepen our knowledge about EH in the Rarh tract's wetlands, assisting decision-makers in creating sustainable wetland management strategies.

Modeling climate-related global risk maps of rice bacterial blight caused by Xanthomonas oryzae (Ishiyama 1922) using geographical information system (GIS)

Rice is a critical staple crop that feeds more than half of the world's population. Still, its production confronts various biotic risks, notably the severe bacterial blight disease produced by Xanthomonas oryzae. Understanding the possible effects of climate change on the geographic distribution of this virus is critical to ensuring food security. This work used ecological niche modeling and the Maxent algorithm to create future risk maps for the range of X. oryzae under several climate change scenarios between 2050 and 2070. The model was trained using 93 occurrence records of X. oryzae and five critical bioclimatic variables. It has an excellent predictive performance, with an AUC of 0.889. The results show that X. oryzae's potential geographic range and habitat suitability are expected to increase significantly under low (RCP2.6) and high (RCP8.5) emission scenarios. Key climatic drivers allowing this development include increased yearly precipitation, precipitation during the wettest quarter, and the wettest quarter's mean temperature. These findings are consistent with broader research revealing that climate change is allowing many plant diseases and other dangerous microbes to spread across the globe. Integrating these spatial predictions with data on host susceptibility, agricultural practices, and socioeconomic vulnerabilities can help to improve targeted surveillance, preventative, and management methods for reducing the growing threat of bacterial blight to rice production. Proactive, multidisciplinary efforts to manage the changing disease dynamics caused by climate change will be critical to assuring global food security in the future decades.

Assessing the Potential Distribution of Oxalis latifolia, a Rapidly Spreading Weed, in East Asia under Global Climate Change

Oxalis latifolia, a perennial herbaceous weed, is a highly invasive species that poses a threat to agricultural lands worldwide. East Asia is under a high risk of invasion of O. latifolia under global climate change. To evaluate this risk, we employed maximum entropy modeling considering two shared socio-economic pathways (SSP2-4.5 and SSP5-8.5). Currently, a small portion (8.02%) of East Asia is within the O. latifolia distribution, with the highest coverages in Chinese Taipei, China, and Japan (95.09%, 9.8%, and 0.24%, respectively). However, our projections indicated that this invasive weed will likely be introduced to South Korea and North Korea between 2041 and 2060 and 2081 and 2100, respectively. The species is expected to cover approximately 9.79% and 23.68% (SSP2-4.5) and 11.60% and 27.41% (SSP5-8.5) of the total land surface in East Asia by these time points, respectively. South Korea and Japan will be particularly susceptible, with O. latifolia potentially invading up to 80.73% of their territory by 2081-2100. Mongolia is projected to remain unaffected. This study underscores the urgent need for effective management strategies and careful planning to prevent the introduction and limit the expansion of O. latifolia in East Asian countries.

Phytochemical Analysis and Habitat Suitability Mapping of Cardiocrinum cordatum (Thunb.) Makino Collected at Chiburijima, Oki Islands, Japan

Cardiocrinum cordatum, known as ubayuri in Japan, has antihypertensive properties and has been shown to inhibit angiotensin-converting enzyme (ACE), which contributes to the production of angiotensin II, a hypotensive substance in the renin−angiotensin system. C. cordatum has been the subject of various studies as a useful plant and is applied as a functional food. Due to the limited distribution, loss of natural habitat by frequent natural disasters, and environmental conditions, the chemical content and biological activity of C. cordatum have been drastically affected. Obtaining a stable supply of Cardiocrinu cordatum material with high biological activity is still a challenge. Understanding the native habitat environment and suitable cultivation sites could help in solving this issue. Therefore, in the current study we investigated the effect of environmental parameters on the hypertensive and antioxidant activities of C. cordatum collected at Chiburijima, Oki Islands, Shimane Prefecture, Japan. We also predicted the habitat suitability of C. cordatum using a geographic information system (GIS) and MaxEnt model with various conditioning factors, including the topographic, soil, environmental, and climatic factors of the study area. A total of 37 individual plant samples along with soil data were collected for this study. In vitro assays of ACE inhibitory and antioxidant activity were conducted on the collected samples. The results show that plants at 14 out of 37 sites had very strong ACE inhibitory activity (IC50 < 1 mg mL−1). However, the collected plants showed no signs of strong antioxidant activity. Statistical analysis using analysis of variance (ANOVA) showed that BIO05 (F value = 2.93, p < 0.05), nitrate−nitrogen (F value = 2.46, p < 0.05), and silt (F value = 3.443, p < 0.05) significantly affected ACE inhibitory activity. On the other hand, organic carbon content (F value = 10.986, p < 0.01) was found to significantly affect antioxidant activity. The final habitat suitability map shows 3.3% very high and 6.8% high suitability regions, and samples with ACE inhibition activity were located within these regions. It is recommended further investigations and studies are conducted on C. cordatum in these locations. The prediction suitability model showed accuracy with AUC-ROC of 96.7% for the study area.

