Soil Salinity Prediction Research Articles

Comparison of machine learning algorithms for soil salinity predictions in three dryland oases located in Xinjiang Uyghur Autonomous Region (XJUAR) of China

ABSTRACT Many different machine learning approaches have been applied for various purposes. However, there has been limited guidance regarding which, if any, machine learning models and covariate sets might be optimal for predicting soil salinity across different oases in the Xinjiang Uyghur Autonomous Region (XJUAR) of China. This study aimed to compare five machine learning algorithms, Least Absolute Shrinkage and Selection Operator (LASSO), Multiple Adaptive Regression Splines (MARS), Classification and Regression Trees (CART), Random Forest tree ensembles (RF), and Stochastic Gradient Treeboost (SGT), to predict soil salinity in three geographically distinct areas (the Qitai, Kuqa, and Yutian oases). A total of 21 data sets from three oases were used to evaluate the performance of the algorithm and to screen the optimal variables. The results show the following indices are considered to be important indicators for quantitative assessment of soil salinity: EEVI, CSRI, EVI2, GDVI, SAIO, and SIT. Comparison results show that SGT is the most suitable algorithm for predicting soil salinity in arid areas. This study provides a comprehensive comparison of machine learning techniques for soil salinity prediction and may assist in the modeling and variable selection of digital soil mapping in the XJUAR of China.

Development and application of long-term root zone salt balance model for predicting soil salinity in arid shallow water table area

A simple root zone salt balance model was developed for long-term soil salt prediction. A groundwater balance module was integrated to obtain the water flux at the bottom of the root zone, since groundwater level is easily and reliably measured data. An assumption that the net percolation at the bottom of root zone equals to the net recharge from root zone to unconfined groundwater system was used, which makes capillary rise and infiltration water leaching of root zone obtainable. The model accuracy and limitation were evaluated by comparing the simulation results with those from HYDRUS-1D under various soil texture and bottom boundary conditions. The mean relative error (MRE) of soil salt content ranged from −6.25% to 9.95%, and the root mean square error (RMSE) from 0.02 kg/100 kg to 0.05 kg/100 kg, which suggest the model accuracy. The determination coefficient (R2) of the cumulative percolation at the bottom of root zone and recharge to groundwater at the end of each calculation period was 0.99, which validates the model assumption. The model was then applied to predict soil salinity in the Longsheng irrigation district of Hetao Irrigation District, China. The soil salinity and water table depth data from 1999 to 2016 were used to calibrate and validate the model parameters. The MRE values of water table depth and average soil salt content were 10.08% and 10.84%, and RMSE values 0.37 m and 0.04 kg/100 kg in the validation period. Subsequently, the validated model was used to simulate the soil salinity in future 100 years under various water saving conditions. Water table depth and autumn irrigation for salt leaching were the two major factors to control soil salinity. The required autumn irrigation to keep root zone in mild salinity level (0.3 kg/100 kg) can decrease by 25.1 mm with 0.1 m water table depth increase, while it should increase by 64 mm with 1.0 kg/m3 groundwater salinity increase in the crop growing season. Moderate groundwater exploitation to increase water table depth is recommended to save autumn irrigation water and control soil salinity.

The development of an overlay model to predict soil salinity risks by using remote sensing and GIS techniques: a case study in soils around Idku Lake, Egypt.

Soil salinization is one of the major environmental problems facing agricultural lands in arid and semiarid areas of the world because of its detrimental impacts on agricultural production and on the sustainable development of land resources. Hence, predicting soil salinity is essential to avoiding further soil degradation. The present study is intended to develop a model for predicting soil salinity in soils around Idku Lake by using remote sensing and geographic information system techniques. This lake is a shallow brackish basin located in the western part of the Nile Delta. For this purpose, Landsat 8-OLI images and shuttle radar topography mission 1Arc-Second Digital Elevation Model data were utilized in this research. A total of 91 surface samples were collected across the study area at a depth between 0 and 30cm and were analyzed via traditional laboratory analysis methods. Five environmental parameters were used in the design of the soil salinity model. A pairwise comparison matrix was used to calculate the factor weight value for each of the layers. A linear regression model was used to plot the relationship between the EC value and raster value of the salinity map derived from the overlay model. According to the results obtained from a pairwise comparison of the factor layers, water table level was the greatest influential factor of soil salinity, followed by landforms. The validation of the model demonstrated a high degree of correlation (R2 = 0.72) between the measured EC values and the salinity values derived from the model. Furthermore, this model could be a useful tool for predicting soil salinity with a suitable validation.

