基于国产高分一号卫星数据的区域土壤盐渍化信息提取与建模

土壤盐渍化是干旱半干旱地区土地退化的主要原因之一,给当地生态系统和社会经济的可持续发展带来了严重的威胁,而对盐渍化空间分布信息的提取是治理盐渍化的基础。因此,选取生态脆弱区渭—库绿洲为研究区,利用2014年7月19日GF-1多光谱影像数据,提取光谱指数及波段信息,结合实际采样点的土壤表层电导率数据(0~10 cm),采用偏最小二乘回归模型(partial least squares regression,PLSR)对土壤盐渍化进行模拟,并对研究区盐渍化分布进行模拟和评估。结果表明:实测土壤表层电导率与光谱指数相关性较好;利用PLSR对渭—库绿洲土壤表层盐渍信息建模,对土壤盐渍化信息提取效果较好,精度较高;充分利用了GF-1影像包含的信息,提高了高分辨率遥感影像盐渍化信息提取的精度;非盐渍化和轻度盐渍化面积分别占总面积的42.88%和17.16%,绿洲中部偏东及东南区域,盐渍化现象稍弱,可成为今后绿洲扩张的重点方向;而中度盐渍化、重度盐渍化和盐土面积分别占总面积的29.51%、8.57%和1.88%,绿洲北部/西部及西南方向的重度盐渍化区域紧挨绿洲区域,已严重威胁了绿洲经济的健康发展,亟待治理。     

基于变量优选与机器学习的干旱区湿地土壤盐渍化数字制图

土壤盐渍化是导致土壤退化和生态系统恶化的主要原因之一，对干旱区的可持续发展构成主要威胁。为了尽可能精确地监测土壤盐渍化的空间变异性，该研究收集新疆艾比湖湿地78个典型样点，其中选取54个样本作为训练集，24个样本作为独立验证集。基于Sientinel-2 多光谱传感器（Multi-Spectral Instrument，MSI）、数字高程模型（Digital Elevation Model，DEM）数据提取3类指数（红边光谱指数、植被指数和地形指数），经过极端梯度提升（Extreme Gradient Boosting，XGBoost）算法筛选有效特征变量，构建了关于土壤电导率（Electrical Conductivity，EC）的随机森林（Random Forest，RF）、极限学习机（Extra Learning Machine，ELM）和偏最小二乘回归（Partial Least Squares Regression，PLSR）预测模型，并选择最优模型绘制了艾比湖湿地盐渍化分布图。结果表明：优选的红边光谱指数基本能够预测EC的空间变化；红边光谱指数与植被指数组合建模效果总体上优于其与地形指数的组合，3类指数组合的建模取得了较为理想的预测精度，其中RF模型表现最优（验证集R2=0.83，RMSE=4.81 dS/m，RPD=3.11）；在整个研究区内，中部和东部地区土壤盐渍化程度尤为严重。因此，XGBoost所筛选出的环境因子结合机器学习算法可以实现干旱区土壤盐渍化的监测。     

基于机器学习的干旱区土壤盐渍化定量估算

  目的  探讨对土壤盐渍化进行快速、准确监测技术与方法。  方法  利用353个地面表观电导率数据,以及从Worldview-2影像获取对应采样点的波段反射率值,结合两波段组合植被指数和三波段组合植被指数,筛选最佳二维、三维波段组合方式,引入人工神经网络、K近邻和支持向量回归来构建区域土壤盐渍化定量反演模型。  结果  ① WV-2影像的红边和近红外波段与ECa呈现显著相关（P < 0.01）。② 二维植被指数（RVI_(B5-B2)、NDVI_(B6-B2)、DVI_(B2-B6)）和三维植被指数（3DVI_(B2-B6-B6)、3DVI_(B3-B5-B6)、3DVI_(B5-B2-B1)、3DVI_(B2-B1-B6)、3DVI_(B2-B1-B6)、3DVI_(B6-B1-B2)、3DVI_(B5-B3-B7)）的波段组合计算提高了其对土壤盐渍化的敏感性。③ 基于不同维度数据的机器学习估算模型中,3DVI和KNN算法结合对土壤盐渍化估算效果最为突出,且模型精度为R2 = 0.773,RMSE = 1.659 dS m?1,RPD = 2.216。  结论  所构建的多维植被指数可应用于类似环境条件下盐渍土地监测和评价研究。     

