Inclusion of fractal dimension in four machine learning algorithms improves the prediction accuracy of mean weight diameter of soil

The study of soil mean weight diameter (MWD), essential for sustainable soil management, has recently received much attention. As the estimation of MWD is challenging, labor-intensive, and time-consuming, there is a crucial need to develop a predictive estimation method to generate helpful information required for the soil health assessment to save time and cost involved in soil analysis. Pedotransfer functions (PTFs) are used to estimate parameters that are ‘difficult to measure’ and time-consuming with the help of ’easy to measure’ parameters. In the current study, empirical PTFs, i.e., multi-linear regression (MLR), and four machine learning based PTFs, i.e., artificial neural network (ANN), support vector machine (SVM), classification and regression trees (CART), and random forest (RF) were used for mean weight diameter prediction in Karnal district of Haryana, India. A total of 121 soil samples from 0‐15 and 15‐30 cm soil depths were collected from seventeen villages of Nilokheri, Nissing, and Assandh blocks of Karnal district. Soil parameters such as bulk density (BD), fractal dimension (D), soil texture (i.e., sand, silt, and clay), organic carbon (OC), and glomalin content were used as the input variables. Two input combinations, i.e., one with texture data (dataset 1) and the other with fractal dimension data replacing texture (dataset 2), were used, and the complete dataset (121) was divided into training and testing datasets in a 4:1 ratio. The model performance was evaluated by statistical parameters such as mean absolute error (MAE), mean absolute percentage error (MAPE), root mean square error (RMSE), normalized root mean square error (NRMSE), and determination coefficient (R²). The comparison results showed that including the fractal dimension in the input dataset improved the prediction capability of ANN, SVM, and RF. MLR and CART showed lower predictive ability than the other three approaches (i.e., ANN, SVM, and RF). In the training dataset, RMSE (mm) for the SVM model was 8.33% lower with D than with texture as the input, whereas, in the testing dataset, it was 16.67% lower. Because SVM is more flexible and effectively captures non-linear relationships, it performed better than the other models in predicting MWD. As seen in this study, the SVM model with input data D is the best in its class and has a high potential for MWD prediction in the Karnal district of Haryana, India.     

基于数字土壤制图技术的土壤有机碳储量估算

精准的土壤属性空间分布信息有助于提升土壤有机碳储量估算的精度。本研究以河南省济源市南山林场为研究区,以地形因子为预测因子,利用模糊C均值(FCM)聚类方法对土壤有机碳含量、土壤容重、土壤厚度和土壤砾石含量进行数字土壤预测制图,基于数字制图结果实现土壤有机碳密度预测制图和土壤有机碳储量估算。结果表明: 基于数字土壤制图方法得到的研究区土壤有机碳密度平均值为4.24 kg·m^-2,其预测图的平均误差(ME)为0.08 kg·m^-2,平均绝对误差(MAE)为2.80 kg·m^-2,均方根误差(RMSE)为5.03 kg·m^-2,与传统类型方法相比,预测结果的精度和稳定性更高,具有较高的可信度,最终估算得到研究区土壤有机碳储量为3.08×10⁸ kg。基于数字土壤制图技术仅采用少量土壤样点即可实现较高精度的土壤有机碳密度制图和储量估算,且能表征土壤有机碳密度空间分布特征。本研究为土壤有机碳储量估算提供了新途径,有助于提升土壤有机碳储量估算的精度和效率。     

A comparative assessment of support vector regression,artificial neural networks,and random forests for predicting and mapping soil organic carbon stocks across an Afromontane landscape

