Novel integrated modelling based on multiplicative long short-term memory (mLSTM) deep learning model and ensemble multi-criteria decision making (MCDM) models for mapping flood risk.

Flood risk assessment is a key step in flood management and mitigation, and flood risk maps provide a quantitative measure of flood risk. Therefore, integration of deep learning - an updated version of machine learning techniques - and multi-criteria decision making (MCDM) models can generate high-resolution flood risk maps. In this study, a novel integrated approach has been developed based on multiplicative long short-term memory (mLSTM) deep learning models and an MCDM ensemble model to map flood risk in the Minab-Shamil plain, southern Iran. A flood hazard map generated by the mLSTM model is based on nine critical features selected by GrootCV (distance to the river, vegetation cover, variables extracted from DEM (digital elevation model) and river density) and a flood inventory map (70% and 30% data were randomly selected as training and test datasets, respectively). The values of all criteria used to assess model accuracy performance (except Cohens kappa for train dataset = 86, and for test dataset = 84) achieved values greater than 90, which indicates that the mLSTM model performed very well for the generation of a spatial flood hazard map. According to the spatial flood hazard map produced by mLSTM, the very low, low, moderate, high and very high classes cover 26%, 35.3%, 20.5%, 11.2% and 7% of the total area, respectively. Flood vulnerability maps were produced by the combinative distance-based assessment (CODAS), the evaluation based on distance from average solution (EDAS), and the multi-objective optimization on the basis of simple ratio analysis (MOOSRA), and then validated by Spearman's rank correlation coefficients (SRC). Based on the SRC, the three models CODAS, EDAS, and MOOSRA showed high-ranking correlations with each other, and all three models were then used in the ensemble process. According to the CODAS-EDAS-MOOSRA ensemble model, 21.5%, 34.2%, 23.7%, 13%, and 7.6% of the total area were classified as having a very low to very high flood vulnerability, respectively. Finally, a flood risk map was generated by the combination of flood hazard and vulnerability maps produced by the mLSTM and MCDM ensemble model. According to the flood risk map, 27.4%, 34.3%, 14.8%, 15.7%, and 7.8% of the total area were classified as having a very low, low, moderate, high, and very high flood risk, respectively. Overall, the integration of mLSTM and the MCDM ensemble is a promising tool for generating precise flood risk maps and provides a useful reference for flood risk management.

Evaluation of pollution indices, health hazards and source identification of heavy metal in dust particles and storm trajectory simulation using HYSPLIT model (Case study: Hendijan center dust, southwest of Iran).

Atmospheric dust is one of the most recent environmental pollutions in Iran. This study examines the concentration of heavy metals and the assessment of environmental and human health risk in the dust samples of Hendijan region as one of the most important centers of wind erosion in the southwestern of Iran. ICP-MSS analysis was performed on 18 samples of fine dust to specify the concentration of heavy metals. Studies showed that the highest concentrations of metals in these fine dust samples belong to Cr, Ni, Zn, Cu, As, Pb and Cd, respectively. Examining fine dust's pollution assessment showed that the highest enrichment and geo-accumulation index belong to As, Ni and Cr metals. Environmental risk assessment shows the low environmental risk of these fine dusts. The hazard quotient in children and adults belongs to Cr, As and Ni, respectively. Human health risk assessment also showed that the highest absorption of metals in both children and adults is through ingestion. The non-carcinogenic risk of heavy metals of dust samples in children is about 9 times more than adults. The highest risk of cancer in the adult group belongs to Ni metal and in the group of children belongs to As and Ni metal. PCA analysis showed that As, Cu, Cd, Cr and Ni are of anthropogenic origin and Zn and Pb are of geogenic origin. The source of the dust phenomenon with the HYSPLIT model and the backward method indicates the tracking of this dust mass through Iraq, and its probable origin was assessed in the centers of northern Iraq and southeastern Syria.

Synthesis, anticancer activity, and ß-lactoglobulin binding interactions of multitargeted kinase inhibitor sorafenib tosylate (SORt) using spectroscopic and molecular modelling approaches.