Soil salinity prediction and mapping by machine learning regression in Central Mesopotamia, Iraq

AbstractSoil salinization affects crop production and food security. Mapping spatial distribution and severity of salinity is essential for agricultural management and development. This study was aimed to test the effectiveness of machine learning algorithms for soil salinity mapping taking the Mussaib area in Central Mesopotamia as an example. A combined dataset consisting of Landsat 5 Thematic Mapper (TM) and ALOS L‐band radar data acquired at the same time was used for fulfilling the task. Relevant biophysical indicators were derived from the TM images, and the soil component was retrieved by removing the vegetation contribution from the L‐band radar backscattering coefficients. Field‐measured salinity at the three corner plots of triangles were averaged to represent the salinity of these triangular areas. These averaged plots were converted into raster by either direct rasterization or buffering‐based rasterization into different cell size to create the training set (TS). One of the three triangle corners was randomly selected to constitute a validation set (VS). Using this TS, the support vector regression (SVR) and random forest regression (RFR) algorithms were then applied to the combined dataset for salinity prediction. Results revealed that RFR performed better than SVR with higher accuracy (93.4–94.2% vs. 85.2–89.4%) and less normalized root mean square error (NRMSE; 6.10–7.69% vs. 10.29–10.52%) when calibrated with both TS and VS. In comparison, prediction by multivariate linear regression (MLR) achieved in our previous study using the same datasets also showed less NRMSE than SVR. Hence, both RFR and MLR are recommended for soil salinity mapping.

Soil salinity mapping utilizing sentinel-2 and neural networks

Soil salinity is the most important soil property that affects the agriculture productivity. Periodical monitoring of its status is considered a crucial factor in the selection of appropriate agricultural practices to attain a sustainable production. The availability of remote sensing data processed by a somewhat novel method such as artificial neural networks (ANN) offer a potential solution that could easily and affordably replace the in-site monitoring methods. The aim of this work is to use high spectral resolution Sentinel-2 (S2) data for soil salinity prediction utilizing neural networks. The study evaluated three approaches in processing the S2 data for inclusion in the artificial neural network for soil salinity prediction. These approaches included S2 spectral reflectance data, spectral indices and principal component analysis (PCA) of the S2 data. The results revealed that a combination of these approaches including the reflectance data of band 11(shortwave infrared band) of S2, the normalized differential vegetation index (NDVI) and the second PCA (dominated by the near infrared band) gave the best performance when used as input when designing the artificial neural networks to predict the soil salinity. The overall accuracy of this approach has a coefficient of determination (R2) of 0.94 between the actual and predicted soil salinity.

Soil Salinity Mapping of Urban Greenery Using Remote Sensing and Proximal Sensing Techniques; The Case of Veale Gardens within the Adelaide Parklands

More well-maintained green spaces leading toward sustainable, smart green cities mean that alternative water resources (e.g., wastewater) are needed to fulfill the water demand of urban greenery. These alternative resources may introduce some environmental hazards, such as salt leaching through wastewater irrigation. Despite the necessity of salinity monitoring and management in urban green spaces, most attention has been on agricultural fields. This study was defined to investigate the capability and feasibility of monitoring and predicting soil salinity using proximal sensing and remote sensing approaches. The innovation of the study lies in the fact that it is one of the first research studies to investigate soil salinity in heterogeneous urban vegetation with two approaches: proximal sensing salinity mapping using Electromagnetic-induction Meter (EM38) surveys and remote sensing using the high-resolution multispectral image of WorldView3. The possible spectral band combinations that form spectral indices were calculated using remote sensing techniques. The results from the EM38 survey were validated by testing soil samples in the laboratory. These findings were compared to remote sensing-based soil salinity indicators to examine their competence on mapping and predicting spatial variation of soil salinity in urban greenery. Several regression models were fitted; the mixed effect modeling was selected as the most appropriate to analyze data, as it takes into account the systematic observation-specific unobserved heterogeneity. Our results showed that Soil Adjusted Vegetation Index (SAVI) was the only salinity index that could be considered for predicting soil salinity in urban greenery using high-resolution images, yet further investigation is recommended.