Global Spatial Suitability Mapping of Wind and Solar Systems Using an Explainable AI-Based Approach

An assessment of site suitability for wind and solar plants is a strategic step toward ensuring a low-cost, high-performing, and sustainable project. However, these issues are often handled on a local scale using traditional decision-making approaches that involve biased and non-generalizable weightings. This study presents a global wind and solar mapping approach based on eXplainable Artificial Intelligence (XAI). To the best of the author’s knowledge, the current study is the first attempt to create global maps for siting onshore wind and solar power systems and formulate novel weights for decision criteria. A total of 13 conditioning factors (independent variables) defined through a comprehensive literature review and multicollinearity analysis were assessed. Real-world renewable energy experiences (more than 55,000 on-site wind and solar plants worldwide) are exploited to train three machine learning (ML) algorithms, namely Random Forest (RF), Support Vector Machine (SVM), and Multi-layer Perceptron (MLP). Then, the output of ML models was explained using SHapley Additive exPlanations (SHAP). RF outperformed SVM and MLP in both wind and solar modeling with an overall accuracy of 90% and 89%, kappa coefficient of 0.79 and 0.78, and area under the curve of 0.96 and 0.95, respectively. The high and very high suitability categories accounted for 23.2% (~26.84 million km2) of the site suitability map for wind power plants. In addition, they covered more encouraging areas (24.0% and 19.4%, respectively, equivalent to ~50.31 million km2) on the global map for hosting solar energy farms. SHAP interpretations were consistent with the Gini index indicating the dominance of the weights of technical and economic factors over the spatial assessment under consideration. This study provides support to decision-makers toward sustainable power planning worldwide.

Genetic Diversity Assessment of Iranian Kentucky Bluegrass Accessions: I. ISSR Markers and Their Association with Habitat Suitability Within and Between Different Ecoregions

Poa pratensis L. is a commonly used cool-season turfgrass and endemic to Iran. This research was carried out to examine the genetic diversity of this plant within and between ecoregions of Iran and the impact of climatic variables and elevation on the distribution of its genotypes, as well as habitat suitability modeling. We used fifty accessions collected from six ecoregions (West, South, North, North-West and North-East) for genetic diversity assessment using 20 ISSR marker primers. The prospective ecoregions for Kentucky bluegrass production were projected using habitat suitability modeling, which took into account important environmental parameters, such as annual mean temperature, annual mean rainfall, and elevation. According to the UPMGA dendrogram, the accessions were divided into two major types and four subclasses. The genetic distance between the North and North-east accessions, as well as the Center accessions, was greater than that of the other genotypes. Center accessions had the greatest levels of polymorphism, effective number of alleles, Shannon index, and Nei's genetic diversity. The FR method was used to create the habitat suitability map based on environmental factors. Rainfall had the largest influence on the genotype distribution of P. pratensis L. The findings of this study can be used as raw materials in future breeding programs to improve and generate new cultivars with superior characteristics. It can also assist programs in identifying rare cultivars as well as preserving and developing native P. pratensis L. genotypes.

Endemic Juniperus Montane Species Facing Extinction Risk under Climate Change in Southwest China: Integrative Approach for Conservation Assessment and Prioritization

Simple Summary

Climate change is one of the most significant drivers of habitat loss and species extinction, particularly montane endemic species such as Juniper trees, which are restricted to unique habitats. Therefore, assessing the impact of climate change on the extinction risk of species is a promising tool or guide for species conservation planning. The loss in species habitat due to global warming indicates the level of extinction or endangerment. Predictions of suitable habitats are outputs from assessment analysis. This will help conservationists discover new populations of endemic species and help raise the awareness of local people to save and rescue these endangered species.

Abstract

Climate change is an important driver of biodiversity loss and extinction of endemic montane species. In China, three endemic Juniperus spp. (Juniperuspingii var. pingii, J.tibetica, and J.komarovii) are threatened and subjected to the risk of extinction. This study aimed to predict the potential distribution of these three Juniperus species under climate change and dispersal scenarios, to identify critical drivers explaining their potential distributions, to assess the extinction risk by estimating the loss percentage in their area of occupancy (AOO), and to identify priority areas for their conservation in China. We used ensemble modeling to evaluate the impact of climate change and project AOO. Our results revealed that the projected AOOs followed a similar trend in the three Juniperus species, which predicted an entire loss of their suitable habitats under both climate and dispersal scenarios. Temperature annual range and isothermality were the most critical key variables explaining the potential distribution of these three Juniperus species; they contribute by 16–56.1% and 20.4–38.3%, respectively. Accounting for the use of different thresholds provides a balanced approach for species distribution models’ applications in conservation assessment when the goal is to assess potential climatic suitability in new geographical areas. Therefore, south Sichuan and north Yunnan could be considered important priority conservation areas for in situ conservation and search for unknown populations of these three Juniperus species.