Sorafenib tosylate (SORt) is an oral multikinase inhibitor used for treatment of advanced renal cell, liver, and thyroid cancers. In this study, this drug was synthesized and its antiproliferative activities against HCT116 and CT26 cells were assessed. The interaction of SORt with ß-lactoglobulin (BLG) was studied using different fluorescence techniques, circular dichroism (CD), zeta potential measurements, and docking simulation. The results of infrared (IR), mass, H NMR, and C NMR spectra demonstrated that the drug was produced with high quality, purity, and efficiency. SORt showed potent cytotoxicity against HCT116 and CT26 cells with IC50 of 8.12 and 5.42 µM, respectively. For BLG binding of SORt, the results showed that static quenching was the cause of the high affinity drug-protein interaction. Three-dimensional fluorescence and synchronous spectra indicated that SORt conformation was changed at different levels. CD suggested that the α-helix content remained almost constant in the BLG-SORt complex, whereas random coil content decreased. Zeta potential values of BLG were more positive after binding with SORt, due to electrostatic interactions between BLG and SORt. Thermodynamic parameters confirmed van der Waals and hydrogen bond interactions in the complex formation. Molecular modelling predicted the presence of hydrogen bonds and electrostatic forces in the BLG-SORt system, which was consistent with the experimental results.

Sediment source fingerprinting: benchmarking recent outputs, remaining challenges and emerging themes.

PURPOSE: This review of sediment source fingerprinting assesses the current state-of-the-art, remaining challenges and emerging themes. It combines inputs from international scientists either with track records in the approach or with expertise relevant to progressing the science. METHODS: Web of Science and Google Scholar were used to review published papers spanning the period 2013-2019, inclusive, to confirm publication trends in quantities of papers by study area country and the types of tracers used. The most recent (2018-2019, inclusive) papers were also benchmarked using a methodological decision-tree published in 2017. SCOPE: Areas requiring further research and international consensus on methodological detail are reviewed, and these comprise spatial variability in tracers and corresponding sampling implications for end-members, temporal variability in tracers and sampling implications for end-members and target sediment, tracer conservation and knowledge-based pre-selection, the physico-chemical basis for source discrimination and dissemination of fingerprinting results to stakeholders. Emerging themes are also discussed: novel tracers, concentration-dependence for biomarkers, combining sediment fingerprinting and age-dating, applications to sediment-bound pollutants, incorporation of supportive spatial information to augment discrimination and modelling, aeolian sediment source fingerprinting, integration with process-based models and development of open-access software tools for data processing. CONCLUSIONS: The popularity of sediment source fingerprinting continues on an upward trend globally, but with this growth comes issues surrounding lack of standardisation and procedural diversity. Nonetheless, the last 2 years have also evidenced growing uptake of critical requirements for robust applications and this review is intended to signpost investigators, both old and new, towards these benchmarks and remaining research challenges for, and emerging options for different applications of, the fingerprinting approach.

The safety of non-incineration waste disposal devices in four hospitals of Tehran.

BACKGROUND: The safe management of hospital waste is a challenge in many developing countries. OBJECTIVES: The aim of this study was to compare volatile organic compounds (VOCs) emissions and the microbial disinfectant safety in non-incineration waste disposal devices. METHODS: VOC emissions and microbial infections were measured in four non-incineration waste disposal devices including: autoclave with and without a shredder, dry heat system, and hydroclave. Using NIOSH and US EPA-TO14 guidelines, the concentration and potential risk of VOCs in emitted gases from four devices were assessed. ProSpore2 biological indicators were used to assess the microbial analysis of waste residue. RESULTS: There was a significant difference in the type and concentration of VOCs and microbial infection of residues in the four devices. Emissions from the autoclave with a shredder had the highest concentration of benzene, ethyl benzene, xylene, and BTEX, and emissions from the hydroclave had the highest concentration of toluene. The highest level of microbial infection was observed in the residues of the autoclave without a shredder. CONCLUSIONS: There is an increased need for proper regulation and control of non-incinerator devices and for monitoring and proper handling of these devices in developing countries.

An interpretable deep learning model to map land subsidence hazard.