Remote sensing approach to assess salt-affected soils in the north-east part of Tadla plain, Morocco

Recently, the remote sensing technologies have been used increasingly in various domains in order to explain or detect different phenomena in a rapid manner and covering large areas. This study aims to use Landsat 8 Oli imagery product to elaborate a map of soil salinity in the north-east part of Tadla plain, by implication of spectral reflectance and electrical conductivity (EC) measured in the laboratory. Based on salinity Index (SI), the Normalized Differential Salinity Index (NDSI), and Landsat bands, we carried out a statistical study via the JMP13 software to determine the most correlated bands with EC measured. The obtained results were very satisfactory with an R2 = 71.3% and root mean square error (RMSE) of 0.084. The elaborated map showed that the salinity is high near Oum Er Rbia River and the two cities of Beni-Mellal and Ouled Yaich, which is due to saline waters of Oum Er Rbia River and Béni-Moussa-East (Dir) groundwater used for irrigation. These results signify that the combination of remote sensing and laboratory EC measurements would be a suitable method for predicting soil salinity.

Prediction of Soil Salinity Using Multivariate Statistical Techniques and Remote Sensing Tools

Soil salinity limits plant growth, reduces crop productivity and degrades soil. Multispectral data from Landsat TM are used to study saline soils in southern Tunisia. This study will explore the potential multivariate statistical analysis, such as principal component analysis (PCA) and cluster analysis to identify the most correlated spectral indices and rapidly predict salt affected soils. Sixty six soil samples were collected for ground truth data in the investigated region. A high correlation was found between electrical conductivity and the spectral indices from near infrared and short-wave infrared spectrum. Different spectral indices were used from spectral bands of Landsat data. Statistical correlation between ground measurements of Electrical Conductivity (EC), spectral indices and Landsat original bands showed that the near and short-wave infrared bands (band 4, band 5 and 7) and the salinity indices (SI 5 and SI 9) have the highest correlation with EC. The use of CA revealed a strong correlation between electrical conductivity EC and spectral indices such abs4, abs5, abs7 and si5. The principal components analysis is conducted by incorporating the reflectance bands and spectral salinity indices from the remote sensing data. The first principal component has large positive associations with bands from the visible domain and salinity indices derived from these bands, while second principal component is strongly correlated with spectral indices from NIR and SWIR. Overall, it was found that the electrical conductivity EC is highly correlated (R2 = -0.72) to the second principal component (PC2), but no correlation is observed between EC and the first principal component (PC1). This suggests that the second component can be used as an explanatory variable for predicting EC. Based on these results and combining the spectral indices (PC2 and abs B4) into a regression analysis, model yielded a relatively high coefficient of determination R2 = 0.62 and a low RMSE = 1.86 dS/m.

Transfer Function and Time Series Outlier Analysis: Modelling Soil Salinity in Loamy Sand Soil by Including the Influences of Irrigation Management and Soil Temperature

AbstractIn variable interval irrigation, simply including soil salinity data in the soil salinity model is not valid for making predictions, because changes in irrigation frequency must also be taken into account. This study on variable interval irrigation used capacitance soil sensors simultaneously to obtain hourly measurements of bulk electrical conductivity (σb), soil temperature (t) and soil water content (θ). Observations of σb were converted so that the electrical conductivity of the pore water (σp) could be estimated as an indicator of soil salinity. Values of θ, t and σp were used to test a mathematical model for studying how σp cross‐correlates with t and θ to predict soil salinity at a given depth. These predictions were based on measurements of σp, t, and θ at a shallow depth. As a result, prediction at shallow depth was successful after integrating intervention analysis and outlier detection into the seasonal autoregressive integrated moving average (ARIMA) model. We then used the (multiple‐input/one‐output) transfer function models to logically predict soil salinity at the depths of interest. The model could also correctly determine the effect of the irrigation event on soil salinity. Copyright © 2017 John Wiley & Sons, Ltd.

