Machine learning prediction of tree species diversity using forest structure and environmental factors: a case study from the Hyrcanian forest, Iran.

The Hyrcanian forest is a global biodiversity hotspot that harbors many endemic and endangered tree species, but its tree diversity is threatened by various human-induced disturbances, such as logging, grazing, and urbanization. To address this issue, we conducted a study using three machine learning methods, i.e., linear regression (LR), random forest (RF), and support vector machine (SVM), to assess and predict tree species diversity within the forest. To do so, we collected an extensive dataset of forest structure and environmental factors from 2725 sample plots located throughout the forest. The Shannon-Wiener diversity index was used to quantify the tree species diversity for each plot. We found that basal area, tree density, and height of trees were the most important predictors of tree diversity, followed by diameter at breast height, elevation, slope, and aspect. We measured the performance of the models using the coefficient of determination (R2), root mean square error (RMSE), and percent of relative error index (PREI), and found RF as the best-performing model in both the training (RMSE = 0.143, R2 = 0.94, and PREI = - 0.09) and validation (RMSE = 0.15, R2 = 0.94, and PREI = - 0.09) phases. RF was able to generalize effectively to new data without losing much accuracy or explanatory power. SVM demonstrated a moderate performance training (training phase: RMSE = 0.23, R2 = 0.57, and PREI = - 0.17) and (validation phase: RMSE = 0.36, R2 = 0.34, and PREI = - 0.21) among the models, while LR performed the worst (training phase: RMSE = 0.41, R2 = 0.13, and PREI = - 0.19) and (validation phase: RMSE = 0.41, R2 = 0.11, and PREI = - 0.36). These findings have broad applications beyond this specific region and can contribute to promoting sustainable land use practices and conservation efforts in other ecosystems facing similar challenges.

Future climate-driven drought events across Lake Urmia, Iran.

Climate change has increased the vulnerability of arid and semi-arid regions to recurrent and prolonged meteorological droughts. In light of this, our study has sought to assess the nature of future meteorological drought in Lake Urmia basin, Iran, within the context of future climate projections. To achieve this, data from 54 general circulation models (GCMs) was calibrated against both in situ and Global Precipitation Climatology Centre datasets. These GCMs were then employed to project drought conditions expected over 2016-2046 under RCP2.6 and RCP8.5 as the most optimistic and pessimistic scenarios, respectively. To provide a comprehensive analysis, these RCPs were combined with two different time scale Standardized Precipitation Index (SPI), leading to eight different scenarios. The SPI was calculated over two temporal scales for the past (1985-2015) and future (2016-2046), including the medium-term (SPI-6) and long-term (SPI-18) index. Results showed that while precipitation is expected to increase by up to 34%, parts of the basin are projected to face severe and prolonged droughts under both RCPs. The most severe drought event is expected to occur around 2045-2046 under the most pessimistic RCP8.5 scenario. Severe droughts with low frequency are also anticipated to increase under other scenarios. By characterizing meteorological drought conditions for Lake Urmia basin under future climate conditions, our findings call for urgent action for adaptation strategies to mitigate the future adverse effects of drought in this region and other regions facing similar challenges. Overall, this study provides valuable insight into the impacts of climate change on future droughts that can adversely influence water resources in arid and semi-arid regions.

A Robust Deep-Learning Model for Landslide Susceptibility Mapping: A Case Study of Kurdistan Province, Iran.

We mapped landslide susceptibility in Kamyaran city of Kurdistan Province, Iran, using a robust deep-learning (DP) model based on a combination of extreme learning machine (ELM), deep belief network (DBN), back propagation (BP), and genetic algorithm (GA). A total of 118 landslide locations were recorded and divided in the training and testing datasets. We selected 25 conditioning factors, and of these, we specified the most important ones by an information gain ratio (IGR) technique. We assessed the performance of the DP model using statistical measures including sensitivity, specificity, accuracy, F1-measure, and area under-the-receiver operating characteristic curve (AUC). Three benchmark algorithms, i.e., support vector machine (SVM), REPTree, and NBTree, were used to check the applicability of the proposed model. The results by IGR concluded that of the 25 conditioning factors, only 16 factors were important for our modeling procedure, and of these, distance to road, road density, lithology and land use were the four most significant factors. Results based on the testing dataset revealed that the DP model had the highest accuracy (0.926) of the compared algorithms, followed by NBTree (0.917), REPTree (0.903), and SVM (0.894). The landslide susceptibility maps prepared from the DP model with AUC = 0.870 performed the best. We consider the DP model a suitable tool for landslide susceptibility mapping.