Soil organic carbon (SOC) is a key indicator of ecosystem health, with a great potential to affect climate change. This study aimed to develop, evaluate, and compare the performance of support vector regression (SVR), artificial neural network (ANN), and random forest (RF) models in predicting and mapping SOC stocks in the Eastern Mau Forest Reserve, Kenya. Auxiliary data, including soil sampling, climatic, topographic, and remotely-sensed data were used for model calibration. The calibrated models were applied to create prediction maps of SOC stocks that were validated using independent testing data. The results showed that the models overestimated SOC stocks. Random forest model with a mean error (ME) of −6.5 Mg C ha⁻¹ had the highest tendency for overestimation, while SVR model with an ME of −4.4 Mg C ha⁻¹ had the lowest tendency. Support vector regression model also had the lowest root mean squared error (RMSE) and the highest R² values (14.9 Mg C ha⁻¹ and 0.6, respectively); hence, it was the best method to predict SOC stocks. Artificial neural network predictions followed closely with RMSE, ME, and R² values of 15.5, −4.7, and 0.6, respectively. The three prediction maps broadly depicted similar spatial patterns of SOC stocks, with an increasing gradient of SOC stocks from east to west. The highest stocks were on the forest-dominated western and north-western parts, while the lowest stocks were on the cropland-dominated eastern part. The most important variable for explaining the observed spatial patterns of SOC stocks was total nitrogen concentration. Based on the close performance of SVR and ANN models, we proposed that both models should be calibrated, and then the best result applied for spatial prediction of target soil properties in other contexts.     

基于地理加权回归克里格模型的帽儿山地区森林碳储量空间分布

森林碳储量对于全球气候变化具有重要影响,以往的模型估算未考虑到模型残差的空间相关性和碳储量数据的非平稳性,影响模型的预测精度.本研究基于东北林业大学帽儿山实验林场的ETM+遥感影像数据和193块固定样地,利用地理加权克里格回归(GWRK)建立森林碳储量与遥感和地形因子的回归模型,同时对比最小二乘模型(OLS)、地理加权回归模型(GWR)的预测精度.结果表明: 对于帽儿山地区的森林碳储量估算,GWRK的平均绝对误差(MAE)、均方根误差(RMSE)低于OLS模型和GWR模型,GWRK模型的平均误差(ME)低于GWR模型,与OLS模型相近.GWRK模型的预测精度为83.2%,较OLS模型(73.7%)和GWR模型(77.3%)分别提高6%和10%,拟合精度明显提高,说明GWRK模型是森林碳储量估算的有效方法.利用GWRK模型预测的研究区森林碳储量平均值为70.31 t·hm^-2,在海拔较高的地区,森林碳储量值相对较高,说明海拔对其有较大影响.     

Artificial neural networking to estimate the leaf area for invasive plant Wedelia trilobata

Leaf area are very important parameter for the understanding of growth and physiological responses of invasive plant species under different environmental factors. This study was conducted to build non-destructive leaf area model of Wedelia trilobata that were grown in greenhouse. Regression analysis and artificial neural network (ANN) approaches were used for the development of leaf area model with the help of leaf length and width of 262 plants samples. In selection of best method under both techniques, the lower value of mean absolute error (MAE), root mean square error (RMSE), mean absolute percentage error (MAPE) and higher value of R² were considered. According to the results it was found that ANN have higher value of (R² = 0.96) and lower value of error (MAE = 0.023, RMSE = 0.379, MAPE = 0.001) than regression analysis (R² = 0.94, MAE = 0.111, RMSE = 1.798, MAPE = 0.0005). It was concluded that error between predicted and actual value was less under ANN. Therefore, ANN model approach can be used as an alternating method for the estimation of leaf area. Through estimation of leaf area, invasive plant growth can predict under different environment conditions.     

基于神经网络模型和地统计学方法的土壤养分空间分布预测

采用径向基函数神经网络模型与普通克里格法相结合的方法,预测川中丘陵区县域尺度土壤养分(有机质和全氮)的空间分布,并与普通克里格法和回归克里格法进行比较.结果表明：各方法对研究区土壤养分的预测结果相似.与多元回归模型相比,神经网络模型对验证样点土壤有机质和全氮的预测值与样点实测值的相关系数分别提高了12.3%和16.5%,表明神经网络模型能更准确地捕捉土壤养分与定量环境因子间的复杂关系.对469个验证样点预测结果的误差分析表明,神经网络模型与普通克里格法相结合的方法对土壤有机质和全氮预测结果的平均绝对误差、平均相对误差、均方根误差较普通克里格法分别降低了6.9%、7.4%、5.1%和4.9%、6.1%、4.6%,降低幅度达到极显著水平(P<0.01);与回归克里格法相比则分别降低了2.4%、2.6%、1.8%和2.1%、2.8%、2.2%,降低幅度达显著水平(P<0.05).     

