Evaluating the performance of artificial intelligence methods for estimation of monthly mean soil temperature without using meteorological data

Soil temperature has an important role in agricultural, hydrological, meteorological and climatological studies. In the present research, monthly mean soil temperature at four different depths (5, 10, 50 and 100 cm) was estimated using artificial neural networks (ANN), adaptive neuro-fuzzy inference system (ANFIS) and gene expression programming (GEP). The monthly mean soil temperature data of 31 stations over Iran were employed. In this process, the data of 21 and 10 stations were used for training and testing stages of used models, respectively. Furthermore, the geographical information including latitude, longitude and altitude as well as periodicity component (the number of months) was considered as inputs in the mentioned intelligent models. The results demonstrated that the ANN and ANFIS models had good performance in comparison with the GEP model. Nevertheless, the ANFIS generally performed better than ANN model.     

Discrimination of Mineralized Rock Types in a Copper-Rich Volcanogenic Massive Sulfide Deposit Through Fast Independent Component and Factor Analysis

Natural Resources Research - Multivariate methods are useful for simplifying the interpretation of variables in geochemical data and are widely used to uncover relationships between elements that...     

Effect of dry density,soil texture and time-spatial variable water content on the soil thermal conductivity

Study of the heat transfer process in saturated and unsaturated soils requires, basically, a relationship between thermal conductivity and the characteristics of the soil, such as water content, dry density and texture of the soil. This study intends to produce a generic model that can predict soil thermal conductivity with the help of easily measurable parameters. The proposed model is first calibrated using measured thermal conductivities from literature data. In order to validate the proposed model the predicted thermal conductivity of this proposed model as well as existing ones are compared with the measured thermal conductivity in literature for different soils. Validation of the proposed model was also performed on our experimental results obtained for a compacted Misillac sand and in-situ clay loam soils. The results show an average of 15% improvement in prediction accuracy for the proposed model compared to the existing models, considering all soil textures. Moreover, we perform a model to estimate thermal conductivity over time throughout the profile of soil in the context of seasonal variation of temperature. The proposed model shows an important effect of heterogeneity on the thermal conductivity variations of a double layered soil.     

Impact of water source dynamics on dissolved reactive phosphorus loadings in heterogenous karst agroecosystems with phosphatic limestones

Karst landscapes underlain with phosphatic limestones are now recognized to be an important contributor of fluvial phosphorus (P) to coastal waters. Specifically, karst agroecosystems may be a hotspot for dissolved reactive P (DRP) due to chronic over-application of organic and inorganic fertilizers that create legacy P accumulation in surface soils. Nevertheless, few studies have assessed the hydrologic controls on DRP transport in these systems at the watershed scale, which is the focus of this study. We analysed soil moisture, soil water extractable P, and storm event hydrologic and water quality data from a small heterogenous karst watershed (10.7 km²) in the Inner-Bluegrass Region of Central Kentucky, USA. Four storm events were sampled in winter, 2020 and were analysed for flow pathways using hydrograph recession analysis and water source connectivity using a tracer-based unmixing model. Based on hydrograph separation results, multiple linear regression analysis was performed to assess drivers of DRP concentrations and loadings. Soil water extractable P results showed stark vertical gradients with greater concentrations at both the surface and deeper soil zones, and minimum concentrations in the root zone. Results for the storm event analysis showed that water source connectivity provided superior prediction of DRP concentrations over the flow pathway analysis, which reflected the heterogeneity of karst maturity masking intermediate flow pathways. Findings from the MLR and loading analysis suggest waters sourced from the soil/epikarst produced significantly higher loadings compared with phreatic and precipitation water source in the three largest events, although concentrations fell between the phreatic (low) and precipitation (high) sources. Findings highlight variable activation of matrix-macropore exchange at different depths throughout the event. Collectively these results suggest existing models and approaches to assess karst hydrology need revision to improve management strategies in this critical landscape.     

