Estimating alpha, beta, and gamma diversity through deep learning

The reliable mapping of species richness is a crucial step for the identification of areas of high conservation priority, alongside other value considerations. This is commonly done by overlapping range maps of individual species, which requires dense availability of occurrence data or relies on assumptions about the presence of species in unsampled areas deemed suitable by environmental niche models. Here we present a deep learning approach that directly estimates species richness, skipping the step of estimating individual species ranges. We train a neural network model based on species lists from inventory plots, which provide ground truthing for supervised machine learning. The model learns to predict species richness based on spatially associated variables, including climatic and geographic predictors, as well as counts of available species records from online databases. We assess the empirical utility of our approach by producing independently verifiable maps of alpha, beta, and gamma plant diversity at high spatial resolutions for Australia, a continent with highly contrasting diversity patterns. Our deep learning framework provides a powerful and flexible new approach for estimating biodiversity patterns.

Desertification in the Sahel Region: A Product of Climate Change or Human Activities? A Case of Desert Encroachment Monitoring in North-Eastern Nigeria Using Remote Sensing Techniques

In Nigeria, desertification has become one of the most pronounced ecological disasters, with the impacts mostly affecting eleven frontline States. This has been attributed to a range of both nat-ural and man-made factors. This study applied a remote sensing-based change detection and indicator analysis to explore land use/land cover changes and detect major conversions from ecologically active land covers to sand dunes. Results indicate that areas covered by sand dunes (a major indicator of desertification) have doubled over the 25 years under consideration (1990 to 2015). Although about 0.71 km2 of dunes have been converted to vegetation, indicative of the success of various international, national, local, and individual afforestation efforts, conversely about 10.1 km2 of vegetation were converted to sand dunes, implying around 14 times more de-forestation compared to afforestation. Juxtaposing the progression of sand dune with climate records of the study area and examining the relationship between indicators of climate change and desertification suggested a mismatch between both processes as increasing rainfall and lower temperatures observed in 1994, 2005, 2012, and 2014 did not translated into positive feedbacks for desertification in the study area. On average, our results reveal that sand dune is progressing at a mean annual rate of about 15.2 km2 in the study area. Based on this study’s land cover change, trend and conversion assessment, visual reconciliation of climate records with land cover data, statistical analysis, observations from ground-truthing, as well as previous literature, it can be inferred that desertification in Nigeria is less a function of climate change, but more a product of human activities driven by poverty, population growth and failed government policies. Further projections by this study also reveal a high probability of more farmlands being converted to sand dunes by the year 2030 and 2045 if current practices prevail.

Harnessing Machine Learning Techniques for Mapping Aquaculture Waterbodies in Bangladesh

Aquaculture in Bangladesh has grown dramatically in an unplanned manner in the past few decades, becoming a major contributor to the rural economy in many parts of the country. National systems for the collection of statistics have been unable to keep pace with these rapid changes, and more accurate, up to date information is needed to inform policymakers. Using Sentinel-2 top of atmosphere reflectance data within Google Earth Engine, we proposed six different strategies for improving fishpond detection as the existing techniques seem unreliable. These techniques include: (1) identification of the best time period for image collection, (2) testing the buffer size for threshold optimization, (3) determining the best combination of image reducer and water-identifying indices, (4) introduction of a convolution filter to enhance edge-detection, (5) evaluating the impact of ground truthing data on machine learning algorithm training, and (6) identifying the best machine learning classifier. Each enhancement builds on the previous one to develop a comprehensive improvement strategy called the enhanced method for fishpond detection. We compared the results of each improvement strategy to known ground truthing fishponds as the metric of success. For machine learning classifiers, we compared the precision, recall, and F1 score to determine the quality of results. Among four machine learning methods studied here, the classification and regression trees performed the best with a precision of 0.738, recall of 0.827, and F1 score of 0.780. Overall, the proposed strategies enhanced fishpond area detection in all districts within the study area.