Toxic Effects of Cadmium on Morus alba L. and Bombyx moril L.

A 3-year micro-plot experiment of mulberry cultivation with Cd-polluted soil and silkworm breeding experiments by feeding with exogenous or endogenous –Cd-polluted mulberry leaves were conducted to evaluate the toxic effects of Cd on mulberry and silkworms. There was no apparent harmful effect on mulberry plant growth at a soil Cd content of 8.49 mg/kg. At a soil Cd content of 75.8 mg/kg, a yield reduction in the leaves became apparent, whereas at 145 mg Cd/kg soil, the plants exhibited marginal growth. There was a significant decrease of the ingestive rate of leaves at an exogenous Cd content of 1.98 mg Cd/kg leaf, whereas the digestive rate decreased significantly at an exogenous Cd content of 0.50 mg/kg. The endogenous Cd content that significantly affected the ingestive and digestive rates was 1.66 mg/kg. The influence of exogenous and endogenous Cd on the biomass of the silkworms decreased with increase of the instar stage. The weight of cocoons and rate of silk reeling was significantly reduced at exogenous and endogenous Cd contents of 5.17 mg/kg and 1.66 mg/kg, respectively.     

基于神经网络模型和地统计学方法的土壤养分空间分布预测

采用径向基函数神经网络模型与普通克里格法相结合的方法,预测川中丘陵区县域尺度土壤养分(有机质和全氮)的空间分布,并与普通克里格法和回归克里格法进行比较.结果表明：各方法对研究区土壤养分的预测结果相似.与多元回归模型相比,神经网络模型对验证样点土壤有机质和全氮的预测值与样点实测值的相关系数分别提高了12.3%和16.5%,表明神经网络模型能更准确地捕捉土壤养分与定量环境因子间的复杂关系.对469个验证样点预测结果的误差分析表明,神经网络模型与普通克里格法相结合的方法对土壤有机质和全氮预测结果的平均绝对误差、平均相对误差、均方根误差较普通克里格法分别降低了6.9%、7.4%、5.1%和4.9%、6.1%、4.6%,降低幅度达到极显著水平(P<0.01);与回归克里格法相比则分别降低了2.4%、2.6%、1.8%和2.1%、2.8%、2.2%,降低幅度达显著水平(P<0.05).     

N、P、K肥对香根草修复土壤镉、锌污染效率的影响

通过盆栽试验研究在30 mg/kg镉(Cd)污染土壤条件下N[CO(NH2)2:100、200、300 mg/kg土]、P(P2O5:50、100、200 mg/kg土)和K(KCl:100、200、300 mg/kg土)处理对香根草修复土壤Cd和锌(Zn)污染效率的影响。结果表明:3种N处理能促进香根草地上部生长,而且显著提高地上部特别是叶的Cd和Zn含量,导致其修复效率成倍显著增加;200 mg/kg K处理显著提高Zn修复效率,但300 mg/kg K和50、200 mg/kg P处理却显著降低Cd、Zn修复效率。因此,为改善香根草对较贫瘠土壤中Cd、Zn污染的修复效率,应对香根草适施N肥,并控制或者不施P、K肥为佳。     