Understanding land degradation induced by gully erosion from the perspective of different geoenvironmental factors.

Complex interrelationships between landscape-level geoenvironmental factors and natural phenomena have rendered land degradation control measures ineffective. For control to be effective, this study argues that the interactions between different geoenvironmental factors and gully erosion (as an indicator of land degradation) should be more fully investigated and spatially mapped. To do so, gully locations of the Konduran watershed, Iran, were detected in the field and modeled in response to seventeen geoenvironmental factors using three machine learning methods, i.e., multivariate adaptive regression splines (MARS), random forest (RF), regularized random forest (RRF), and Bayesian generalized linear model (Bayesian GLM). The models' performance was validated, the relationship of gully occurrence with each factor was quantified, the probability of gully erosion (i.e., land degradation) was retrospectively estimated, and the spatially explicit maps of land degradation susceptibility were produced. Based on the area under the receiver operating characteristic curve (AUC), the RRF and MARS models with AUC = 0.98 achieved the greatest goodness-of-fit with the training dataset, whereas the RF model with AUC = 0.83 showed the greatest ability in predicting future gully occurrences. Further scrutinization using the sensitivity and specificity metrics demonstrated the efficiency of the RF model for correctly classifying the gully (sensitivity-training = 92%; sensitivity-validation = 90%) and non-gully (specificity-training = 95%; specificity-validation = 68%) pixels. Nearly 13% of the study area ended up being the hardest hit region due to their general characteristics of distance from roads and rives, altitude, and normalized difference vegetation index (NDVI) that were identified as the most influential factors in gully erosion occurrence. Given the resolution quality and reliable predictive accuracy, our spatially explicit maps of land susceptibility to gully erosion can be used by authorities and urban planners for identifying the target areas for rehabilitation and making more informed decisions for infrastructure development. Although our study was strictly focused on a certain region, our recommendations and implications are of global significance.

Fuzzy-based vulnerability assessment of coupled social-ecological systems to multiple environmental hazards and climate change.

Climate change and combining related parameters of environmental hazards have left a considerable challenge in assessing social-ecological vulnerability. Here we integrated a fuzzy-based approach in the vulnerability assessment of mangrove social-ecological systems combining environmental parameters, socio-economic, and vegetative components from exposure dimensions, sensitivity and adaptive capacity along the northern coasts of the Persian Gulf and the Gulf of Oman for the first time. This study aims to provide critical information for habitat-scale management strategies and adaptation plans by assessing the vulnerability of mangrove social-ecological systems. This study provides a methodology framework that consists of five steps. Step 1: We combined the fuzzy weighted maps of seven environmental hazards, including tidal range, maximum wind speeds, drought magnitude, maximum temperatures, extreme storm surge, sea-level rise, significant wave height, and social vulnerability. This map combination determined that the computed exposure index is from 1.07 to 4.32 across the study areas, with an increasing trend from the coasts of the Persian Gulf to the Gulf of Oman. Step 2: We integrated the fuzzy weighted maps of four sensitivity variables, including area change, health change, seaward edge retreat, and production potential change. The findings show that the sensitivity index is from 1.40 to 2.64 across the study areas, increasing the trend from the Persian Gulf coast to the Gulf of Oman. Step 3: Besides, we combined the fuzzy weighted maps of three adaptive capacity variables, including the availability of migration areas, recruitment, and local communities' participation in restoration projects and education programs. The result showed that the index value across the study areas varies between 0.087 and 2.38, decreasing the trend from the Persian Gulf coast to the Gulf of Oman. Step 4: Implementing fuzzy hierarchical analysis process to determine the relative weight of variables corresponding to exposure, sensitivity and adaptive capacity. Step 5: The integration of exposure, sensitivity and adaptive capacity and the vulnerability index maps in the study areas showed variation from 0.25 to 5.92, with the vulnerability of mangroves from the west coast of the Persian Gulf (Nayband) decreasing towards Khamir, then increasing to the eastern coasts of the Gulf of Oman (Jask and Gwadar). Overall, the results indicate the importance of the proposed approach to the vulnerability of mangroves at the habitat scale along a coastal area and across environmental gradients of climatic, maritime and socio-economic variables. This study validated the findings based on the ground truth measurements, and high-resolution satellite data incorporated the Consistency Rate (CR) in the Fuzzy Analytic Hierarchy Process (FAHP). The overall accuracy of all classified remote sensing images and maps consistently exceeded 90%, and the CR of the 25 completed questionnaires was <0.1. Finally, this study indicates differences in vulnerability of various habitats, leading to focus conservation completion and rehabilitation and climate change adaptation planning to support the Sustainable Development Goal (SDG)-13 implementation.