A simple numerical model to estimate water availability in saline soils

We developed a numerical model to predict soil salinity from knowledge of evapotranspiration rate, crop salt tolerance, irrigation water salinity, and soil hydraulic properties. Using the model, we introduced a new weighting function to express the limitation imposed by salinity on plant available water estimated by the integral water capacity concept. Lower and critical limits of soil water uptake by plants were also defined. We further analysed the sensitivity of model results to underlying parameters using characteristics given for corn, cowpea, and barley in the literature and two clay and sandy loam soils obtained from databases. Results showed that, between two irrigation events, soil salinity increased nonlinearly with decreasing soil water content especially when evapotranspiration and soil drainage rate were high. The salinity weighting function depended greatly on the plant sensitivity to salinity and irrigation water salinity. This research confirmed that both critical and lower limits (in terms of water content) of soil water uptake by plants increased with evapotranspiration rate and irrigation water salinity. Since the presented approach is based on a physical concept and well-known plant parameters, soil hydraulic characteristics, irrigation water salinity, and meteorological conditions, it may be useful in spatio-temporal modelling of soil water quality and quantity and prediction of crop yield.

Soil water-salt dynamics state and associated sensitivity factors in an irrigation district of the loess area: a case study in the Luohui Canal Irrigation District, China

Soil salinisation determines the distribution pattern of crop processes in irrigation districts. The research presented here was conducted in the Luohui Canal Irrigation District, which is located in the loess area of Shaanxi Province, China. A back-propagation artificial neural network (BPANN) for the soil water-salt state was established to predict soil salinity and alkalinity. The degree of influence of numerous factors on the dynamics was quantitatively determined using the default factor testing method and verified with grey relational analysis. The results show that the BPANN prediction accuracy is very high, and it can efficiently depict the comprehensive relationships between the influential factors and dynamic states. The influence of soil moisture, evaporation, groundwater salt and groundwater depth on the dynamics is significant in the region. The current irrigation method, used for many years, cannot meet the water necessities for the vegetation, causing the groundwater levels to decline and a lowering of the soil moisture zone, leading to the occurrence of serious soil salinisation. Under the action of evaporation, more salt accumulates in the upper part of the soil, resulting in extensive soil salinisation. The higher the groundwater salt content, the more salt is carried by rising capillary water, and the more the soil is salinised. If groundwater depth was to exceed the critical water table, then groundwater and salt would move to the soil surface by moisture evaporation, and salt would build up on the soil surface in this irrigation district. The interactions between each factor forms a complex coupling relationship state.

Hierarchical Bayesian models for predicting soil salinity and sodicity characteristics in a coastal reclamation region

Knowledge of soil saline-alkaline distributions and salt migration mechanism is important when developing strategies that reduce soil salinity, improve soil fertility, and enhance soil management in coastal reclamation regions. Soil saline-alkaline characteristics are indicated by the soil salt content (SSC) and sodium adsorption ratio (SAR). However, direct measurements of these parameters are typically expensive, labor intensive and time consuming. Data that is more easily obtained for a property and that may already exist, such as soil organic matter (SOM), may be used to predict SSC and SAR if SOM contents are closely related to soil sodicity and salinity. Data collected in a coastal reclamation area in Rudong County, Jiangsu Province, China, was used to generate regression equations for the relationships between SSC and SOM and between SAR and SOM based on hierarchical Bayesian methods. Results showed that SSC and SAR both decreased with increases in SOM. The deviance information criterion (DIC) indicated that a modified power function model and a log-quadratic model could best describe the two relationships, respectively. Therefore, predicting SSC could be achieved from the relationship between ln[SSC] and a quadratic function of ln[SOM], while SAR used the relationship between ln[SAR] and a quadratic function of SOM. This further suggests that increasing SOM may enhance reductions in soil salinity and sodicity, which would improve soil structure and fertility in the study region and probably at other coastal reclamation sites.