基于混合效应的大别山地区杉木树高-胸径模型比较

基于安徽省大别山区马鬃岭林场杉木人工林30块样地1087组数据,选用7个常用树高-胸径(H-D)模型(线性模型、Chapman-Richards模型、Logistic模型等),采用最小二乘法拟合并选出最优基础模型(式11,只含D变量的Chapman-Richards模型),然后基于该模型构建含林分变量优势木平均高度、密度的H-D模型(式12),同时考虑样地水平的随机效应,分别基于式11、12构建混合模型(式13、14),并用幂函数、指数函数消除误差异方差,利用决定系数(R²)、均方根误差(RMSE)、平均绝对误差(MAE)和平均相对误差绝对值(MAPE)等指标来评价模型的拟合与预测能力,最终获取最优树高预测模型.结果表明:含林分变量的模型的拟合精度(式12,R²=0.863、RMSE=1.381、MAE=0.971)优于基础模型(式11,R²=0.827、RMSE=1.554、MAE=0.101).对于误差方差,幂函数、指数函数均能较好地消除异方差,但幂函数相对最好.混合模型的拟合与预测能力均优于式11、12,但混合模型(式13、14)之间的拟合与预测精度相差不大.基于混合效应的H-D模型(式13)能够较好地描述不同林分间H-D关系的差异,实际运用中可选用该模型来预测杉木树高,具有较高的预测精度.     

基于几丁质酶特征序列预测其最适pH的研究

基于Z标度法对C/18家族几丁质酶的特征序列PS01095和PS00232进行数字转换,将得到的数据集采用逐步回归方法回归预测,构建了几丁质酶特征序列与其最适pH间关系的数学模型.当模型的相关系数为R＝0.964,显著水平P<0.001,得到了最佳的预测效果.模型对pH值拟合的平均绝对百分比误差为0.05 %,同时具有良好的预测效果,预测的平均绝对误差为0.26 个pH单位,比基于几丁质酶氨基酸组成的支持向量机模型更好.     

Sentinel-2 Satellite Imagery Application to Monitor Soil Salinity and Calcium Carbonate Contents in Agricultural Fields

The estuary tides affect groundwater dynamics; these areas are susceptible to waterlogging and salinity issues. A study was conducted on two fields with a total area of 60 hectares under a center pivot irrigation system that works with solar energy and belong to a commercial farm located in Northern Sudan. To monitor soil salinity and calcium carbonate in the area and stop future degradation of soil resources, easy, non-intrusive, and practical procedures are required. The objective of this study was to use remote sensing-determined Sentinel-2 satellite imagery using various soil indices to develop prediction models for the estimation of soil electrical conductivity (EC) and soil calcium carbonate (CaCO₃). Geo-referenced soil samples were collected from 72 locations and analyzed in the laboratory for soil EC and CaCO₃. The electrical conductivity of the soil saturation paste extract was represented by average values in soil dataset samples from two fields collected from the topsoil layer (0 to 15 cm) characteristic of the local salinity gradient. The various soil indices, used in this study, were calculated from the Sentinel-2 satellite imagery. The prediction was determined using the root mean square error (RMSE) and cross validation was done using coefficient of determination. The results of regression analysis showed linear relationships with significant correlation between the EC analyzed in laboratory and the salinity index-2 “SI2” (Model-1: R² = 0.59, p = 0.00019 and root mean square error (RMSE = 1.32%) and the bare soil index “BSI” (Model-2: R² = 0.63, p = 0.00012 and RMSE = 6.42%). Model-1 demonstrated the best model for predicting soil EC, and validation R² and RMSE values of 0.48% and 1.32%, respectively. The regression analysis results for soil CaCO₃ determination showed linear relationships with data obtained in laboratory and the bare soil index “BSI” (Model- 3: R² = 0. 45, p = 0.00021 and RMSE = 1.29%) and the bare soil index “BSI” & Normalized difference salinity index “NDSI” (Model-4: R² = 0.53, p = 0.00015 and RMSE = 1.55%). The validation confirmed the Model-3 results for prediction of soil CaCO₃ with R² and RMSE values of 0.478% and 1.29%, respectively. Future soil monitoring programs might consider the use of remote sensing data for assessing soil salinity and CaCO₃ using soil indices results generated from satellite image (i.e., Sentinel-2).     

Soil Nitrogen Fertilization Effects on Phytoextraction of Cadmium and Lead by Tobacco (Nicotiana tabacum L.)