Genetic and firefly metaheuristic algorithms for an optimized neuro-fuzzy prediction modeling of wildfire probability.

In the terrestrial ecosystems, perennial challenges of increased frequency and intensity of wildfires are exacerbated by climate change and unplanned human activities. Development of robust management and suppression plans requires accurate estimates of future burn probabilities. This study describes the development and validation of two hybrid intelligence predictive models that rely on an adaptive neuro-fuzzy inference system (ANFIS) and two metaheuristic optimization algorithms, i.e., genetic algorithm (GA) and firefly algorithm (FA), for the spatially explicit prediction of wildfire probabilities. A suite of ten explanatory variables (altitude, slope, aspect, land use, rainfall, soil order, temperature, wind effect, and distance to roads and human settlements) was investigated and a spatial database constructed using 32 fire events from the Zagros ecoregion (Iran). The frequency ratio model was used to assign weights to each class of variables that depended on the strength of the spatial association between each class and the probability of wildfire occurrence. The weights were then used for training the ANFIS-GA and ANFIS-FA hybrid models. The models were validated using the ROC-AUC method that indicated that the ANFIS-GA model performed better (AUCsuccessrate = 0.92; AUCpredictionrate = 0.91) than the ANFIS-FA model (AUCsuccessrate = 0.89; AUCpredictionrate = 0.88). The efficiency of these models was compared to a single ANFIS model and statistical analyses of paired comparisons revealed that the two meta-optimized predictive models significantly improved wildfire prediction accuracy compared to the single ANFIS model (AUCsuccessrate = 0.82; AUCpredictionrate = 0.78). We concluded that such predictive models may become valuable toolkits to effectively guide fire management plans and on-the-ground decisions on firefighting strategies.

Multi-hazards vulnerability assessment of southern coasts of Iran.

Coastal vulnerability assessment has become one of the most important tools for decision making and providing effective managerial solutions to reduce adverse socio-economic impacts of multiple environmental hazards on coupled social-ecological systems of coastal areas. The aim of this study was to assess the vulnerability of the northern coasts of the Persian Gulf (PG) and the Gulf of Oman (GO) in the Hormozgan province of Iran. Nine variables of vulnerability that included the rate of coastline change, relative sea level rise, coastal slope, mean tidal range, coastal geomorphology, significant wave height (SWH), extreme storm surge, population density, and fishing intensity were weighted, mapped, and combined into the Coastal vulnerability index (CVI). Experts viewed sea level rise, shoreline change and extreme storm surge as most important for imparting vulnerabilities on the northern coasts of PG and GO. Socio-economic variables (i.e., population density and fishery intensity) were considered least important. Of the total length of the provincial shoreline, 27% were classified into the very low vulnerability class, 31% into the low, 17.4% into the moderate, 15.4% into the high, and 9.2% into the very high vulnerability class. About 1295 km (58%) of shorelines were classified into the low and very low vulnerability classes (CVI value ≤ 8.32) and mainly consisted of shorelines on the western coast along the PG. In contrast, 553 km (24.6%) of shorelines were classified into the high and very high vulnerability classes (CVI values > 13.39) and were located along the central coasts (especially in the Qeshm Island and Strait of Hormuz) and on the east coasts of the GO. At least a quarter of all shorelines in the province have high and very high vulnerability to environmental hazards that are the harbingers of climate change.