The main goal of this research is the interpretability of deep learning (DL) model output (e.g., CNN and LSTM) used to map land susceptibility to subsidence hazard by means of different techniques. For this purpose, an inventory map of land subsidence (LS) is prepared based on fieldwork and a record of LS presence points, and 16 features controlling LS were mapped. Thereafter, 11 effective features controlling LS were identified by means of a particle swarm optimization (PSO) algorithm, which was then used as input in the CNN and LSTM predictive models. To address the inherent black box nature of DL models, six interpretation methods (feature interaction, permutation importance plot (PFIM), bar plot, SHapley Additive exPlanations (SHAP) main plot, heatmap plot, and waterfall plot) were used to interpret the predictive model outputs. Both models (CNN and LSTM) had AUC > 90 and therefore provided excellent accuracy for mapping LS hazard. According to the most accurate model-the CNN predictive model-the range from very low to very high hazard classes occupied 20%, 20%, 25%, 16.3%, and 18.7% of the study area, respectively. According to three plots (bar plot, SHAP main plot, and heatmap plot), which were constructed based on the SHAP value, distance from the well, GDR and DEM were identified as the three most important features with the highest impact on the DL model output. The results of the waterfall plot indicate two effective features consisting of distance from the well and coarse fragment, and two effective features comprising landuse and DEM, contributed negatively and positively to LS, respectively. Overall, these explanation techniques can address critical concerns with respect to the interpretability of sophisticated DL predictive models.

Using an interpretable deep learning model for the prediction of riverine suspended sediment load.

The prediction of suspended sediment load (SSL) within riverine systems is critical to understanding the watershed's hydrology. Therefore, the novelty of our research is developing an interpretable (explainable) model based on deep learning (DL) and Shapley Additive ExPlanations (SHAP) interpretation technique for prediction of SSL in the riverine systems. This paper investigates the abilities of four DL models, including dense deep neural networks (DDNN), long short-term memory (LSTM), gated recurrent unit (GRU), and simple recurrent neural network (RNN) models for the prediction of daily SSL using river discharge and rainfall data at a daily time scale in the Taleghan River watershed, northwestern Tehran, Iran. The performance of models was evaluated by using several quantitative and graphical criteria. The effect of parameter settings on the performance of deep models on SSL prediction was also investigated. The optimal optimization algorithms, maximum iteration (MI), and batch size (BC) were obtained for modeling daily SSL, and structure of the model impact on prediction remarkably. The comparison of prediction accuracy of the models illustrated that DDNN (with R2 = 0.96, RMSE = 333.46) outperformed LSTM (R2 = 0.75, RMSE = 786.20), GRU (R2 = 0.73, RMSE = 825.67), and simple RNN (R2 = 0.78, RMSE = 741.45). Furthermore, the Taylor diagram confirmed that DDNN has the highest performance among other models. Interpretation techniques can address the black-box nature of models, and here, SHAP was applied to develop an interpretable DL model to interpret of DL model's output. The results of SHAP showed that river discharge has the strongest impact on the model's output in estimating SSL. Overall, we conclude that DL models have great potential in watersheds to predict SSL. Therefore, different interpretation techniques as tools to interpret DL model's output (DL model is as black-box model) are recommended in future research.

Intrinsic and extrinsic techniques for quantification uncertainty of an interpretable GRU deep learning model used to predict atmospheric total suspended particulates (TSP) in Zabol, Iran during the dusty period of 120-days wind.

Total suspended particulates (TSP), as a key pollutant, is a serious threat for air quality, climate, ecosystems and human health. Therefore, measurements, prediction and forecasting of TSP concentrations are necessary to mitigate their negative effects. This study applies the gated recurrent unit (GRU) deep learning model to predict TSP concentrations in Zabol, Iran, during the dust period of the 120-day wind (3 June - 4 October 2014). Three uncertainty quantification (UQ) techniques consisting of the blackbox metamodel, heteroscedastic regression and infinitesimal jackknife were applied to quantify the uncertainty associated with GRU model. Permutation feature importance measure (PFIM), based on the game theory, was employed for the interpretability of the predictive model's outputs. A total of 80 TSP samples were collected and were randomly divided as training (70%) and validation (30%) datasets, while eight variables were used in the TSP prediction model. Our findings showed that GRU performed very well for TSP prediction (with r and Nash Sutcliffe coefficient (NSC) values above 0.99 for both datasets, and RMSE of 57 µg m-3 and 73 µg m-3 for training and validation datasets, respectively). Among the three UQ techniques, the infinitesimal jackknife was the most accurate one, while all the observed and predicted TSP values fell within the continence limitation estimated by the model. PFIM plots showed that wind speed and air humidity were the most and least important variables, respectively, impacting the predictive model's outputs. This is the first attempt of using an interpretable DL model for TSP prediction modelling, recommending that future research should involve aspects of uncertainty and interpretability of the predictive models. Overall, UQ and interpretability techniques have a key role in reducing the impact of uncertainties during optimization and decision making, resulting in better understanding of sophisticated mechanisms related to the predictive model.