The Hatnub Quarries Industrial Landscape Survey 2017: Mobile-GIS Ground-truthing of the Satellite Remote-survey

This paper presents the results of the 2017 mobile-GIS survey of 1 km2 around the Hatnub Egyptian alabaster quarries and analysis of the accuracy of the remote-survey of the same area, which was completed in 2016 using satellite imagery. The analysis revealed that remote-survey was a very accurate method for recording archaeological features in clear and unobstructed parts of the desert, while targeted mobile-GIS survey of obscure areas and questionable features was an effective method for reducing inaccuracies in remote-survey data. The results will inform future phases of the Hatnub Industrial Landscape Project and the fieldwork also identified several avenues of future research into routes and roads across the desert.

Vertical Configuration of a Side Scan Sonar for the Monitoring of Posidonia oceanica Meadows

Posidonia oceanica meadows are ecosystem engineers that play several roles in marine environment maintenance. In this sense, monitoring of the spatial distribution and health status of their meadows is key to make decisions about protecting them against their degradation. With the aim of checking the ability of a simple low-cost acoustic method to acquire information about the state of P. oceanica meadows as ecosystem indicators, ground-truthing and acoustic data were acquired over several of these meadows on the Levantine coast of Spain. A 200 kHz side scan sonar in a vertical configuration was used to automatically estimate shoot density, canopy height and cover of the meadows. The wide athwartship angle of the transducer together with its low cost and user friendliness entail the main advantages of this system and configuration: both improved beam path and detection invariance against boat rolling. The results show that canopy height can be measured acoustically. Furthermore, the accumulated intensity of the echoes from P. oceanica in the first 30 centimeters above the bottom is indirectly related to shoot density and cover, showing a relation that should be studied deeply.

Measurement of Seafloor Acoustic Backscatter Angular Dependence at 150 kHz Using a Multibeam Echosounder

Acoustic seafloor measurements with multibeam echosounders (MBESs) are currently often used for submarine habitat mapping, but the MBESs are usually not acoustically calibrated for backscattering strength (BBS) and cannot be used to infer absolute seafloor angular dependence. We present a study outlining the calibration and showing absolute backscattering strength values measured at a frequency of 150 kHz at around 10–20 m water depth. After recording bathymetry, the co-registered backscattering strength was corrected for true incidence and footprint reverberation area on a rough and tilted seafloor. Finally, absolute backscattering strength angular response curves (ARCs) for several seafloor types were constructed after applying sonar backscattering strength calibration and specific water column absorption for 150 kHz correction. Thus, we inferred specific 150 kHz angular backscattering responses that can discriminate among very fine sand, sandy gravel, and gravelly sand, as well as between bare boulders and boulders partially overgrown by red algae, which was validated by video ground-truthing. In addition, we provide backscatter mosaics using our algorithm (BBS-Coder) to correct the angle varying gain (AVG). The results of the work are compared and discussed with the published results of BBS measurements in the 100–400 kHz frequency range. The presented results are valuable in extending the very sparse angular response curves gathered so far and could contribute to a better understanding of the dependence of backscattering on the type of bottom habitat and improve their acoustic classification.

Continuous Monitoring of Tree Responses to Climate Change for Smart Forestry: A Cybernetic Web of Trees

Trees are long-lived organisms that contribute to forest development over centuries and beyond. However, trees are vulnerable to increasing natural and anthropic disturbances. Spatially distributed, continuous data are required to predict mortality risk and impact on the fate of forest ecosystems. In order to enable monitoring over sensitive and often remote forest areas that cannot be patrolled regularly, early warning tools/platforms of mortality risk need to be established across regions. Although remote sensing tools are good at detecting change once it has occurred, early warning tools require ecophysiological information that is more easily collected from single trees on the ground.Here, we discuss the requirements for developing and implementing such a tree-based platform to collect and transmit ecophysiological forest observations and environmental measurements from representative forest sites, where the goals are to identify and to monitor ecological tipping points for rapid forest decline. Long-term monitoring of forest research plots will contribute to better understanding of disturbance and the conditions that precede it. International networks of these sites will provide a regional view of susceptibility and impacts and would play an important role in ground-truthing remotely sensed data.