Ensemble models based on radial basis function network for landslide susceptibility mapping.

Ensemble learning techniques have shown promise in improving the accuracy of landslide models by combining multiple models to achieve better predictive performance. In this study, several ensemble methods (Dagging, Bagging, and Decorate) and a radial basis function classifier (RBFC) were combined to predict landslide susceptibility in the Trung Khanh district of the Cao Bang Province, Vietnam. The ensemble models were developed using a geospatial database containing 45 historical landslides (1074 points) and thirteen influencing variables characterizing the topography, geology, land use/cover, and human activities of the study area. The performance of the models was evaluated based on the area under the receiver operating characteristic curve (AUC) and several other performance metrics, including positive predictive value (PPV), negative predictive value (NPV), sensitivity (SST), specificity (SPF), accuracy (ACC), and root mean square error (RMSE). The Bagging-RBFC model with PPV = 86%, NPV = 95%, SST = 95%, SPF = 87%, ACC = 91%, RMSE = 0.297, and AUC = 98% was found to be the most accurate model for the prediction of landslide susceptibility, followed by the Dagging-RBFC, Decorate-RBFC, and single RBFC models. The study demonstrates the efficacy of ensemble learning techniques in developing reliable landslide predictive models, which can ultimately save lives and reduce infrastructure damage in landslide-prone regions worldwide.

Spatial modeling of exposure of mangrove ecosystems to multiple environmental hazards.

Determining the level of ecosystems exposure to multiple environmental hazards or risk factors is of paramount importance for developing, adopting, and planning management strategies to minimize the harmful effects of these hazards. We quantified the level of exposure of mangroves on the northern coasts of the Persian Gulf (PG) and the Gulf of Oman (GO) between 1986 and 2019 to eight environmental hazards, i.e., drought, maximum temperatures, rising sea levels, change of freshwater inflows to coasts, extreme storm surges, significant wave height (SWH), seaward edge retreat in the mangroves, and fishery intensity. Based on expert opinion, fuzzy weights were used to integrate these exposures into a single index (EI) for the region. Experts gave the greatest weight/importance to the risks posed by sea-level rise and seaward retreat of mangroves and the lowest risk to significant wave height and fishery intensity in coastal waters. The overall EI and six of eight individual variables (except fishery intensity and maximum temperatures) pointed to exposure levels of mangroves that increased from the coasts of the PG (EI 0.69) to the GO (EI 6.69). Since these hazards are expected to continue in the future, local/regional management responses should focus on minimizing regional anthropogenic threats and halt conversion of natural areas to agricultural and open areas to maintain freshwater inputs to coastal areas, particularly on the GO. Further, uplands that may serve as future refugia into which mangroves may expand over time as sea levels continue to rise should be protected from development. This was the first study that used an analytic framework to compute a mangrove exposure index to a suite of physical and socio-economic hazards across a region. This framework may provide insights into cost-effective resilience-based design and management of socio-ecologically coupled ecosystems in an era of increasing types and intensities of environmental hazards.

Vulnerability of coastal communities to climate change: Thirty-year trend analysis and prospective prediction for the coastal regions of the Persian Gulf and Gulf of Oman.