Machine learning performance for predicting soil salinity using different combinations of geomorphometric covariates

Conventional methods of monitoring salt accumulation in irrigation schemes require regular field visits to collect soil samples for laboratory analysis. Identifying areas prone to salt accumulation by means of geomorphometry (i.e. terrain analyses using digital elevation models (DEMs)) can potentially save time and costs. This study evaluated the extent to which DEM derivatives and machine learning (ML) algorithms (k-nearest neighbour, support vector machine, decision tree (DT) and random forest) can be used for predicting the location and extent of salt-affected areas within the Vaalharts and Breede River irrigation schemes of South Africa. In accordance with local management policies, salt-affected areas were defined as regions with soil electrical conductivity (EC) values >4dS/m. Two DEMs, namely the one-arch second Shuttle Radar Topography Mission (SRTM) DEM and a photogrammetrically-extracted digital surface model (DSM), were used for deriving the derivatives. Wetness indices as well as hydrological and morphometric terrain analysis techniques were used to generate predictive variables. For comparative purposes, the predictive variables were also used as input to regression modelling and kriging with external drift (KED). Thresholds were applied to the regression models and KED results to obtain a binary classification. EC values based on in situ soil samples were used for model development, classifier training and accuracy assessment.The results show that KED achieved the highest overall accuracy (OA) in Vaalharts (79.6%), whereas KED and ML (DT) showed the most promise in the Breede River (75%). The findings suggest that the use of elevation data and its derivatives as input to geostatistics and ML holds much potential for monitoring salt accumulation in irrigated areas, particularly for simulating sub-surface conditions. More work is needed to investigate the potential of using ML and DEM-derivatives, along with other geospatial datasets such as satellite imagery (that have been shown to be effective for monitoring surface conditions), for the operational modelling of salt accumulation in large irrigation schemes.

Investigation of soil salinity to distinguish boundary line between saline and agricultural lands in Bonab Plain, southeast Urmia Lake, Iran

Gradual drying of Urmia Lake has left vast saline areas all around it, increasing the risk of salinization of agricultural lands next to the Lake. The current research was aimed to predict soil salinity and distinguish the boundary line between saline and agricultural lands by taking in to account the spatial variability of soil salinity in Bonab Plain, Iran. To do so, soil samples were taken from depth 0-25 cm in 78 points with spatial intervals of 500 m and were analyzed for their electrical conductivity in saturated paste extractions. Data analysis showed that soil salinity mean wasn’t stationary and was varying among dataset. Therefore to build up a variogram, the spatial components of the mean trend were computed and subtracted from laboratory measured ECe values, which resulted in residuals. The semivariogram function was then calculated and modelled based on the residuals. Cross-validation results showed that kriging method along with modified semi-variogram, resulted in better predictions of soil salinity with ME and MSE equal to 0.12 and 0.3. Setting 4 dSm-1 as the limit between saline and non-saline soils in kriging algorithms resulted in a sharp boundary line between saline and non-saline lands in the study area. The presence of highly saline soils next to the agricultural lands in the area can increase the risk of secondary salinization of the Bonab Plain which is one of the important agricultural production centers in the area. Therefore, careful monitoring of lands near salinity boundary in the area should be of high priority. Key Words : Bonab Plain, Kriging, variogram, soil salinity, spatial prediction, Urmia Lake.

