Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract The Soil Conservation Services Curve Number model (SCS-CN) is well documented and should be applied in emerging land uses like dynamic urbanizing watersheds. The present research attempts at primarily to investigate the reliability of Curve Number (CN) method in the urbanizing watershed of Upper Bhima from Western Maharashtra for runoff estimation and secondarily validating the CNs using two approaches at basin scale from storm event data. The mean CNtab for the basins were generated using the standard SCS-CN tables whereas event-based rainfall–runoff data was employed to compute the CNemp for eight basins. It was observed that the CN obtained by both methods does not vary significantly for almost all the basins. The uncertainty analysis reveals that there is least uncertainty in these CNs based on mean conditions. All the basins follow the standard behavior pattern wherein curve numbers reach an asymptote level at a lower end of the CNs. In case of most of the basins, the measured and simulated runoff values have little deviation. The Nash–Sutcliffe Efficiency (NSE) test results yielded higher results indicating that the model used for estimation (SCS-CN) has a higher performance rating and low level of uncertainty. The coefficient of determination (r2 or D) values for all the basins were well above 0.70 and the correlation between estimated and measured runoff was significant at 0.05 level. The moderate to very low level of uncertainty in the functioning of the SCS-CN model and satisfactory to very good ratings on the performance scale for all the basins indicates that the procedure adopted for the estimation of runoff in these watershed would yield high level of certainty in the results. PubDate: 2023-09-15
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract Climate change is a major threat to biodiversity, with global greenhouse gas emissions exceeding the Paris Agreement, which has a significant impact on the distribution of species at risk of facing extinction. Thus, predicting climate change's influence on species distribution is crucial. In Sub-Saharan Africa, particularly in Benin, some useful plants such as Parkia biglobosa, Vitex doniana, and Vitellaria paradoxa contribute greatly to improving socio-economic standards. However, they are subjected to overexploitation and climate change, which potentially could lead to their extinction. To predict the habitat suitability of these native agroforestry species for their conservation and cultivation, we assessed the best-performing algorithm among Maximum Entropy, Random Forest, Support Vector Machine, Generalized Linear Models and Boosted Regression Tree. Data were collected from field occurrences and Global Biodiversity Information Facility, and coupled with environment variables selected based on collinearity tests, contribution of variables, and Jackknife tests. We analyzed the main variables affecting their distribution under Representative Concentration Pathways (RCP) 4.5 and RCP 8.5 scenarios by the year 2055. Results showed that Random Forest (RF) was the most appropriate model for predicting the distribution of the three species, with an area under the curve (AUC) > 0.90. Cation exchange capacity, isothermality, and potential evapotranspiration are the environmental factors that all three species depend on. Under current environmental conditions, P. biglobosa, V. paradoxa, and V. doniana covered 52.10%, 76.91%, and 70.22% of the suitable habitats throughout the study area (11,540 km2). A probable expansion of the suitable habitats was noted, with up to 76.19% for P. biglobosa and 82.82% for V. paradoxa. Exceptionally, V. doniana will lose 7.36% of its suitable habitats under the pessimistic (RCP 8.5) scenarios by the year 2055. These findings represent a step forward in the process of conserving P. biglobosa, V. paradoxa, and V. doniana in appropriate habitats in the context of climate change. PubDate: 2023-09-13
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract The influence of interaction effects from adjacent excavations on multiple unsupported excavations could lead to the collapse of excavation walls during the construction of footings. This paper presents the predictions of the stability number of multiple unsupported excavations in cohesive-frictional soils, by using finite element limit analysis (FELA) and an artificial neural network (ANN). A numerical model which is capable of capturing the Mohr–Coulomb (MC) failure criterion is utilized, to determine the lower-bound (LB) and upper-bound (UB) solutions from finite element limit analysis (FELA). The input parameters are excavated depth ratio H/B, excavated spacing ratio S/B, and friction angle ϕ; where H is depth of excavation, S is spacing between excavations, and B is width of excavation. The paper illustrates the effects of these input parameters on the stability number N of multiple unsupported excavations in cohesive-frictional soils. The results obtained from FELA by using OptumG3 (Optum CE in OptumG3 software, Optum Computational Engineering, Copenhagen, 2020) are displayed in dimensionless charts, showing the relationships between the stability number and input parameters, which will be useful for practical applications. The FELA solutions are then employed as a necessary dataset for the ANN model of multiple unsupported excavations in cohesive-frictional soils. The ANN results show good agreement with the results obtained from FELA. Additionally, the relative importance (RI) of the input parameters is calculated using the ANN model to examine the relationship between the input parameters and the stability number. The results indicate that friction angle ϕ is the most critical parameter, followed by excavated depth ratio H/B and excavated spacing ratio S/B. PubDate: 2023-09-09