This study relates changes in social vulnerability of 20 counties on the northern coasts of the Persian Gulf (PG) and the Gulf of Oman (GO) over a 30-year period (1988-2017) to changing socio-economic conditions and environmental (climate) hazard. Social vulnerability in 2030, 2040 and 2050 is predicted based on the RCP8.5 climate change scenario that projects drought intensities and rising sea levels. Social vulnerability was based on the three dimensions of sensitivity, exposure, and adaptive capacity using 18 socio-economic and five climate indicators identified by experts. All but one indicator related very strongly to the dimension it sought to represent. Despite improvements in adaptive capacity over time, social vulnerability increased between 1988 and 2017 and rates of change accelerated after change point years that occurred between 1998 and 2002 in most counties. Extrapolating past changes of each indicator over time enabled forecasts of social vulnerability in the future. While social variability decreased between 2017 and 2030, it increased again between 2030 and 2050. The lowest future social vulnerability is expected along the eastern PG coast, the greatest along the western PG and the GO. The worsening of socio-economic indicators contributed to increased sensitivity, and increased drought intensities plus the expected rise in sea levels will lead to social vulnerabilities in 2050 comparable to present levels. Between 1.4 and 1.7 M people will live in areas that are likely submerged by water in the future. About 80% of these people live in six counties with variable social vulnerabilities. While counties with lower social variabilities might be better able to cope with the challenges posed by climate change, adaptation programs to enhance the resilience of the residents in these and the remaining counties along the PG and the GO need to be implemented soon to avoid uncontrolled mass migration of millions of people from the region.

Groundwater Potential Mapping Combining Artificial Neural Network and Real AdaBoost Ensemble Technique: The DakNong Province Case-study, Vietnam.

: The main aim of this study is to assess groundwater potential of the DakNong province, Vietnam, using an advanced ensemble machine learning model (RABANN) that integrates Artificial Neural Networks (ANN) with RealAdaBoost (RAB) ensemble technique. For this study, twelve conditioning factors and wells yield data was used to create the training and testing datasets for the development and validation of the ensemble RABANN model. Area Under the Receiver Operating Characteristic (ROC) curve (AUC) and several statistical performance measures were used to validate and compare performance of the ensemble RABANN model with the single ANN model. Results of the model studies showed that both models performed well in the training phase of assessing groundwater potential (AUC ≥ 0.7), whereas the ensemble model (AUC = 0.776) outperformed the single ANN model (AUC = 0.699) in the validation phase. This demonstrated that the RAB ensemble technique was successful in improving the performance of the single ANN model. By making minor adjustment in the input data, the ensemble developed model can be adapted for groundwater potential mapping of other regions and countries toward more efficient water resource management. The present study would be helpful in improving the groundwater condition of the area thus in solving water borne disease related health problem of the population.

Monitoring and Assessment of Water Level Fluctuations of the Lake Urmia and Its Environmental Consequences Using Multitemporal Landsat 7 ETM+ Images.

The declining water level in Lake Urmia has become a significant issue for Iranian policy and decision makers. This lake has been experiencing an abrupt decrease in water level and is at real risk of becoming a complete saline land. Because of its position, assessment of changes in the Lake Urmia is essential. This study aims to evaluate changes in the water level of Lake Urmia using the space-borne remote sensing and GIS techniques. Therefore, multispectral Landsat 7 ETM+ images for the years 2000, 2010, and 2017 were acquired. In addition, precipitation and temperature data for 31 years between 1986 and 2017 were collected for further analysis. Results indicate that the increased temperature (by 19%), decreased rainfall of about 62%, and excessive damming in the Urmia Basin along with mismanagement of water resources are the key factors in the declining water level of Lake Urmia. Furthermore, the current research predicts the potential environmental crisis as the result of the lake shrinking and suggests a few possible alternatives. The insights provided by this study can be beneficial for environmentalists and related organizations working on this and similar topics.

Landslide Susceptibility Mapping Using Machine Learning Algorithms and Remote Sensing Data in a Tropical Environment.