基于物候特征的盐渍化信息数据挖掘研究

PDF HTML阅读 XML下载 导出引用 引用提醒 基于物候特征的盐渍化信息数据挖掘研究 DOI: 10.5846/stxb201607201479 作者: 作者单位: 新疆大学资源与环境科学学院,新疆大学资源与环境科学学院,新疆大学资源与环境科学学院,新疆大学资源与环境科学学院,新疆大学资源与环境科学学院 作者简介: 通讯作者: 中图分类号: 基金项目: 新疆维吾尔自治区重点实验室专项基金（2016D03001，2014KL005）；新疆维吾尔自治区科技支疆项目（201591101）；2014级新疆大学博士生科技创新项目（XJUBSCX-2014013）；国家自然科学基金项目（U1303381，41261090，41161063）；教育部促进与美大地区科研合作与高层次人才培养项目 Research on data mining of salinization information based on phenological characters Author: Affiliation: Xinjiang University,College of Resources and Environment Science,Xinjiang University,College of Resources and Environment Science,,, Fund Project: 摘要 | 图/表 | 访问统计 | 参考文献 | 相似文献 | 引证文献 | 资源附件 | 文章评论 摘要:盐渍化是影响植被和作物长势的重要因素，精确反演盐渍化的时空分布信息至关重要。基于MOD13A1-NDVI数据反演生长季开始日期（SOS）、生长季结束日期（EOS）、生长季长度（LEN）等物候参数和计算出能高精度反演盐渍化空间分布的多种植被指数、盐分指数、地形指数、干旱指数等参数后作为BP-ANN人工神经网络的输入因子来反演盐渍化信息，同时按照植被类型和地貌类型进行分区来反演盐渍化信息，以探讨盐渍化受植被和地貌类型的影响。主要结论如下：①盐渍化的形成受多种因素的影响，与物候参数大多呈非线性关系，不能单纯的以某拟合公式来进行表达，需要借助人工神经网络超强的非线性拟合能力来反演盐渍化信息。②通过深入挖掘植被物候信息，在融入物候参数后的反演精度显著提高。可决系数R2从0.68（非物候参数）增加到0.79（包括物候参数），但是需要加入地形、影像数据和土壤水分等方面的信息来更加精确的反演盐渍化信息。生物累积量指标LSI（Large seasonal integral）和SSI（Small seasonal integral）能够很好的表征盐渍化的信息。③划分植被类型后的盐渍化提取精度进一步提高，可决系数R2达到了0.88。④以地貌特征作为类型分区后，反演结果的R2达到了0.85，精度较高，比以植被类型作为分区的精度略小。高程较低区域的盐渍化现象普遍较重，盐渍化程度受到地形和地貌因素的影响显著。⑤农用地区域多为非盐渍化和轻度盐渍化地，稀疏植被区多为重盐渍化地。研究区的非盐渍化和轻盐渍化地、中盐渍化地和重度盐渍化地比例分别为53.42%，13.71%，32.87%。以上的研究结果提出了一种融合物候信息和非物候参数来反演盐渍化信息的方法，进行深入的协同植被物候监测盐渍化信息方面的数据挖掘，在融入了物候参数后，盐渍化的预测精度显著提高。 Abstract:Soil salinization is an important factor that affects crop and vegetation growth condition and can result in environmental impacts with considerable economic consequences. Therefore, it is necessary to determine an effective method to monitor spatiotemporal salinity distribution. We used MOD13A1 time-series NDVI data to determine the vegetation phenology, including start of season (SOS), end of season (EOS), length of season (LEN), etc., and calculated several vegetation, salinity, terrain, and drought indexes, and spatial models. These were used as input parameters for the BP-ANN model. Meanwhile, we predict the soil salinity through vegetation and geomorphological partitioning, which described the correlations between vegetation or geomorphic type and salinization. The main conclusions are as follows: salinity is influenced by many factors, and many of them show non-linear relationships between phenological indicators and salinization, so we utilized artificial neural networks to predict soil salinity than mathematical equations; through a combination of phenology parameters, the precision of inversion salinity R2 improved from 0.68 (no phenologcial indicators were included) to 0.79 (phenological indicators were included). However, additional auxiliary data to predict soil salinity, such as terrain, image, and soil moisture parameters should also be included. After the classification of the vegetation, the inversion precision improved obviously, where R2 increased to 0.88. Phenological characters, such as large seasonal integrals (LSIs) and small seasonal integrals (SSIs) are good indicators to represent soil salinity. After geomorphological partitioning, R2 increased to 0.85, indicating that it could be a good salinity predictor, but the ability of comprehensive inversion was lower than vegetation type partitioning. In farmland, the salinity level was low. The low, intermediate, and high salinization was 53.42, 13.71, and 32.87% respectively. Generally, salinization was higher at lower altitudes, and the salinity level was affected by terrain and geomorphological factors. The above conclusions indicate an effective method for the inversion of salinization levels that combines phenology and other parameters for comprehensively determining the effect of phenological information on salinity monitoring ability in data mining. The inversion of soil salinity is enhanced by the inclusion of phenological parameters. 参考文献 相似文献 引证文献