An assessment of global land susceptibility to wind erosion based on deep-active learning modelling and interpretation techniques.

Spatial accurate mapping of land susceptibility to wind erosion is necessary to mitigate its destructive consequences. In this research, for the first time, we developed a novel methodology based on deep learning (DL) and active learning (AL) models, their combination (e.g., recurrent neural network (RNN), RNN-AL, gated recurrent units (GRU), and GRU-AL) and three interpretation techniques (e.g., synergy matrix, SHapley Additive exPlanations (SHAP) decision plot, and accumulated local effects (ALE) plot) to map global land susceptibility to wind erosion. In this respect, 13 variables were explored as controlling factors to wind erosion, and eight of them (e.g., wind speed, topsoil carbon content, topsoil clay content, elevation, topsoil gravel fragment, precipitation, topsoil sand content and soil moisture) were selected as important factors via the Harris Hawk Optimization (HHO) feature selection algorithm. The four models were applied to map land susceptibility to wind erosion, and their performance was assessed by three measures consisting of area under of receiver operating characteristic (AUROC) curve, cumulative gain and Kolmogorov Smirnov (KS) statistic plots. The results revealed that GRU-AL model was considered as the most accurate, revealing that 38.5%, 12.6%, 10.3%, 12.5% and 26.1% of the global lands are grouped at very low, low, moderate, high and very high susceptibility classes to wind erosion hazard, respectively. Interpretation techniques were applied to interpret the contribution and impact of the eight input variables on the model's output. Synergy plot revealed that the soil carbon content exhibited high synergy with DEM and soil moisture on the model's predictions. ALE plot showed that soil carbon content and precipitation had negative feedback on the prediction of land susceptibility to wind erosion. Based on SHAP decision plot, soil moisture and DEM presented the highest contribution on the model's output. Results highlighted new regions at high latitudes (southern Greenland coast, hotspots in Alaska and Siberia), which exhibited high and very high land susceptibility to wind erosion.

Stacking- and voting-based ensemble deep learning models (SEDL and VEDL) and active learning (AL) for mapping land subsidence.

This contribution presents a novel methodology based on the feature selection, ensemble deep learning (EDL) models, and active learning (AL) approach for prediction of land subsidence (LS) hazard and rate, and its uncertainty in an area involving two important plains - the Minab and Shamil-Nian plains - in the Hormozgan province, southern Iran. The important features controlling LS hazard were identified by ridge regression. Then, two EDL models were constructed by stacking (SEDL) and voting (VEDL) five dense deep learning (DL) models (model 1 to model 5) for mapping LS hazard. Thereafter, the predictive model performance was assessed by a precision-recall curve and Kolmogorov-Smirnov (KS) plot. A partial dependence plot (PDP), individual conditional expectation plots (ICEP), game theory, and a sensitivity analysis were used for the interpretability of the predictive DL model. According to SEDL - a model with higher accuracy - 34% (1624 km2), 14.7% (698 km2), and 19.2% (912 km2) of the total area were classified as being of very low, low, and moderate hazards, whereas 17.7% (845 km2) and 14.4% (683 km2) of area were classified as being of high and very high hazards, respectively. Based on all interpretability techniques, aquifer loss or groundwater drawdown is the most important feature controlling LS hazard, and it having the greatest impact on the SEDL model output. Based on a Taylor diagram and R2 as model performance assessment indicators, SEDL-AL (with R2 > 95% for training and test datasets) performed better than SEDL for quantify LS rate, the rate of LS ranging between 0 and 48.1 cm. The highest rate of LS occurred in the Minab plain - an area located downstream of the Minab Esteghlal dam. SEDL-AL was used to quantify the uncertainty associated with the LS rate. The observed values fell within predictions provided by SEDL-AL, which indicates a high accuracy of our predictive model. Overall, our newly developed modeling techniques are helpful tools for the spatial mapping of LS susceptibility and rate, and its uncertainty.