A greenhouse experiment using 24 plastic pots filled with 6 kg of Pb- and Cd-contaminated soil was carried out. In all 24 pots, soils were heavy metal–contaminated with 10 mg Cd kg^?1 soil and 500 mg of Pb kg^?1 soil by using CdCl and PbNO₃. Two-month-old tobacco (Nicotiana tabacum L.) plants were used to extract these heavy metals. Results showed that tobacco is able to remove Cd and Pb from contaminated soils and concentrate them in its harvestable part, that is, it could be very useful in phytoextraction of these heavy metals. Increasing additions of ammonium nitrate to soil (50, 100, and 150 mg N kg^?1 soil) significantly (p ≤ .05) increased aboveground Cd and Pb accumulation during a 50-day experimental period, whereas increasing additions of urea to soil (50 and 100 mg N kg^?1 soil) did not show these effects at the same significance levels. Increasing additions of ammonium nitrate to soil shows as dry matter increases, both accumulated Cd and accumulated Pb also increase when tobacco plants are growing under Pb- and Cd-contaminated soil conditions. Higher Pb concentrations depress Cd/Pb ratios for concentrations and accumulations, suggesting that Pb negatively affects Cd concentration and/or accumulation.     

Spatio-temporal air quality analysis and PM2.5 prediction over Hyderabad City,India using artificial intelligence techniques

Air pollution is one of the most serious environmental issues faced by humans, and it affects the quality of life in cities. PM_2.5 forecasting models can be used to create strategies for assessing and warning the public about anticipated harmful levels of air pollution. Accurate pollutant concentration measurements and forecasting are critical criteria for assessing air quality and are the foundation for making the right strategic decisions. Data-driven machine learning models for PM_2.5 forecasting have gained attention in the recent past. In this study, PM_2.5 prediction for Hyderabad city was carried out using various machine learning models viz. Multi-Linear Regression (MLR), decision tree (DT), K-Nearest Neighbors (KNN), Random Forest (RF), and XGBoost. A deep learning model, the Long Short-Term Memory (LSTM) model, was also used in this study. The results obtained were finally compared based on error and R² value. The best model was selected based on its maximum R² value and minimal error. The model's performance was further improved using the randomized search CV hyperparameter optimization technique. Spatio-temporal air quality analysis was initially conducted, and it was found that the average winter PM_2.5 concentrations were 68% higher than the concentrations in summer. The analysis revealed that XGBoost regression was the best-performing machine learning model with an R² value of 0.82 and a Mean Absolute Error (MAE) of 7.01 μg/ m³, whereas the LSTM deep learning model performed better than XGBoost regression for PM_2.5 modeling with an R² value of 0.89 and an MAE of 5.78 μg/ m³.     

基于随机森林法的山核桃林地土壤速效养分含量空间分布特征研究

土壤氮磷钾是土壤肥力管理的重要元素,是植物生长的必要养分元素。对土壤氮磷钾的空间分布进行特征解译,可为精准管理临安山核桃产区林地土壤肥力,促进山核桃林产业可持续发展提供理论依据。研究以临安山核桃主产区为研究区域,利用随机森林（RF）、普通克里格（OK）和Shapley加性解释（SHAP）方法,结合地形因子、气候因子、土壤因子、遥感因子等环境变量,对山核桃林地土壤碱解氮（AN）、有效磷（AP）、速效钾（AK）的空间分布特征进行分析。研究结果表明：相比于OK模型,基于环境协变量所构建的RF模型对AN、AP和AK含量空间分布预测表现最佳,R²分别为0.68、0.60和0.64,均方根误差（RMSE）分别为20.005、10.287和22.426,平均绝对误差（MAE）分别为15.425、7.709 和21.628。RF模型SHAP分析显示,AN和AK含量分布主要受土壤有机质（SOM）的影响,并且SOM与AN和AK存在正相关性;AP主要受pH的影响,其次为色调指数,AP与pH和色调指数均具有负相关性;AK和AP同时受到海拔和坡向的影响。两种模型预测的氮磷钾空间分布趋势总体相似,不同速效养分存在明显的空间异质性。碱解氮高值区域主要分布于研究区东部;有效磷高值区域主要分布于研究区西部,但分散度高;速效钾高值区域则主要分布于研究区中部。总体而言,基于随机森林模型可以高精度模拟山核桃林地土壤氮磷钾含量空间分布特征,并依据主要环境协变量对土壤氮磷钾的影响关系,提出相应改良措施。在有效磷含量低值区域可以施用石灰来缓解土壤酸化,同时补追磷肥;碱解氮含量高值区域可以合理减少氮肥施用;速效钾含量低值区域合理施加钾肥;对于海拔较高及迎风坡多降雨的区域,可以构建林下高效水土保持植被,减轻水土流失;在林地施用有机肥料,改善土壤理化性质,增加土壤养分含量。     