Comparative Study among Different Semi-Empirical Models for Soil Salinity Prediction in an Arid Environment Using OLI Landsat-8 Data

Salt-affected soils, caused by natural or human activities, are a common environmental hazard in semi-arid and arid landscapes. Excess salts in soils affect plant growth and production, soil and water quality and, therefore, increase soil erosion and land degradation. This research investigates the performance of five different semi-empirical predictive models for soil salinity spatial distribution mapping in arid environment using OLI sensor image data. This is the first attempt to test remote sensing based semi-empirical salinity predictive models in this area: the Kingdom of Bahrain. To achieve our objectives, OLI data were standardized from the atmosphere interferences, the sensor radiometric drift, and the topographic and geometric distortions. Then, the five semi-empirical predictive models based on the Normalized Difference Salinity Index (NDSI), the Salinity Index-ASTER (SI-ASTER), the Salinity Index-1 (SI-1), the Soil Salinity and Sodicity Index-1 and Index-2 (SSSI-1 and SSSI-2), developed for slight and moderate salinity in agricultural land, were implemented and applied to OLI image data. For validation purposes, a fieldwork was organized and different important spots-locations representing different salinity levels were visited, photographed, and localized using an accurate GPS (σ ≤ ±30 cm). Based on this a priori knowledge of the soil salinity, six validation sites were selected to reflect non-saline, low, moderate, high and extreme salinity classes, descriptive statistics extracted from polygons and/or transects over these sites were used. The obtained results showed that the models based on NDSI, SI-1 and SI-ASTER all failed to detect salinity bounds for both extreme salinity (Sabkhah) and non-saline conditions. In Fact, NDSI and SI-ASTER gave respectively only 35% dS/m and 25% dS/m in extreme salinity validation site, while SI-1 and SI-ASTER indicated 38% dS/m and 39% dS/m in non-saline validation site. Therefore, these three models were deemed inadequate for the study site. However, both SSSI-1 and SSSI-2 allowed a detection of the previous salinity bounds and furthermore described similarly and correctly the urban-vegetation areas and the open-land areas. Their predicted EC is around 10% dS/m for non-saline urban soil, about 25% dS/m for low salinity urban-vegetation soil, approximately 30% to 75% dS/m, respectively, for moderate to high salinity soils. SSSI-2 based semi-empirical salinity models was able to differentiate the high salinity versus extreme salinity in areas where both exist and was very accurate to highlight the pure salt where SSSI-1 has reach saturation for both salinity classes. In conclusion, reliable salinity map was produced using the model based on SSSI-2 and OLI sensor data that allows a better characterization of the soil salinity problem in an Arid Environment.

Quantitative remote sensing of soil electrical conductivity using ETM+ and ground measured data

ABSTRACTIn this article, the possibility of predicting soil electrical conductivity (EC) from the enhanced thematic mapper plus (ETM+) data is studied using the multiple regression technique. In this regard, soil EC was measured in 188 points in an area of 5000 ha, Western Urmia Lake, northwest of Iran. ETM+ data for sampling dates were also acquired. Then, the measured EC (as the dependent variable) and reflectance from the original bands of ETM+ data and their ratios besides the different extracted indices (as independent variables) were used to construct different regression relations to predict soil salinity. Different data reduction algorithms, including principal component (PC) analysis, minimum noise fraction (MNF) transformation, pixel purity index (PPI), and n-dimensional visualizer (nDV) algorithms, were also applied to extract the independent factors of original bands. During the construction of regression relations, different criteria including normalized difference vegetation index (NDVI) and (R is reflectance and NIR and SWIR1 are near-infrared and shortwave infrared 1 bands of ETM+ data, respectively) were applied to group data sets in order to increase the prediction accuracies. Results revealed that the constructed regression relations were robust enough to predict the soil salinity showing adjusted determination coefficient (R2) up to 0.875. The best equation was obtained for the data set with NDVI values higher than 0.35. In general, the results show that the multiple regression technique, along with remotely sensed data, has enough accuracy to predict soil salinity.

