Machine learning models for wetland habitat vulnerability in mature Ganges delta

Transformation trajectory of wetland and suitability of migratory water bird habitat in the moribund Ganges delta

Wetland is a suitable habitat for water birds, and it enhances cultural ecosystem services. But the rapid transformation of such habitat, especially in floodplain environments, is an emerging crisis. Wetland reclamation and fragmentation are two major issues leading to poor habitat and landscape. The present paper aimed to explore the spatio-temporal changes in the suitability of wetland bird habitat, wetland landscape pattern, and the connection between them. Two wetlands, including a wetland of national importance, were taken as cases for this study. Time series Landsat and Sentinel images were taken for developing modeling parameters and Land Use Land Cover (LULC) for the years 2016 and 2020. The first transformation of wetland was accounted from the LULC maps of both years. Machine learning algorithm-based spatial models were developed for mapping the poor landscape condition of the existing wetland parts. Finally, semi-subjective analytic hierarchy approach (AHP)-based models were developed for assessing waterbird habitat suitability. Results demarcated more than 48% area belonging primarily to edges and tiny patches of wetlands under a poor state in 2020. Although the total wetland area was reduced between 2016 and 2020, the wetland area found to be highly suitable habitat increased from 25.5 to 59.44% of the total area during that period. The suitability of edge-preferring bird habitat showed a 10% increase. The increasing poverty of the landscape was caused by declining edge-preferring bird habitat suitability. From 1990 to 2020, 27% of wetlands were converted to single-cropped lands, and 5% were converted to multi-cropped agricultural land. Since the study spatially identified the potential suitable area and trend of wetland habitat transformation, this could help policymakers define suitable planning for the restoration and conservation of such promising bird habitat.

Effect of channel morphological changes on wetland transformation

Keeping aside the traditional approaches to investigating floodplain wetland transformation, the current study investigated various aspects of it through changes in river channel morphology and drainage pattern. The study analyzed wetland transformation using satellite image-based machine learning and intensive fieldwork. Ordinary Least Square (OLS) regression was applied to identify dominant influencing factors among 24 contributing factors under six clusters to eight dependent phenomena of transformation. The result showed that 57 % of wetland area lost since 1991, and existing wetland has also experiencing hydrological scarcity. From 1991 to 2021, the area under low water depth (<1 m.) inflated from 18.55 % to 50.54 %, the hydro-period narrowed down, and the appearance of water become inconsistent. The OLS result showed that changes in channel morphology (bottle neck channel, embankment-driven carrying capacity enhancement, etc.), interruptions in river and wetland connecting channels (source closure, breaching the continuity, conversion in to agricultural land, etc.), and changes in flood ambience (regulated by dam construction, erection of embankments, etc.) majorly contributed to wetland transformation. Very high explainability was found in the cases of rate of wetland loss, decreasing water depth under greater depth, narrowing hydro-period (R2 > 0.9). The findings of this work would be a good policy document for floodplain wetland management.

Anthropogenic impacts on urban blue space and its reciprocal effect on human and socio-ecological health

Quantifying anthropogenic impacts on blue space (BS) and its effect on human and socio-ecological health was least explored. The present study aimed to do this in reference to the urban BS transformation scenario of Eastern India. To measure BS transformation, Landsat image-based water indices were run from 1990 to 2021. Anthropogenic impact score (AIS) and 7 components scores of 78 selected BS on 70 parameters related data driven from the field. Total 345 respondents were taken for human and socio-ecological health assessment. For this, depression (DEP), anxiety (ANX), stress (STR), physical activities (PA), social capital (SC), therapeutic landscape (TL) and environment building (EB) parameters were taken. The result exhibited that BS was reduced. About 50% of urban core BS was reported highly impacted. Human and socio-ecological health was identified as good in proximity to BS, but it was observed better in the cases of larger peripheral BS. AIS on BS was found to be positively associated with mental health (0.47-0.63) and negatively associated with PA, SC, TL and EB (-0.50 to -0.90). Standard residual in ordinary least square was reported low (-1.5 to 1.5) in 95% BS. Therefore, BS health restoration and management is crucial for sustaining the living environment.

Evaluating the ecological vulnerability of Chongqing using deep learning

This study used deep learning to evaluate the ecological vulnerability of Chongqing, China, discuss the deep learning evaluations of ecological vulnerability, and generate vulnerability maps that support local ecological environment protection and governance decisions and provide reference for future studies. The information gain ratio was used to screen the influencing factors, selecting 16 factors that influence ecological vulnerability. Deep neural network (DNN) and convolutional neural network (CNN) methods were used for modeling, and two ecological vulnerability maps of the study area were generated. The results showed that the mean absolute error and root mean square error of the DNN and CNN models were relatively small, and the fitting accuracy was high. The area under the receiver operating characteristic curve of the CNN model was 0.926, which was better than that of the DNN model (0.888). Random forest was applied to calculate the importance of the influencing factors in the two models. Because the main factor was geological features, the relative ecological vulnerability was mainly affected by karst topography. Through the analysis of the ecological vulnerability map, the areas with higher vulnerability are the karst mountains of Dabashan, Wushan, and Qiyaoshan in the northeast and southeast, as well as the valley between mountains and cities in the center and west of the study area. According to the investigation of these areas, the primary ecological problems are low forest quality, structural irregularities caused by self-geological factors, severe desertification, and soil erosion. Human activity is also an important factor that causes ecological vulnerability in the study area. In conclusion, deep learning, particularly CNN models, can be used for ecological vulnerability assessments. The ecological vulnerability maps conformed to the basic cognition of field surveys and can provide references for other deep learning vulnerability studies. While the overall vulnerability of the study area is not high, ecological problems that lead to its vulnerability should be addressed by future ecological protection and management measures.