露地栽培向设施栽培转变对菜田土壤重金属含量及形态的影响

为了研究露地栽培向设施大棚栽培转变对土壤重金属含量的影响,对武汉市郊区露地和设施塑料大棚两种栽培条件下菜田土壤重金属Cd、Cr和Pb各形态含量及分布特征进行了研究。结果显示,露地和大棚栽培条件下土壤重金属元素Cr和Pb各种形态含量之间没有明显差异,但Cd各种形态含量间有显著差异;从露地到大棚,土壤中Cd酸可提取态含量从露地的0.62 mg/kg上升到大棚的1.19 mg/kg,其次是Cd残渣态、有机结合态、氧化态和碳酸盐结合态;Cd总含量从露地的0.79 mg/kg升高到大棚的1.58 mg/kg,显著超过土壤环境质量标准中的Cd含量标准值(0.3 mg/kg),达到严重污染水平。Cd碳酸盐结合态和氧化态占总量的比例有所降低,而酸可提取态占总量的比例有所升高。说明从露地到设施大棚栽培,促使了土壤中部分Cd碳酸盐结合态和氧化结合态向酸可提取态转变,提高了土壤中Cd的生物有效性。因此,在设施大棚栽培快速发展的情况下,要加强重金属Cd对土壤污染的治理,减少重金属Cd对蔬菜的毒害。     

Cd and Pb Contents in Soil,Plants, and Grasshoppers along a Pollution Gradient in Huludao City,Northeast China

Cd and Pb contents in soil, plants, and two grasshopper species (Locusta migratoria manilensis and Acrida chinensis) were examined to quantify the influence ranges of zinc smelting on heavy metal contamination. Samples were collected simultaneously from Huludao City, a chemical and nonferrous smelting base in Northeast China. Cd and Pb contamination in soil and plants were serious. Cd and Pb contents were 13.32 and 8.83 mg/kg in L. migratoria manilensis and 16.67 and 15.00 mg/kg in A. chinensis, respectively. Correlation analysis indicated the same metal source for Cd and Pb in soil, plants, and grasshoppers. Cd and Pb contents in soil, plants, and grasshoppers were all significantly related to distances far from the zinc smelter in good negative logarithm model. The fitting curves indicated that the influence radius of the smelter on heavy metal contamination was about 4,000 m for soil and plants and about 2,000 m for grasshoppers.     