Application of sediment rating curves to evaluate efficiency of EPM and MPSIAC using RS and GIS

Erosion potential method (EPM) and Modified Pacific Southwest Interagency Committee (MPSIAC) are two empirical models for estimating soil erosion and sediment delivery. These models use a relatively simple formulation, but they are still applied in various areas with different environmental conditions. However, evaluation of their efficiency is challenging. Accordingly, the main purpose of this study is investigating the performance of EPM and MPSIAC in estimating soil erosion and sediment yield using sediment rating curve (SRC) methods. Talar watershed in Iran was selected as the study area and suspended sediment load (SSL) of two Shirgah–Talar and Valikbon stations were used to assess the output of the models. Remote sensing and geographic information system were utilized in implementing the models. The estimated sediment yield values by the models were evaluated using the results of least square error regression and quantile regression (QR) SRC methods. Then, sediment yield values were obtained from 20-year discharge data (1992–2011). Despite the high uncertainty of QR results, the annual sediment delivery values of the models were achieved in an acceptable range. The most likely (with a probability of 0.5) average annual SSL values were between 713?×?10³ and 840?×?10³ ton for Shirgah–Talar station. Those values for Valikbon station were between 3142?×?10¹ and 3702?×?10¹. Moreover, the estimated average sediment yield in Shirgah–Talar station using MPSIAC and EPM were 591392 and 514054 ton/year, respectively. Those values for Valikbon station were 51881 and 27449 ton/year. Then, the results proved the better performance of MPSIAC in estimating SSL in the study area compared with EPM.     

Determination of Key Beds from the Cap Rocks of Oil Reservoirs Using a Novel Method,Case Study: The Gachsaran Formation,Southwest Iran

Lithology and Mineral Resources - The use of key beds in the cap rocks of the oil reservoirs is crucial. Lack of awareness of these key beds will have serious risks and damages. The Gachsaran oil...     

Exploring the sample size and replications scenarios effect on spatial prediction of flood,using MARS and MaxEnt methods case study: saliantape catchment,Golestan, Iran

Flood has always been a destructive natural hazard during the recent years. Hence, this research aimed to predict the potentiality and probability of flood phenomenon by using the two well-known models, i.e., the MARS algorithm (multivariate adaptive regression splines) and MaxEnt (maximum entropy) method in the Saliantapeh catchment, Golestan province, Iran, covering 4515.47 km². First, documentary sources report and field surveys were used to provide a flood database map. Then, to prepare the flood spatial potentiality map (FSPM), we select sixteen influential variables as predictors. Furthermore, the relative contributions of predicting factors are estimated using the MaxEnt method. For the analysis of data sensitivity and the uncertainty of the proposed models, different scenarios including the sample size (50%/50%, 80%/20%, and 70%/30%, respectively, for training and validation), and the number of replications (5, 10, and 20) were used. Along with the area under the ROC curve (AUC), the highest accuracy for both models corresponds to the first scenario of sample size (80/20%). Contrarywise, it can be concluded that for this scenario, the MARS technique indicated higher predictive skill (AUC?=?98.51%). Regarding the second scenario, which is corresponding to the replicate, the MARS model with 20 replications still has the highest accuracy (94.70%) compared to the other scenarios and the MaxEnt model. The findings of robustness demonstrated that the scenarios with the greatest AUC value have the highest robustness. This work demonstrates that the utilization of the best accurate model with high certainty along with FSPM may be useful to identify and manage the areas that are most susceptible to flood.

     

Object-based classification of hyperspectral data using Random Forest algorithm

This paper presents a new framework for object-based classification of high-resolution hyperspectral data. This multi-step framework is based on multi-resolution segmentation (MRS) and Random Forest classifier (RFC) algorithms. The first step is to determine of weights of the input features while using the object-based approach with MRS to processing such images. Given the high number of input features, an automatic method is needed for estimation of this parameter. Moreover, we used the Variable Importance (VI), one of the outputs of the RFC, to determine the importance of each image band. Then, based on this parameter and other required parameters, the image is segmented into some homogenous regions. Finally, the RFC is carried out based on the characteristics of segments for converting them into meaningful objects. The proposed method, as well as, the conventional pixel-based RFC and Support Vector Machine (SVM) method was applied to three different hyperspectral data-sets with various spectral and spatial characteristics. These data were acquired by the HyMap, the Airborne Prism Experiment (APEX), and the Compact Airborne Spectrographic Imager (CASI) hyperspectral sensors. The experimental results show that the proposed method is more consistent for land cover mapping in various areas. The overall classification accuracy (OA), obtained by the proposed method was 95.48, 86.57, and 84.29% for the HyMap, the APEX, and the CASI data-sets, respectively. Moreover, this method showed better efficiency in comparison to the spectral-based classifications because the OAs of the proposed method was 5.67 and 3.75% higher than the conventional RFC and SVM classifiers, respectively.