Artificial Intelligence in Geospatial Analysis for Flood Vulnerability Assessment: A Case of Dire Dawa Watershed, Awash Basin, Ethiopia

This study presents the novelty artificial intelligence in geospatial analysis for flood vulnerability assessment in Dire Dawa, Ethiopia. Flood-causing factors such as rainfall, slope, LULC, elevation NDVI, TWI, SAVI, K-factor, R-factor, river distance, geomorphology, road distance, SPI, and population density were used to train the ANN model. The weights were generated in the ANN model and prioritized. Initial values were randomly assigned to the NN and trained with the feedforward processes. Ground-truthing points collected from the historical flood events of 2006 were used as targeting data during the training. A rough flood hazard map generated in feedforward was compared with the actual data, and the errors were propagated back into the NN with the backpropagation technique, and this step was repeated until a good agreement was made between the result of the GIS-ANN and the historical flood events. The results were overlapped with ground-truthing points at 88.46% and 89.15% agreement during training and validation periods. Therefore, the application of the GIS-ANN for the assessment of flood vulnerable zones for this city and its catchment was successful. The result of this study can also be further considered along with the city and its catchment for practical flood management.

The Relationship Between Feeding Patch Quality and Fodder Species of Wild Elephants in the Teknaf Wildlife Sanctuary, Cox’s Bazar, Bangladesh

We examined the relationship between the presence or absence of elephants in patches of land and the most common ecological factors, such as fodder species, water bodies, resting places, elephant movement trails, and soil types, across ten transect sites in the Teknaf Wildlife Sanctuary (TWS), Bangladesh. By ground-truthing 360 line transects and 1080 quadrate blocks, we recorded a total of 184 fodder species, including 71 monocotyledons, 58 dicotyledons, and 55 domesticated plant species. Three categories of domesticated fodder species were recorded that consisted of 13 cultivated crops, 24 vegetables, and 18 homestead garden plants. We also applied dung-pile dissection techniques to a total of 250 dung piles between August 2018 and July 2019. Highly statistically significant differences among the abundances of different fodder species and presence of elephants were found across different transect sites. The average fodder species density was found to be 3.44 plant species per site per km2, while the elephant density was 0.63 individuals per site per km2. A significant strong correlation was found between fodder species density and the number of elephants among the transect sites (P = 0.02). The numbers of ground-recorded fodder species were higher than those found in dung piles. The presence of elephants across transect sites was influenced not only by fodder species but also by other ecological factors, such as water bodies, resting places, movement trails, and soil types.

DETECTION AND SPATIAL ANALYSIS OF LAND-USE: A CASE OF BUTUAN CITY WITH HISTORY OF MAJOR INFORMAL SETTLEMENTS

This study detects the significant informal settlements in Butuan City proper. It determines the growth rate in 15 years with the given five-year interval. Machine learning algorithms and spatial analysis were applied to obtain the possible locations of informal settlement buildings. The projected locations of informal settlement buildings were validated thru aerial image validation using Remote Sensing and GIS-based techniques in ArcGIS software. Eight (8) barangays satisfy all the informal settlement building characteristics during the aerial validation process and ground-truthing, namely, Golden Ribbon, Holy Redeemer, Limaha, New Society, Ong Yiu, Port Puyohon, San Ignacio, and Tandang Sora. The eight (8) barangays were manually digitized from the given 5-years interval from 2005, 2010, 2015, and 2010. The value of the major informal settlement buildings area was computed to excel. The area growth rate was calculated using the growth rate formula. This study showed that the significant informal settlement in the study area increased. Among the eight (8) focused barangays, Tandang Sora ranked the highest informal settlements growth from 2005 to 2020. Its area increases up to 178.52%, a total of 24,608.43 square meters. Finally, the results revealed that the area of informal settlement buildings in Butuan City from 2005–2020, in 15-years, its value increases up to 9.74%, a total of 19,172.88 square meters.