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract Water is an essential element for life on the planet Earth, which comes on the Earth naturally through precipitation. Continuous increase in human population has diminished the open land water sources and increased surface water pollution as well as atmospheric pollution. The increasing human population and the presence of atmospheric pollutants affect the natural formation of cloud droplets and their conversion into raindrops, respectively, and thus results in rainfall suppression, as observed in the last few decades. In this study, we formulate a nonlinear mathematical model to assess the influence of artificially introduced conducive aerosols in the regional atmosphere through a weather modification technique ‘cloud-seeding’ for rainfall enhancement to rescind the effect of atmospheric pollutants on natural rainfall. In the modeling phenomenon, it is assumed that atmospheric pollutants reduce the condensation and nucleation processes of cloud droplets to form raindrops, but the introduction of conducive aerosols in the regional atmosphere increases the condensation and nucleation processes of cloud droplets, thus increases the formation of raindrops. The local as well as global stability behavior of the feasible equilibria are discussed, and sensitivity analysis is performed for the key parameters. Along with the analytical findings, the numerical simulation is presented to corroborate the obtained analytical findings. The model analysis provides some interesting results to overcome rainfall suppression. PubDate: 2023-09-08
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract In this article, we adopt a simplified approach to describe the processes utilized in the parameterization of a parametric model, focusing primarily on the analysis of hurricanes. The model in question has demonstrated a significant capacity to represent variations in the radial wind profile of hurricanes. One of the advantages of this model is the precise determination of certain parameters, such as maximum wind, radial wind, among others. Additionally, the results satisfactorily depict the evolution of these parameters over time, particularly during the intensification phase. Another crucial aspect is that the model has managed to detect the high friction velocity region near the "eye" of the system. This capability persisted even during periods of system intensification and dissipation, being capable of identifying the core of the system, despite facing challenges with increasing central pressure. Regarding the latent heat flux, its increase near the system's "eye" is clearly identified. As the system develops, there is a greater release of latent heat on the eye wall due to the condensation of water vapor in the troposphere. Simultaneously, there is an increase in the sensible heat flux as the hurricane evolves, indicating that the instability mechanism of heat exchange at the surface responds to the rising cyclonic wind over the surface during this evolutionary phase. PubDate: 2023-09-08
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract This research evaluates the chief shaft in the Azad Dam hydroelectric power plant in Iran using a raise borer machine (RBM). Core samples were taken from the exploratory boreholes of the 210-m-long pressurized shaft. This study aims to determine new experimental models using linear and nonlinear regression approaches to estimate the operating parameters of RBM reaming, including daily progress rate, torque, thrust force, instantaneous penetration rate, power consumption, and the field’s specific energy. Data on geological and operating parameters were sampled and recorded from two separate phases during field studies of the main shaft drilling. The results of field studies in secondary reaming, geological mapping, and laboratory results were used statistically. Statistical modeling showed that the rock quality index of RBM (Qr) has the highest correlation coefficient in linear relationships with the operating parameters. Rock quality designation (RQD) and Qr are the most significant parameters affecting the RBM’s rotational speed and torque estimate. Analyzing the development of nonlinear models showed that RQD performs well in estimating RBM’s rotational speed and torque. Also, the field penetration index was a great measure to estimate the drilling’s specific energy. The correlation coefficient in predicting the operating parameters of the RBM’s penetration rate, rotational speed, and torque equaled 0.84, 0.71, and 0.7, respectively. Also, the RMSE value based on nonlinear regression for all three parameters equaled 0.24%, 1.35%, and 3.1%, respectively, indicating a low error in estimating RBM performance. PubDate: 2023-09-07