基于国产高分一号卫星数据的区域土壤盐渍化信息提取与建模

土壤盐渍化是干旱半旱地区土地退化的主要原因之一，给当地生态系统和社会经济的可持续发展带来了严重的威胁，而对盐渍化空间分布信息的提取是治理盐渍化的基础。因此，选取生态脆弱区渭—库绿洲为研究区，利用2014年7月19日GF-1 多光谱影像数据，提取光谱指数及波段信息，结合实际采样点的土壤表层电导率数据（0～10 cm），采用偏最小二乘回归模型（Partial least squares regression，PLSR）对土壤盐渍化进行模拟，并对研究区盐渍化分布进行模拟和评估。结果表明：实测土壤表层电导率与光谱指数相关性较好；利用PLSR对渭—库绿洲土壤表层盐渍信息建模，对土壤盐渍化信息提取效果较好，精度较高；充分利用了GF-1影像包含的信息，提高了高分辨率遥感影像盐渍化信息提取的精度；非盐渍化和轻度盐渍化面积分别占总面积的42.88%和17.16%，绿洲中部偏东及东南区域，盐渍化现象稍弱，可成为今后绿洲扩张的重点方向；而中度盐渍化、重度盐渍化和盐土面积分别占总面积的29.51%、8.57%和1.88%，绿洲北部/西部及西南方向的重度盐渍化区域紧挨绿洲区域，已严重威胁到了绿洲经济的健康发展，亟待治理。     

基于反射光谱与Landsat 8 OLI多光谱数据的艾比湖湿地土壤盐分估算

土壤盐渍化是农业生产中最关键的生态问题之一,是一种降低土壤质量,严重影响作物产出的缓慢土壤退化过程。因此,土壤盐分的监测及评估对干旱区的土地管理与恢复至关重要。选取艾比湖湿地为研究区,基于2016年干湿两季(5月/9月)的Landsat8 OLI遥感影像,147个土壤实测样点的电导率(Electrical Conductivity, EC)数据及其对应的室内反射光谱数据,通过光谱重采样技术,建立该研究区土壤EC的偏最小二乘(partial least-squares regression, PLSR)定量估算模型,并尝试对干湿两季的土壤盐渍化状况进行对比分析。结果表明:(1)艾比湖湿地土壤盐渍化较为严重,湿季土壤EC(23.90 mS·cm~(-1))高于干季(11.62 mS·cm~(-1));(2)基于重采样处理后的光谱数据及13个光谱指数所建立的PLSR模型具有较好的精度(R2=0.91,RMSE=6.48mS·cm~(-1),RPD=2.45),说明该模型可以有效地对区域尺度的土壤EC进行定量估算。(3)从2016年5月至9月,艾比湖湿地的中度和重度盐渍土面积增加,轻度盐渍土和盐土面积减少。本研究将两种不同分辨率的数据进行联合建模,既提升了传统光学遥感影像模型的精度,又将高光谱数据扩展至像元尺度上,为土壤盐分信息的遥感提取提供了科学参考。     

基于黄瓜种植的设施菜地土壤硝酸盐型次生盐渍化的分级研究

在江苏省泰州市基于黄瓜种植(种植的黄瓜品种为东方明珠)的设施菜地中进行了硝酸盐型土壤次生盐渍化的分级研究。供试土壤经过不同浓度的硝酸盐处理对应形成不同程度的次生盐渍化。用土壤电导率(EC)标示次生盐渍化的等级,经测定,对照及处理的EC分别为:对照(CK)0.89 dS m-1,处理1(T1)2.78 dS m-1,处理2(T2)3.65 dS m-1,处理3(T3)4.66 dS m-1,处理4(T4)6.15 dS m-1。统计结果表明,表层土壤(0～10 cm)硝酸盐含量与土壤EC之间呈显著正相关关系。表明由硝酸盐引起的不同程度的土壤次生盐渍化即硝酸盐型次生盐渍化的分级可以通过土壤电导率来标示。实验表明,不同程度的硝酸盐型次生盐渍化显著影响黄瓜的株高和产量(p<0.05)。根据不同程度的次生盐渍化土壤对黄瓜株高和产量的影响,对长江三角洲地区基于黄瓜种植的设施菜地土壤硝酸盐型次生盐渍化进行了初步分级,即等级Ⅰ:EC<2.03 dS m-1,等级Ⅱ:2.03 dS m-16.15 dS m-1。     

基于SDI校正指数的滨海平原盐渍化生态风险评价

为了研究滨海平原区土壤盐渍化风险情况,本研究以黄骅市为研究区,将Landsat 8夏季遥感影像获得的SDI(盐渍化植被监测指数)校正指数作为生态终点,采用相关性分析法筛选土壤生产潜力和土壤健康方面指标,最终选取含盐量、电导率及有机质、K~+、Na~+、Mg~(2+)、SO_4~(2-)、Cl~-含量8个指标,借助SDI指数标准化校正的灰色关联度法构建盐渍化生态风险评价模型,对滨海平原区盐渍化生态风险进行定量评价及空间分析。结果表明:研究区内盐渍化生态风险值介于0.24～0.73,均值高达0.42,其中99.25%的区域处于中度及以上盐渍化生态风险水平,整体上土壤生态风险较高,呈现东部沿海高、西部平原低的趋势;盐渍化生态风险评价在空间上与土壤含盐量、土壤电导率较为相似,因此,土壤含盐量及电导率对盐渍化生态风险评价起主导作用。研究区内盐渍化生态风险及评价因子的空间分布可为生态环境、农业健康持续发展、地下水的限采禁采、土壤改良和盐渍化防治提供参考依据。     