人工长白落叶松立木叶面积预估模型

叶面积影响着树木干物质的生产,进而影响树木乃至整个林分的生长,而叶面积准确估计对分析树木和林分生长具有重要作用.本研究基于黑龙江省长白落叶松人工林中76株解析木数据,分别建立枝条层面和单木层面的叶面积预估模型.结果表明: 考虑样木层次随机效应的最优枝条叶面积混合效应模型包含lnBD(BD为枝条基径)、lnRDINC(RDINC为相对着枝深度)和lnCR(CR为冠长率)3个随机效应参数,具体形式为:lnBLA=β₁+(β₂+b₂)lnBD+(β₃+b₃)lnRDINC+β₄lnDBH+β₅lnHT/DBH+(β₆+b₆)lnCR,其中:β_i和b_i分别是模型的固定效应参数和随机效应参数;DBH为树木胸高处直径;HT/DBH为树高与胸径的比值.模型的修正决定系数(R_a²)为0.90,均方根误差(RMSE)为0.5477,平均偏差(ME)为-0.03,平均绝对偏差(MAE)为0.24,预测精度(P)为91%,枝条叶面积预估模型的预估效果较好.以枝条叶面积预估模型为基础,计算树冠叶面积并建立树冠叶面积预估模型,最终形式为:lnCLA=γ₀+γ₁lnDBH+γ₂CR,其中,γ_i为模型参数.似然比检验结果(P>0.05)说明该模型不用考虑样地层次的随机效应.本研究所建立的立木树冠叶面积预估模型的决定系数(R²)为0.87,RMSE为0.3847,拟合效果好,可以很好地预测人工长白落叶松立木树冠叶面积,为以后叶面积分布和光合作用的研究提供了理论基础.     

镉胁迫下植物促生菌密歇根克雷伯氏菌TS8和Lelliottia jeotgali MR2对拟南芥生长及镉富集的影响北大核心CSCD

为了研究镉胁迫下植物促生菌密歇根克雷伯氏菌(Klebsiella michiganensis)TS8和Lelliottia jeotgaliMR2对拟南芥(Arabidopsis thaliana)生长及镉富集的影响,文中以野生型拟南芥为试验材料,将其种植在不同镉浓度的土壤基质中,并施入MR2和TS8菌悬液。低浓度镉处理组(LC)为购买的基质营养土,初始镉浓度为14.17 mg/kg,高浓度处理组(HC)为在购买的基质营养土上额外喷洒200 mg/kg Cd^(2+)。结果表明,相比对照组,不同浓度镉胁迫下喷施MR2菌悬液均可显著促进拟南芥的生长,而TS8和MR2_TS8混合菌液仅在高浓度镉胁迫下表现出一定的促生效果。但值得关注的是,不同浓度镉胁迫下TS8菌悬液可显著降低拟南芥的地下部对重金属镉的富集(60%和59%),并有效提高地上部对重金属镉的富集(234%和35%)。此外,单菌和混合菌均能显著提高土壤中可还原态镉向酸可提取态镉转化,促进植物吸收,降低土壤总镉含量。因此,针对不同环境下,合理配施植物促生细菌在提高作物产量或修复土壤镉污染中具有一定的应用价值。     

Spatio-temporal modelling of particulate matter concentrations using satellite derived aerosol optical depth over coastal region of Chennai in India

The present study primarily focuses on describing aerosol optical depth (AOD), its distribution pattern and seasonal variation, and modelling Particulate Matter Concentrations in Chennai. The frequency distribution of AOD and PM_2.5 demonstrates that AOD can be used as a proxy for estimating PM_2.5 in the study region as the occurrence of AOD almost resonates with that of PM_2.5. The seasonal variation of AOD and PM_2.5 revealed that during the winter (October–January) and summer (February–May) seasons, AOD reasonably followed the trend of PM_2.5. However, during the monsoon period, AOD showed random variations. Different models like linear and non-linear regression models and machine learning models such as random forest (RF) have been developed for PM_2.5 estimation. The model's performance in different stations and seasons has been assessed. The effect of meteorology and other factors in the model has also been assessed. From linear and non-linear model analysis, AOD was a significant parameter in estimating PM_2.5. The Random Forest model was the stable model for the study region, with a model R² of 0.53 and an RMSE of 15.89 μg/m³. The inclusion of meteorological parameters like relative humidity, wind speed, and wind direction decreased the error in prediction by 17.45 μg/m³. The seasonal and spatial analysis indicates that the prediction capability of models varies with stations and seasons. The best performing model was found to be Model RF, and the model could explain about 53.14% of the variability in PM_2.5 concentration occurrence in the study region with a prediction error of 15.89 μg/m³.