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Water erosion of soils is increased by global warming and has a detrimental effect on natural resources. The situation is worse in the High Atlas, where a combination of natural and human forces accelerate erosion and reduce the income of local families. In this work, we assessed the existing erosion of the Haouz plain and projected water erosion of the soils into the future. The survey was conducted in three stages. In the first step, we estimated the annual erosion rate of the Haouz plain from 1992 to 2020 using the Revised Universal Soil Equation (RUSLE), publicly available data and the most recent land observations. The efficiency of the multi-criteria analysis method, taking into account how the second step’s water erosion is affected by the five RUSLE equation parameters. Analytical Hierarchy Process (AHP) was used to determine a weight for each criterion. Using new erosion parameters and the projected erosivity of precipitation from the sixteenth phase of the Coupled Model Intercomparison Project (CMIP6) models, the third stage involved predicting soil water erosion in 2040. According to the results, the average annual soil erosion rate of the Haouz plain is currently 3.53 t ha-1 y-1. According to our predictions, the Haouz Plain will experience an increase in erosion to 4.41 t ha-1 y-1 and 5.31 t ha-1y-1 by 2040, respectively, under the circumstances indicated by RCP2.6 and RCP8.5. Policy makers seeking to adopt environmentally sound measures to halt the depletion of soil and water resources in semi-arid environments could use the current assessment and future predictions of soil water erosion in the Haouz Plain as a basis for data. Graphical abstract PubDate: 2023-09-07
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract The coastal basin of Tarfaya is part of the Meso-Cenozoic basins of the Moroccan Atlantic margin which developed following the bursting of Pangea. In this study, Landsat-8 OLI/TIR images were used as a tool for the extraction of geological lineaments. A more detailed statistical analysis of the spatial distribution of the lineaments was applied by comparing the means of their lengths and their orientations in four areas of the basin using the Kruskal–Wallis test. In addition, an ascending hierarchical classification was used to group individuals according to lineament orientation. The results showed that the basin is fractured by a network of 1603 lineaments. The variance analysis of the lineament orientation means proved the subdivision of the basin into four structural zones. The ascending hierarchical classification (CHA) classified the lineaments in the coastal basin of Tarfaya into six groups high-frequency orientations organized into three directions classes (ENE–WSW, NW–SE to NNW–SSE and NE–SW to NNE–SSW) with proportions of 29%, 28% and 43% respectively. These directions reveal the fracturing of the studied area by the reactivation of pre-existing structures during the Variscan and Atlasic tectonic episodes. Thus, during the Hercynian Orogeny, the basin underwent NNW–SSE directed compression and rotation related to late Variscan deformation that led to the partial stress reversal that became NE–SW to NNE–SSW. The ENE–WSW directions represent the reactivation of the Alpine orogeny in the Meso-Cenozoic formations of the basin. The application of multivariate statistical methods for analyzing lineaments in the basin introduces an innovative approach to fracture analysis. PubDate: 2023-09-05
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract Estimating spring discharge in karst aquifers is challenging due to non-linear and non-stationary hydrological processes caused by spatial and temporal variations. This study mimicked the phenomenon by simulating spring discharge using a laboratory physical model. The hydrological processes adopted in the simulation include systems such as infiltration, fissure-conduit, and drainage. We then recorded spring discharge and precipitation values from the simulated model along side the corresponding air temperature and humidity—in order to analyse the time series behaviour of the system. To estimate spring discharge from the simulation, a deep learning algorithm is developed taking temperature, humidity and precipitation as the input. In this work, the Bayesian optimisation was used to sweep through a range of hyperparameter values to search for the top 5 optimal training options for a Long Short Term Memory (LSTM) neural network. In addition, XGBoost was employed to identify the key predictors of spring discharge, resulting in enhanced predictability. The results show that LSTM-1, LSTM-2, LSTM-3, and LSTM-4 are the recommended recurrent neural network designs for predicting spring discharge using all three input parameters. LSTM-1, LSTM-2, and LSTM-3 network architectures are optimal for utilising two input variables: precipitation intensity and temperature. LSTM-5 has shown that a single parameter is inadequate for estimating spring discharge. The LSTMs yielded an RMSE value of ∼0.04, as well as a \({R}^{2}\) value of ∼98.01%. The study showed that using different input parameters, the suggested LSTM model can effectively simulate spring discharge in a karst environment. PubDate: 2023-09-05
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract This letter discusses the models of climate change that do and do not consider greenhouse gases in climate change effects on oil palm. The importance of correct interpretation of previous data when discussing model systems is emphasised. PubDate: 2023-09-01