High-resolution, spatially resolved quantification of wind erosion rates based on UAV images (case study: Sistan region, southeastern Iran).

Estimating the quantitative distribution of wind erosion rates is one of the most important requirements for managing affected environments and optimizing locations to control wind erosion. This study develops high-resolution maps for wind erosion with unmanned aerial vehicles or UAV images in the Sistan region-an area in southeastern Iran with severe wind erosion. Aerial imaging by UAV was done during period of erosive winds. Changes in the amount of wind erosion were measured for 7 months. Digital elevation models or DEM with a spatial resolution of 6 mm, orthophoto mosaic images with a resolution of 3 mm, were prepared before and after the erosion event. Three erosive facies consisting of surface, edge, and blowout were identified. The amount of erosion in different geomorphological landscapes or facies was measured according to differences of DEMs (DOD). The effect of physical factors of the geomorphological landscapes on wind erosion was investigated by calculating the correlation between the erosion, roughness, and slope in the geomorphological landscapes. The results showed that the highest and lowest mean of eroded soil were 22 mm and 4 mm in the blowout and surface facies, respectively. The average rate of wind erosion was 201 t/ha during the study period, which indicates the high intensity of wind erosion in the Sistan plain. Overall, UAV-as an aerial imaging technique collecting ground data-can be a helpful tool in the aeolian geomorphology especially for collecting data for measuring the rate of soil erosion by the wind in the aeolian landscapes located in remote regions.

Interpretability of simple RNN and GRU deep learning models used to map land susceptibility to gully erosion.

Gully erosion possess a serious hazard to critical resources such as soil, water, and vegetation cover within watersheds. Therefore, spatial maps of gully erosion hazards can be instrumental in mitigating its negative consequences. Among the various methods used to explore and map gully erosion, advanced learning techniques, especially deep learning (DL) models, are highly capable of spatial mapping and can provide accurate predictions for generating spatial maps of gully erosion at different scales (e.g., local, regional, continental, and global). In this paper, we applied two DL models, namely a simple recurrent neural network (RNN) and a gated recurrent unit (GRU), to map land susceptibility to gully erosion in the Shamil-Minab plain, Hormozgan province, southern Iran. To address the inherent black box nature of DL models, we applied three novel interpretability methods consisting of SHaply Additive explanation (SHAP), ceteris paribus and partial dependence (CP-PD) profiles and permutation feature importance (PFI). Using the Boruta algorithm, we identified seven important features that control gully erosion: soil bulk density, clay content, elevation, land use type, vegetation cover, sand content, and silt content. These features, along with an inventory map of gully erosion (based on a 70 % training dataset and 30 % test dataset), were used to generate spatial maps of gully erosion using DL models. According to the Kolmogorov-Smirnov (KS) statistic performance assessment measure, the simple RNN model (with KS = 91.6) outperformed the GRU model (with KS = 66.6). Based on the results from the simple RNN model, 7.4 %, 14.5 %, 18.9 %, 31.2 % and 28 % of total area of the plain were classified as very-low, low, moderate, high and very-high hazard classes, respectively. According to SHAP plots, CP-PD profiles, and PFI measures, soil silt content, vegetation cover (NDVI) and land use type had the highest impact on the model's output. Overall, the DL modelling techniques and interpretation methods used in this study proved to be helpful in generating spatial maps of soil erosion hazard, especially gully erosion. Their interpretability can support watershed sustainable management.

Novel deep learning hybrid models (CNN-GRU and DLDL-RF) for the susceptibility classification of dust sources in the Middle East: a global source.