We used AdaBoost (AB), alternating decision tree (ADTree), and their combination as an ensemble model (AB-ADTree) to spatially predict landslides in the Cameron Highlands, Malaysia. The models were trained with a database of 152 landslides compiled using Synthetic Aperture Radar Interferometry, Google Earth images, and field surveys, and 17 conditioning factors (slope, aspect, elevation, distance to road, distance to river, proximity to fault, road density, river density, normalized difference vegetation index, rainfall, land cover, lithology, soil types, curvature, profile curvature, stream power index, and topographic wetness index). We carried out the validation process using the area under the receiver operating characteristic curve (AUC) and several parametric and non-parametric performance metrics, including positive predictive value, negative predictive value, sensitivity, specificity, accuracy, root mean square error, and the Friedman and Wilcoxon sign rank tests. The AB model (AUC = 0.96) performed better than the ensemble AB-ADTree model (AUC = 0.94) and successfully outperformed the ADTree model (AUC = 0.59) in predicting landslide susceptibility. Our findings provide insights into the development of more efficient and accurate landslide predictive models that can be used by decision makers and land-use managers to mitigate landslide hazards.

Shallow Landslide Susceptibility Mapping: A Comparison between Logistic Model Tree, Logistic Regression, Naïve Bayes Tree, Artificial Neural Network, and Support Vector Machine Algorithms.

Shallow landslides damage buildings and other infrastructure, disrupt agriculture practices, and can cause social upheaval and loss of life. As a result, many scientists study the phenomenon, and some of them have focused on producing landslide susceptibility maps that can be used by land-use managers to reduce injury and damage. This paper contributes to this effort by comparing the power and effectiveness of five machine learning, benchmark algorithms-Logistic Model Tree, Logistic Regression, Naïve Bayes Tree, Artificial Neural Network, and Support Vector Machine-in creating a reliable shallow landslide susceptibility map for Bijar City in Kurdistan province, Iran. Twenty conditioning factors were applied to 111 shallow landslides and tested using the One-R attribute evaluation (ORAE) technique for modeling and validation processes. The performance of the models was assessed by statistical-based indexes including sensitivity, specificity, accuracy, mean absolute error (MAE), root mean square error (RMSE), and area under the receiver operatic characteristic curve (AUC). Results indicate that all the five machine learning models performed well for shallow landslide susceptibility assessment, but the Logistic Model Tree model (AUC = 0.932) had the highest goodness-of-fit and prediction accuracy, followed by the Logistic Regression (AUC = 0.932), Naïve Bayes Tree (AUC = 0.864), ANN (AUC = 0.860), and Support Vector Machine (AUC = 0.834) models. Therefore, we recommend the use of the Logistic Model Tree model in shallow landslide mapping programs in semi-arid regions to help decision makers, planners, land-use managers, and government agencies mitigate the hazard and risk.

Modeling multi-decadal mangrove leaf area index in response to drought along the semi-arid southern coasts of Iran.

Leaf Area Index (LAI; as an indicator of the health) of the mangrove ecosystems on the northern coasts of the Persian Gulf and the Gulf of Oman was measured in the field and modeled in response to observed (1986-2017) and predicted (2018-2100) drought occurrences (quantified using the Standardized Precipitation Index [SPI]). The relationship of LAI with the normalized difference vegetation index (NDVI) obtained from satellite images was quantified, the LAI between 1986 and 2017 retrospectively estimated, and a relationship between LAI and SPI developed for the same period. Long-term climate data were used as input in the RCP8.5 climate change scenario to reconstruct recent and forecast future drought intensities. Both the NDVI and the SPI were strongly related with the LAI, indicating that realistic LAI values were derived from historic satellite data to portray annual changes of LAI in response to changes in SPI. Our findings show that projected future drought intensities modeled by the RCP8.5 scenario increase more and future LAIs decreased more on the coasts of the Gulf of Oman than the coasts of the Persian Gulf in the coming decades. The year 1998 was the most significant change-point for mean annual rainfall amounts and drought occurrences as well as for LAIs and at no time between 1998 and 2017 or between 2018 and 2100 are SPI and LAI values expected to return to pre-1998 values. LAI and SPI are projected to decline sharply around 2030, reach their lowest levels between 2040 and 2070, and increase and stabilize during the late decades of the 21st century at values similar to the present time. Overall, this study provides a comprehensive picture of the responses of mangroves to fluctuating future drought conditions, facilitating the development of management plans for these vulnerable habitats in the face of future climate change.