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract Understanding the properties of fluids present in an oil and gas reservoir is important for the exploration and development of a field. Detecting hydrocarbons early on allows for a more accurate analysis of the reservoir’s fluid characteristics and helps minimize uncertainty in reservoir assessments and future field planning. Two methods commonly used in the oil and gas industry for this purpose are gas while drilling (GWD) and wireline logs analysis. However, each method has its own limitations and advantages, so an integrated approach that combines these methods can provide more accurate and comprehensive information about fluids in a reservoir. The main objective of this work is the desire to develop an understanding of GWD interpretation in the Zubair Formation in North Rumaila oilfield, southern Iraq using real-time chromatographic analysis of hydrocarbons can be used to effectively characterize the reservoir fluids. Our research incorporates three wells from North Rumaila oilfield in southern Iraq, focusing on the “Zubair Formation” which serves as the main target formation for hydrocarbon. The datasets consist of drilling parameters, mud logging data, LWD/wireline log, geological reports, PVT results, wells test data, and formation pressure data. Total gas and chromatograph components reading were synchronized with depth and corresponding lithology. All of the data were checked for potential errors that may have been caused by factors such as drilling mud properties and additives, borehole conditions, and the nature of the formation, etc. Gas quality ratio (GQR) was calculated in order to determine any contamination in the data, whereas gas ratio analysis was conducted to interpret gas readings using the following methods: wetness, balance, character ratio, pixler ratio, fluid saturation, reserval ratio, and an additional light–gas ratio plot. Gas analysis results were compared with production test data and suggesting that this method is reliable and capable of helping to justify a decision for further drilling action. This study explains that integrated gas while drilling (GWD) analysis can be a powerful method for reservoir fluid characterization within minutes after the Zubair reservoir was drilled using real-time mud gas data. In addition, our interpretation suggests that this method is useful to identify fluid characterization and contacts, assisting pressure test (permeability), selecting pay zone, and distinguish between non-productive and productive hydrocarbon. PubDate: 2023-09-01
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract In this study, we assess the spatiotemporal projections of precipitation over Algeria derived from the multi-model ensemble mean (MME) of eleven global climate model datasets within the Coupled Model Intercomparison Project Phase 6 (CMIP6). The patterns of six extreme precipitation indices defined by the Expert Team on Climate Change Detection and Indices were analyzed for two future time slices: the near-future period 2021–2040 and far-future period 2081–2100 relative to the reference period (1995–2014), under two Shared Socio-economic Pathways (SSP) scenarios: medium emission SSP2-4.5 and worst-case scenario (SSP5-8.5). These indices cover those representing the consecutive dry days CDD, simple daily precipitation intensity (SDII), very heavy precipitation days (R20mm), consecutive dry days (CDD), consecutive wet days (CWD) and very wet days (R95p). The MME Projections show a substantial reduction in total precipitation over most parts of the country by the mid and the end of the twenty-first century. The northern region close to the Mediterranean Sea will experience the highest drying, particularly during the October–December season. The reduction in precipitation would exacerbate the prevailing protracted drought over the country and drastically alter the environmental infrastructure. The projections depict an increase of the very heavy precipitation days and the very wet days by the end of the century and under both scenarios, hence, this could increase the flooding and landslides risk over the country, albeit with reduced total precipitation. The findings of this study are useful for hydrological related preparedness in the era of global warming and climate change. PubDate: 2023-09-01
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract Association of dissolved uranium with other water quality attributes such as sulphate, chloride, alkalinity, nitrate, calcium and magnesium have been investigated in the groundwater of the study area. Through correlation test, a significant correlation was found between uranium with chemical attributes like sulphate, chloride, calcium, nitrate and alkalinity. In view of the toxicity and dissolution characteristics and anthropogenic abundance, nitrate was considered as secondary contaminant for spatial correlation with uranium. To accomplish the objective of spatial correlation, EBK modeling was applied for both uranium and nitrate. Further annual effective dose was calculated for the population of this area. A significant spatial correlation (r = 0.56; p-value < 0.05) was observed between U and nitrate for the study area. In addition, to check the redundancy in correlated raster bands standardized principal component analysis was applied on 1 km × 1 km raster bands. Through statistical analysis, this research concluded that presence of dissolved uranium in groundwater in the form of dissolved uranyl nitrate was due to anthropogenic intervention. The annual effective dose (AED) due to ingestion of uranium in drinking water was estimated for two age groups (children and adults). The calculated annual effective dose reflected that children in the region were more vulnerable to U in drinking water. However, 95th percentile of the AED was well below the reference dose of 0.1 (mSv year−1). PubDate: 2023-09-01