Dust storms have many negative consequences, and affect all kinds of ecosystems, as well as climate and weather conditions. Therefore, classification of dust storm sources into different susceptibility categories can help us mitigate its negative effects. This study aimed to classify the susceptibility of dust sources in the Middle East (ME) by developing two novel deep learning (DL) hybrid models based on the convolutional neural network-gated recurrent unit (CNN-GRU) model, and the dense layer deep learning-random forest (DLDL-RF) model. The Dragonfly algorithm (DA) was used to identify the critical features controlling dust sources. Game theory was used for the interpretability of the DL model's output. Predictive DL models were constructed by dividing datasets randomly into train (70%) and test (30%) groups, six statistical indicators being then applied to assess the DL hybrid model performance for both datasets (train and test). Among 13 potential features (or variables) controlling dust sources, seven variables were selected as important and six as non-important by DA, respectively. Based on the DLDL-RF hybrid model - a model with higher accuracy in comparison with CNN-GRU-23.1, 22.8, and 22.2% of the study area were classified as being of very low, low and moderate susceptibility, whereas 20.2 and 11.7% of the area were classified as representing high and very high susceptibility classes, respectively. Among seven important features selected by DA, clay content, silt content, and precipitation were identified as the three most important by game theory through permutation values. Overall, DL hybrid models were found to be efficient methods for prediction purposes on large spatial scales with no or incomplete datasets from ground-based measurements.

Fingerprinting the spatial sources of fine-grained sediment deposited in the bed of the Mehran River, southern Iran.

Accurate information on the sources of suspended sediment in riverine systems is essential to target mitigation. Accordingly, we applied a generalized likelihood uncertainty estimation (GLUE) framework for quantifying contributions from three sub-basin spatial sediment sources in the Mehran River catchment draining into the Persian Gulf, Hormozgan province, southern Iran. A total of 28 sediment samples were collected from the three sub-basin sources and six from the overall outlet. 43 geochemical elements (e.g., major, trace and rare earth elements) were measured in the samples. Four different combinations of statistical tests comprising: (1) traditional range test (TRT), Kruskal-Wallis (KW) H-test and stepwise discriminant function analysis (DFA) (TRT + KW + DFA); (2) traditional range test using mean values (RTM) and two additional tests (RTM + KW + DFA); (3) TRT + KW + PCA (principle component analysis), and; 4) RTM + KW + PCA, were used to the spatial sediment source discrimination. Tracer bi-plots were used as an additional step to assess the tracers selected in the different final composite signatures for source discrimination. The predictions of spatial source contributions generated by GLUE were assessed using statistical tests and virtual sample mixtures. On this basis, TRT + KW + DFA and RTM + KW + DFA yielded the best source discrimination and the tracers in these composite signatures were shown by the biplots to be broadly conservative during transportation from source to sink. Using these final two composite signatures, the estimated mean contributions for the western, central and eastern sub-basins, respectively, ranged between 10-60% (overall mean contribution 36%), 0.3-16% (overall mean contribution 6%) and 38-77% (overall mean contribution 58%). In comparison, the final tracers selected using TRT + KW + PCA generated respective corresponding contributions of 1-42% (overall mean 20%), 0.5-30% (overall mean 12%) and 55-84% (overall mean 68%) compared with 17-69% (overall mean 41%), 0.2-12% (overall mean 5%) and 29-76% (overall mean 54%) using the final tracers selected by RTM + KW + PCA. Based on the mean absolute fit (MAF; ≥ 95% for all target sediment samples) and goodness-of-fit (GOF; ≥ 99% for all samples), GLUE with the final tracers selected using TRT + KW + PCA performed slightly better than GLUE with the final signatures selected by the three other combinations of statistical tests. Based on the virtual mixture tests, however, predictions provided by GLUE with the final tracers selected using TRT + KW + DFA and RTM + KW + DFA (mean MAE = 11% and mean RMSE = 13%) performed marginally better than GLUE with RTM + KW + PCA (mean MAE = 14% and mean RMSE = 16%) and GLUE with TRT + KW + PCA (mean MAE = 17% and mean RMSE = 19%). The estimated source proportions can help watershed engineers plan the targeting of conservation programmes for soil and water resources.

Assessment of the uncertainty and interpretability of deep learning models for mapping soil salinity using DeepQuantreg and game theory.