How far the types and wetland hydrological conditions influence its provisioning services in the Indian mature Ganges delta

Present work intended to explore how far the Provisioning Service Value (PSV) of the mature Ganges deltaic wetlands is determined by its typology and a few physical attributes like hydrology and aquatic vegetations. Firstly, a field investigation was carried out in the representative sample sites, and field-measured PSV was calibrated with wetland types, hydrological security, and aquatic plant biomass to perform spatial estimation and mapping of PSV. The estimation yielded average annual PSV of entire wetlands as 146.5 × 10⁵ Indian Rupee (INR)/km²/year, with the highest over bheries (embankments for fish and shrimp aquaculture) 176 × 10⁵ INR/km²/year and lowest over marshy wetlands 107 × 10⁵ INR/km²/year. Sensitivity analysis of this estimation showed in cases of 55% field visited sites, the field-measured PSV was outside the range of low standard regression residuals (-0.5 to 0.5). While searching for the reason behind such error in the estimation, the variability of the field-measured PSV was measured. Various inequality measures showed high inequality in inter and intra-hydrological conditions of the wetland. Analysis of variance (ANOVA) proved statistical significance of within-class variability. To explain the variability of PSV, Kernel Density Estimation (KDE) plotting was performed, incorporating a few other regional conditioning factors like wetland size, fish and shrimp aquaculture, perenniality, expenditure, and external feeding from the experience of the field. From this excesize, external feeding and expenditure were essential factors that should be incorporated along with the wetland characteristics and physical attributes for accurate estimation. Since producing spatial data layers of these factors with a finer resolution is difficult, the study suggests case-specific estimation of PSV instead of general spatial mapping.

Linking river flow modification with wetland hydrological instability, habitat condition, and ecological responses

Flow modification pursuing dams is widely found. Some works also focused on its impact on floodplain wetland hydrology. However, how this change can pose an impact on habitat conditions, ecological conditions, and trophic state is also a matter of investigation. The very least attention has been paid to this so far. Therefore, the present study focused on these, taking the dam-induced Lower Tangon river basin of India and Bangladesh as a case. The degree of flow alteration in the river was presented in a heat map. Multi-parametric machine learning (ML) approaches were applied to model hydrological instability and habitat condition. The ecological consequences like evaluating eco-deficit using flow duration curve (FDC) approach, trophic state using trophic state index (TSI), fish habitat zone using image-based hydrological parameters, etc. were measured. The study exhibited that after damming, the degree of river flow modification was about 41%. Consequently, the wetland hydrological instability and habitat conditions were degraded. In the post-dam period, > 50% of wetland area was lost, and hydrological instability was enhanced considerably over wider parts of the wetland. Habitat conditions of the existing wetland also witnessed fragility (poor and very poor areas increased by about 22.23 and 9.34%). As a result of this, adverse ecological responses were found. For instance, the eco-deficit area was increased by 36.19%, a good proportion (100%) of wetlands was witnessed the transformation of TSI from oligotrophic to mesotrophic state, and optimum fish habitat area was declined. The ecological strength map integrating all the cause-effect model parameters showed that good ecological strength was reduced from 49 to 2% in the post-dam. The result of the study would be very useful for wetland restoration for ecological and human well-being.

Machine learning algorithm-based risk assessment of riparian wetlands in Padma River Basin of Northwest Bangladesh

Wetland risk assessment is a global concern especially in developing countries like Bangladesh. The present study explored the spatiotemporal dynamics of wetlands, prediction of wetland risk assessment. The wetland risk assessment was predicted based on ten selected parameters, such as fragmentation probability, distance to road, and settlement. We used M5P, random forest (RF), reduced error pruning tree (REPTree), and support vector machine (SVM) machine learning techniques for wetland risk assessment. The results showed that wetland areas at present are declining less than one-third of those in 1988 due to the construction of the dam at Farakka, which is situated at the upstream of the Padma River. The distance to the river and built-up area are the two most contributing drivers influencing the wetland risk assessment based on information gain ratio (InGR). The prediction results of machine learning models showed 64.48% of area by M5P, 61.75% of area by RF, 62.18% of area by REPTree, and 55.74% of area by SVM have been predicted as the high and very high-risk zones. The results of accuracy assessment showed that the RF outperformed than other models (area under curve: 0.83), followed by the SVM, M5P, and REPTree. Degradation of wetlands explored in this study demonstrated the negative effects on biodiversity. Therefore, to conserve and protect the wetlands, continuous monitoring of wetlands using high resolution satellite images, feeding with the ecological flow, confining built up area and agricultural expansion towards wetlands, and new wetland creation is essential for wetland management.