Assessing the susceptibility of schools to flood events in Iran

Catastrophic floods cause deaths, injuries, and property damages in communities around the world. The losses can be worse among those who are more vulnerable to exposure and this can be enhanced by communities’ vulnerabilities. People in undeveloped and developing countries, like Iran, are more vulnerable and may be more exposed to flood hazards. In this study we investigate the vulnerabilities of 1622 schools to flood hazard in Chaharmahal and Bakhtiari Province, Iran. We used four machine learning models to produce flood susceptibility maps. The analytic hierarchy process method was enhanced with distance from schools to create a school-focused flood-risk map. The results indicate that 492 rural schools and 147 urban schools are in very high-risk locations. Furthermore, 54% of rural students and 8% of urban students study schools in locations of very high flood risk. The situation should be examined very closely and mitigating actions are urgently needed.

Modelling for risk and biosecurity related to forest health

Modelling the invasion and emergence of forest pests and pathogens (PnPs) is necessary to quantify the risk levels for forest health and provide key information for policy makers. Here, we make a short review of the models used to quantify the invasion risk of exotic species and the emergence risk of native species. Regarding the invasion process, models tackle each invasion phase, e.g. pathway models to describe the risk of entry, species distribution models to describe potential establishment, and dispersal models to describe (human-assisted) spread. Concerning the emergence process, models tackle each process: spread or outbreak. Only a few spread models describe jointly dispersal, growth, and establishment capabilities of native species while some mechanistic models describe the population temporal dynamics and inference models describe the probability of outbreak. We also discuss the ways to quantify uncertainty and the role of machine learning. Overall, promising directions are to increase the models’ genericity by parameterization based on meta-analysis techniques to combine the effect of species traits and various environmental drivers. Further perspectives consist in considering the models’ interconnection, including the assessment of the economic impact and risk mitigation options, as well as the possibility of having multi-risks and the reduction in uncertainty by collecting larger fit-for-purpose datasets.

A machine learning framework for multi-hazards modeling and mapping in a mountainous area

This study sought to produce an accurate multi-hazard risk map for a mountainous region of Iran. The study area is in southwestern Iran. The region has experienced numerous extreme natural events in recent decades. This study models the probabilities of snow avalanches, landslides, wildfires, land subsidence, and floods using machine learning models that include support vector machine (SVM), boosted regression tree (BRT), and generalized linear model (GLM). Climatic, topographic, geological, social, and morphological factors were the main input variables used. The data were obtained from several sources. The accuracies of GLM, SVM, and functional discriminant analysis (FDA) models indicate that SVM is the most accurate for predicting landslides, land subsidence, and flood hazards in the study area. GLM is the best algorithm for wildfire mapping, and FDA is the most accurate model for predicting snow avalanche risk. The values of AUC (area under curve) for all five hazards using the best models are greater than 0.8, demonstrating that the model’s predictive abilities are acceptable. A machine learning approach can prove to be very useful tool for hazard management and disaster mitigation, particularly for multi-hazard modeling. The predictive maps produce valuable baselines for risk management in the study area, providing evidence to manage future human interaction with hazards.

Entropy-Based Metrics for Occupancy Detection Using Energy Demand

Smart Meters provide detailed energy consumption data and rich contextual information that can be utilized to assist electricity providers and consumers in understanding and managing energy use. The detection of human activity in residential households is a valuable extension for applications, such as home automation, demand side management, or non-intrusive load monitoring, but it usually requires the installation of dedicated sensors. In this paper, we propose and evaluate two new metrics, namely the sliding window entropy and the interval entropy, inspired by Shannon’s entropy in order to obtain information regarding human activity from smart meter readings. We emphasise on the application of the entropy and analyse the effect of input parameters, in order to lay the foundation for future work. We compare our method to other methods, including the Page–Hinkley test and geometric moving average, which have been used for occupancy detection on the same dataset by other authors. Our experimental results, using the power measurements of the publicly available ECO dataset, indicate that the accuracy and area under the curve of our method can keep up with other well-known statistical methods, stressing the practical relevance of our approach.