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Groundwater is a key resource for agriculture, which uses approximately 9–105 Mm3 of groundwater in producing a global output valued at $210–$230 billion each year (Shah et al. Water for food, water for life: a comprehensive assessment of water management in agriculture. Earthscan and Colombo, International Water Management Institute, London, 2007). The analysis of groundwater abstraction and water levels shows that the area's excessive water abstraction almost fully controls the aquifer's hydrology. The hydro-abstraction well is a hydraulic structure which is used as an alternative source instead of Ranney well. The problems with Ranney wells are that Ranney wells are difficult and need a complex hydro-geological system where the sub-surface strata should be medium-to-coarse sand along with water table condition within 6 m from the ground surface. If the clay layers come in between, depth of the Ranney well which is 32 m, then sinking of Ranney well cannot be done uniformly. As a result, the rings/walls of the Ranney well develop cracks and fail due to differential settlement where the RCC of the rings comes under tensile position. Irrespective of Ranney well, the hydro-abstraction well can be installed anywhere along the river bank with all types of strata and to a depth of 15 m water level from the ground surface. To address the seepage and ground water issues and manage them, the paper compares the two types of wells. The investigation that was done on a location that is often influenced by ground water hazards is described in the report. For the interpretation of the sub-surface geological feature, fieldwork justifications are given. PubDate: 2023-09-01
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract The hydrologic model is a simplified representation of an existing hydrologic system that helps water resources comprehension, forecasting, and management. Hydrological models are a vital component and essential tool for water resources, environmental planning and management. Urbanization and industrialization significantly impact hydrologic processes locally and globally due to the rapid expansion of population worldwide. Thus, development planning and managing various water resources must meet multiple water demands. However, acquiring gauge discharge data have always been difficult since measurements cannot be taken at every point along the river. Hydrological models are tools, used extensively to simulate many processes of the hydrological cycle. The various ongoing researches are on topics in which the model gives more compatible results with observed discharges. However, it is argued that complex modelling does not provide better results due to soil heterogeneity and climatic changes that play vital roles in streamflow behaviour. Recently, several studies have been conducted to examine the compatibility of model results with streamflow measurements. This paper aims is to provide comprehensive state-of-the-art technology hydrological modelling by briefly discussing different hydrological models and evaluate their application based on Nexus assessment. Furthermore, this paper discussed the different loss methods such as Soil Conservation Service Curve Number (SCS-CN), Soil Moisture Accounting (SMA), Green and Ampt (G.A.), Deficit and constant (D.C.) available in Hydrologic Engineering Center-Hydrologic Modeling System (HEC-HMS). The literature review suggested that that the HEC-HMS is feasible compared to other models. In additions the review demonstrates that the HEC-HMS performed well for dendritic watershed drainage patterns. This study observed that the SCS-CN method and the SMA method are the most widely methods for event-based and continuous modelling. Compared to other models the D.C. loss approach of the HEC-HMS is the least utilized but found to be straightforward and provide accurate results. This study guides modellers in identifying the type of hydrological models that need to employ to a particular catchment for a specific problem. It also equally helps water resources managers and policymakers by providing them with an executive summary of hydrological studies and sustainable development. PubDate: 2023-09-01