This research introduces a new combined modelling approach for mapping soil salinity in the Minab plain in southern Iran. This study assessed the uncertainty (with 95% confidence limits) and interpretability of two deep learning (DL) models (deep boltzmann machine-DBM) and a one dimensional convolutional neural networks (1DCNN)-long short-term memory (LSTM) hybrid model (1DCNN-LSTM) for mapping soil salinity by applying DeepQuantreg and game theory (Shapely Additive exPlanations (SHAP) and permutation feature importance measure (PFIM)), respectively. Based on stepwise forward regression (SFR)-a technique for controlling factor selection, 18 of 47 potential controls were selected as effective factors. Inventory maps of soil salinity were generated based on 476 surface soil samples collected for measuring electrical conductivity (ECe). Based on Taylor diagrams, both DL models performed well (RMSE < 20%), but the 1DCNN-LSTM hybrid model performed slightly better than the DBM model. The uncertainty range associated with the ECe values predicted by both models estimated using DeepQuantilreg were similar (0-25 dS/m for the 1DCNN-LSTM hybrid model and 2-27 dS/m for DBM model). Based on the SFR and PFIM (permutation feature importance measure)-a measure in game theory, four controls (evaporation, sand content, precipitation and vertical distance to channel) were selected as the most important factors for soil salinity in the study area. The results of SHAP (Shapely Additive exPlanations)-the second measure used in game theory-suggested that five factors (evaporation, vertical distance to channel, sand content, cation exchange capacity (CEC) and digital elevation model (DEM)) have the strongest impact on model outputs. Overall, the methodology used in this study is recommend for applications in other regions for mapping environmental problems.

PCR-SSCP variation of GH and STAT5A genes and their association with estimated breeding values of growth traits in Baluchi sheep.

Growth hormone (GH) selected for its important role in economically relevant traits and signal transducers and activators of transcription 5A (STAT5A) is also known as a main mediator of growth hormone action on target genes. A total number of 190 lambs of Iranian purebred Baluchi sheep were genotyped at exon 5 of GH and exons 7 and 8 of STAT5A genes by using PCR-SSCP analysis. GH gene revealed three (G1, G2, and G3) conformational patterns; however, STAT5A loci were not polymorphic. Breeding values of growth traits including birth weight, weaning weight, 6 months weight, 9 months weight, and yearling weight were estimated by using the Best Linear Unbiased Prediction based on an animal model with a relationship matrix. Studied growth traits were examined for association analysis. Our findings suggest that animals with G2 genotype have highest breeding value for six month weight, while these animals have lowest breeding value for pre-weaning traits. Higher performance of G2 animals in adult ages may be related to the growth hormone role in puberty ages. The other traits showed no relationship to the genotypes examined.

The simultaneous carrier ability of natural antioxidant of astaxanthin and chemotherapeutic drug of 5-fluorouracil by whey protein of ß-lactoglobulin: spectroscopic and molecular docking study.

In the present study, the simultaneous carrier ability of natural antioxidant of astaxanthin (ATX) and chemotherapeutic drug of 5-fluorouracil (5-FU) by whey protein of ß-lactoglobulin (ß-LG) using various spectroscopic techniques (UV-visible, fluorescence, circular dichroism (CD) and dynamic light scattering) in combination with molecular docking were investigated (at room and physiological temperatures). According to the fluorescence quenching tests, the binding parameters between drug and ATX with protein showed that the number of their binding sites was the same in the single and competitive states. Molecular docking results have showed completely consistent with the fluorescence data that presented the independent binding sites for 5-FU and ATX on ß-LG. Also, analysis of Far-UV-CD showed that the simultaneous binding of the drugs to the protein partially enhances its stability, which is associated with the decreasing in ß-sheet structure and increasing in α-helix. According to the Zeta potential measurements in the presence of different concentrations of the drugs, they have stronger binding to the protein at lower concentrations. Therefore, given the remarkable features of ß-LG, including the ability to interact simultaneously with the natural compound of ATX and the antitumor drug of 5-FU, this study could provide useful information for the development and improvement of new protein carrier systems with synergism potency. Communicated by Ramaswamy H. Sarma.

Spatial modelling of soil salinity: deep or shallow learning models?