Prediction of soil salinity in the Yellow River Delta using geographically weighted regression

ABSTRACTIt is essential to determine the content and spatial distribution of soil salinity in a timely manner because soil salinization can cause land degradation on a regional scale. Geographically weighted regression (GWR) is a local regression method that can achieve the spatial extension of dependent variables based on the relationships between the dependent variables and environment variables and the spatial distances between the sample points and predicted locations. This study aimed to explore the feasibility of GWR in predicting soil salinity because the existing interpolation methods for soil salinity in the Yellow River Delta are still of low precision. Additionally, multiple linear regressions, cokriging and regression kriging were added to compare the accuracy of GWRs. The results showed that GWR predicted soil salinity with high accuracy. Furthermore, the accuracy was improved when compared to other methods. The root mean square error, correlation coefficient, regression coefficient and adjustment coefficients between the observed values and predicted values of the validation points were 0.31, 0.65, 0.57 and 0.42, respectively, which were better than that of other methods, indicating that GWR is an optimal method.

Prediction of Soil Salinity Using Near-Infrared Reflectance Spectroscopy with Nonnegative Matrix Factorization.

As a key, yet difficult, issue currently in the quantitative remote sensing analysis of soil, the accurate and stable monitoring of soil salinity content (SSC) in situ should be studied and improved. The purpose of this study is to explore the method of fusing spectra outdoors with spectra indoors and improve the estimation precision of SSC based on near-infrared (NIR) reflectance hyper-spectra. First, samples of saline soil from the Yellow River delta of China were collected and analyzed. We measured three groups of sample spectra using a spectrometer: (1) situ-spectra, measured at sampling points in situ; (2) out-spectra, measured outdoors on air-dried samples; and, (3) lab-spectra, measured in a dark laboratory with the above air-dried samples. Second, four algorithms (multiplicative update, alternating least-squares, sparse affine non-negative matrix factorization (NMF), and gradient projection algorithms) of NMF were used to fuse the situ-spectra or out-spectra with the lab-spectra for the calibration of SSC. Finally, estimation models of SSC were built using the multiple linear regression method based on the first derivatives of the un-fused and fused spectra. The results indicate that using the NMF method to fuse the situ-spectra or out-spectra with the lab-spectra can heighten the correlation between SSC and the outdoor spectra in most wavelength ranges and improve the accuracy of the prediction model. The gradient projection algorithm shows the best performance with fewer variables and highest accuracy of the SSC model based on the NIR spectra.

Modification of transient state analytical model under different saline groundwater depths, irrigation water salinities and deficit irrigation for quinoa

Salinization of soil is primarily caused by capillary rise from saline shallow groundwater or application of saline irrigation water. In this investigation, the transient state analytical model was modified to predict water uptake from saline shallow groundwater, actual crop evapotranspiration, soil water content, dry matter, seed yield and soil salinity under different saline groundwater depths, irrigation water salinities and deficit irrigation for quinoa. Considering the effect of salinity on soil saturated hydraulic conductivity and maximum root depth in presence of shallow saline groundwater, the model resulted in good agreement between the measured and predicted saline groundwater uptake, soil salinity increase at different groundwater depths (300-800 mm) and water salinity (10-40 dS m -1 ). Therefore, the modified model is applicable for quinoa yield and soil salinity prediction and it could be a valuable tool for soil salinity management in presence of shallow saline groundwater. Furthermore, prediction of quinoa yield by the modified model can be used for better irrigation water salinity management under different saline groundwater depths, irrigation water salinities and deficit irrigation.