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract Bus rapid transit (BRT) vehicles are common microenvironments in urban areas. In some cities, these BRT vehicles are diesel-powered, which makes them highly pollutant. Recent studies report high levels and exposure risk to particulate matter in BRT vehicles. Nevertheless, extensive research has yet to be published, including gaseous pollutants (e.g., CO). Nevertheless, extensive research including gaseous pollutants (e.g., CO) has not been published. This research aims to evaluate the self-pollution of BRT buses in terms of exhaust gasses. For this, measurements and computational fluid dynamics (CFD) were used. Results suggest that pollutant concentrations stay low during most of the trips. However, some areas of the buses have significant swings and peaks due to the transit cycle. Here, we used CFD modeling to evaluate the dispersion of the exhaust CO inside and outside the bus. CFD results show that the bus rear has the highest concentrations, with a mean self-pollution ratio of 12%. Additionally, we developed a method based on the source-receptor relationship to quantify the impact of exhaust emissions reduction on self-pollution, showing that the technological replacement of current diesel buses would reduce self-pollution and, therefore, passenger exposure. Finally, since modeling results may be inaccurate, an uncertainty analysis was developed using the Monte Carlo method to obtain a confidence interval of 90% for the variables linked to the self-pollution. PubDate: 2023-09-01
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract Soil organic matter (SOM) and pH are important indicators to evaluate soil quality. This study applied several machine learning (ML) techniques, namely Random Forest (RF), k-Nearest Neighbor method (kNN) and Cubist, to digital map SOM and pH (H2O and KCl) contents in the foothills of the Ural Mountains, Russia. For this purpose, a total of 52 soil samples were collected from the topsoil depth (0–20 cm). The environmental variables were derived from a digital elevation model (DEM), satellite imagery (Sentinel-2A) and land use/land cover (LULC) map. The ML models were calibrated and validated by the leave-one-out cross-validation approach. The coefficient of determination (R2) and root-mean-square error (RMSE) were used to determine the ML model performance. According to the R2 and RMSE metrics, Cubist method resulted in the most accurate spatial prediction for SOM (R2 = 0.64 and RMSE = 1.95), while RF approach showed the highest performance to predict pH H2O and KCl (R2 = 0.49; RMSE = 0.45 and R2 = 0.44; RMSE = 0.61, respectively). Results showed that remote sensing data were the key variables to explain the variability of all soil properties. Sentinel-2A bands B8A, B7 and B8 were the most effective covariates in predicting SOM, whereas spectral indices GNDVI and SAVI explained most of the spatial distribution of both soil pH. According to the generated maps, the highest SOM concentrations (4–8%) were found under the forest and especially at the bottom of the slopes, which is consistent with favorable conditions of organic carbon accumulation and its redeposition under the influence of water erosion. More acidic soils (pH H2O < 6) were also located under the forest, consisting of mixed coniferous broad-leaved species, which also affects the soil acidity. This study confirms the necessity to use various ML techniques to predict individual soil properties. Overall, our findings and generated maps can provide useful information for future digital soil mapping of areas in similar climatic and geographic conditions. PubDate: 2023-09-01
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract Monitoring and managing lake water quality are some of the most reminiscent of aquatic environmental challenges. Prediction of water quality parameters plays a major role in the enrichment of water resource management. This study creates and compares deep learning models such as ANFIS (adaptive neuro-fuzzy inference system), LSTM (long short-term memory), and NAR neural networks in water quality prediction and proposes the most effective prediction model. Obtaining adequate water quality data with high precision to train and test deep learning models is sometimes challenging due to cost or technological constraints. A solution to this problem was established by developing the best band regression empirical water quality (BREWQ) model, which was used to extract seasonal water quality dataset (n = 490) from multi-resolution satellite imagery (Sentinel 2A) for 2016–2021. The accuracy assessment parameters, such as coefficient of determination (R2), mean squared error (MSE), root mean squared error (RMSE), and mean absolute percentage error (MAPE), is confined to illustrate the performance accuracy of the incorporated models. The results from this study inferred that the ANFIS model approach of Gbellmf linear, Gaussmff with the fuzzy set combination of [3 3 3 3] has the potential to predict magnesium-87.03% (MSE:8.2436, RMSE:9.92, MAPE:4.50), potassium-94.34% (MSE:6.24, RMSE: 8.06, MAPE: 31.55), sodium-96.19% (MSE:5.82, RMSE:7.36, MAPE:4.82). Secchi Disc Depth (SDD)-99.87% (MSE:3.72, RMSE:6.14, MAPE:16.03), temperature-96.2% (MSE:2.19, RMSE:3.67, MAPE:45.33), total hardness-90.31% (MSE:8.09, RMSE:10.13, MAPE: 8.91) and turbidity-60.52% (MSE:6.24, RMSE: 8.08, and MAPE:1688.43) with the least error, whereas LSTM showed the lowest error in predicting parameters such as Chlorophyll-a (Chl-a)- 96.76% (MSE:8.78, RMSE:10.4502 & MAPE:16.83), and Total Suspended Solids (TSS)-84.31% (MSE:3.58, RMSE: 6.68, MAPE:35.69). On the other hand, it is revealed from this study that NARNET had less accuracy than LSTM and ANFIS in predicting water quality due to their simple network structure. In the ultramodern era, such deep-learning prediction models aid in continuously monitoring water bodies to prevent pollution. PubDate: 2023-09-01
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.