Understanding the spatial distribution of soil salinity is required to conserve land against degradation and desertification. Against this background, this study is the first attempt to predict soil salinity in the Jaghin basin, in southern Iran, by applying and comparing the performance of four deep learning (DL) models (deep convolutional neural networks-DCNNs, dense connected deep neural networks-DenseDNNs, recurrent neural networks-long short-term memory-RNN-LSTM and recurrent neural networks-gated recurrent unit-RNN-GRU) and six shallow machine learning (ML) models (bagged classification and regression tree-BCART, cforest, cubist, quantile regression with LASSO penalty-QR-LASSO, ridge regression-RR and support vectore machine-SVM). To do this, 49 environmental landsat8-derived variables including digital elevation model (DEM)-extracted covariates, soil-salinity indices, and other variables (e.g., soil order, lithology, land use) were mapped spatially. For assessing the relationships between soil salinity (EC) and factors controlling EC, we collected 319 surficial (0-5 cm depth) soil samples for measuring soil salinity on the basis of electrical conductivity (EC). We then selected the most important features (covariates) controlling soil salinity by applying a MARS model. The performance of the DL and shallow ML models for generating soil salinity spatial maps (SSSMs) was assessed using a Taylor diagram and the Nash Sutcliff coefficient (NSE). Among all 10 predictive models, DL models with NSE ≥ 0.9 (DCNNs was the most accurate model with NSE = 0.96) were selected as the four best models, and performed better than the six shallow ML models with NSE ≤ 0.83 (QR-LASSO was the weakest predictive model with NSE = 0.50). Based on DCNNs-, the values of the EC ranged between 0.67 and 14.73 dS/m, whereas for QR-LASSO the corresponding EC values were 0.37 to 19.6 dS/m. Overall, DL models performed better than shallow ML models for production of the SSSMs and therefore we recommend applying DL models for prediction purposes in environmental sciences.

Detection and prediction of lake degradation using landscape metrics and remote sensing dataset.

Monitoring changes in natural ecosystems is considered essential to natural resource management. Despite the global importance of the lakes' quality monitoring, there is currently a research gap in the simultaneous predictive modeling of lakes' land-use changes and ecosystem measurements. In the present study for projecting the water bodies of lakes and their surrounding ecosystems, the land-use changes and the landscape analysis of different periods, i.e., 1987, 2002, 2018, and 2030, are studied using remote sensing data and various metrics. The trend of land-use and landscape changes is projected for 2030. The results indicate significant degradation of rangelands and forests due to the conversion to agriculture and construction and the declining trend of lakes' water body and their transformation to salt lake and salt lands. The increase of agricultural lands and the overuse of groundwater wells upstream of the lakes could be one of the reasons for this decline. Decreasing the lakes' water body and subsequently increasing salt lands are considered a severe threat to human health and the ecosystem services of the lakes. Besides, the dust generated by salt lands could also decrease crop yield in the study area.

Desertification of Iran in the early twenty-first century: assessment using climate and vegetation indices.

Remote sensing of specific climatic and biogeographical parameters is an effective means of evaluating the large-scale desertification status of drylands affected by negative human impacts. Here, we identify and analyze desertification trends in Iran for the period 2001-2015 via a combination of three indices for vegetation (NPP-net primary production, NDVI-normalized difference vegetation index, LAI-leaf area index) and two climate indices (LST-land surface temperature, P-precipitation). We combine these indices to identify and map areas of Iran that are susceptible to land degradation. We then apply a simple linear regression method, the Mann-Kendall non-parametric test, and the Theil-Sen estimator to identify long-term temporal and spatial trends within the data. Based on desertification map, we find that 68% of Iran shows a high to very high susceptibility to desertification, representing an area of 1.1 million km2 (excluding 0.42 million km2 classified as unvegetated). Our results highlight the importance of scale in assessments of desertification, and the value of high-resolution data, in particular. Annually, no significant change is evident within any of the five indices, but significant changes (some positive, some negative) become apparent on a seasonal basis. Some observations follow expectations; for instance, NDVI is strongly associated with cooler, wet spring and summer seasons, and milder winters. Others require more explanation; for instance, vegetation appears decoupled from climatic forcing during autumn. Spatially, too, there is much local and regional variation, which is lost when the data are considered only at the largest nationwide scale. We identify a northwest-southeast belt spanning central Iran, which has experienced significant vegetation decline (2001-2015). We tentatively link this belt of land degradation with intensified agriculture in the hinterlands of Iran's major cities. The spatial and temporal trends identified with the three vegetation and two climate indices afford a cost-effective framework for the prediction and management of future environmental trends in developing regions at risk of desertification.