博斯腾湖湖滨绿洲土壤电导率高光谱估算模型

以博斯腾湖湖滨绿洲为研究区，采用分数阶微分对光谱指数进行波段优化，筛选高光谱数据的特征波段，利用偏最小二乘回归(PLSR)和支持向量机(SVM)构建土壤电导率高光谱数据的估算模型。研究结果表明:(1)分数阶微分的高光谱数据与土壤电导率的相关性:随着分数阶微分阶数的增加，特征波段数呈现逐渐增加的趋势，2 阶是特征波段数量最多的阶数，特征波段数量为335(P=0.01)，相关系数绝对值最大为0.64。(2)分数阶微分优化光谱指数的高光谱数据:随着分数阶微分阶数的增加，光谱矩阵图表现为相关系数在正负值之间波动较大，0.8 阶在光谱指数DSI 的相关系数绝对值最大为0.75；平方根、对数、倒数的相关系数绝对值最大为0.64。(3)基于PLSR 和SVM 构建土壤电导率估算模型:基于0.8 阶微分和光谱指数DSI 筛选的特征波段建立的估算模型估算效果较好，其中SVM 构建的估算模型最优，模型精度为RSVMc2=0.89，RMSESVMc=0.03，RSVMv2=0.80，RMSESVMv=1.12。利用SVM 估算模型可以有效地对研究区土壤电导率进行定量估算。     

基于WorldView-2影像的土壤含盐量反演模型

针对WorldView-2影像高空间分辨率评价其定量反演土壤含盐量的能力,以盐渍化现象较为明显的新疆克里雅河流域为研究对象,基于WorldView-2影像和实测高光谱数据,利用偏最小二乘回归(partial least squares regression,PLSR)和BP人工神经网络(back propagation artificial neural networks,BP ANN)方法建立定量反演该流域土壤含盐量模型并做出研究区高空间分辨率土壤含盐量分布图。结果表明:1)利用实测高光谱数据和影像数据分别建立的2种模型中BP神经网络模型预测精度都高于PLSR模型,其中基于影像数据建立的6:8:1结构的3层BP神经网络模型决定系数R2、均方根误差RMSE、相对分析误差RPD分别为0.851、0.979、2.337,模型的稳定性和预测能力都优于PLSR模型(R2、RMSE、RPD分别为0.814、1.139、2.007)。2)利用WorldView-2影像提高了土壤含盐量制图的空间分辨率,归一化植被指数NDVI和比例植被指数RVI较有效降低了植被覆盖与土壤水分对预测精度的影响。该文建立的考虑植被覆盖与土壤水分定量反演土壤含盐量的模型不需要复杂的参数,一定程度上满足了干旱、半干旱地区的盐渍化监测需求,可以促进WorldView-2等高空间分辨率卫星在盐渍化监测中的进一步应用。     

基于多源遥感数据的艾比湖流域盐土SWAT模型参数修正

在SWAT(soil and water assessment tool)模型模拟地表分量过程中,常默认土壤剖面电导率(electrical conductivity,EC)值为0或0.1,将其应用于土壤盐渍化程度较高的流域时,不符合下垫面实际情况。为确保水文模拟逼近真实地表模拟过程,进一步提高模拟精度,该文利用GF-1号卫星16 m分辨率多光谱遥感影像结合分类回归树法反演艾比湖流域区域尺度0~100 cm土壤剖面电导率,模拟值与实测值均方根最大值误差为4.81 dS/m,相对误差最大值为15.17%。模拟值用于修正EC值,结果表明:EC值修正后的SWAT模型土壤水分模拟值,较修正前模拟值精度提高23.84个百分点。该方法在实现SWAT模型参数本地化的同时,有效提高了土壤水分模拟精度,可为土壤盐渍化区域水文模拟提供参考。     

气候因子和地表覆盖对沿海滩涂土壤盐分动态的影响

为探明气候因子对沿海滩涂表层土壤盐分季节性变化规律的影响,并探讨植被和秸秆覆盖对滩涂土壤脱盐效果及控盐的作用。2014年5月—2015年5月,在江苏沿海滩涂盐碱地(中重度盐分),设置4种处理进行田间试验,分别为对照(裸地,CK)、秸秆覆盖(覆盖量为15 t·hm-2,SM)、植被覆盖(PC)和植被+秸秆覆盖(覆盖量为7.5 t·hm-2,PC+1/2SM),监测了气候因子和表层土壤盐分的季节性动态变化。结果表明:1)在沿海滩涂裸地中,土壤盐分具有一定程度的季节性规律,表现为在10—12月具有明显的积盐效果,且在10月EC1︰5达到最大值为3.90 d S·m-1。2)相关分析表明:采样前7 d降雨累积量与土壤盐分变化有着极密切负相关关系;气候因子的多因子及互作逐步分析表明:降雨量增加可以促进土壤脱盐作用,大气温度升高可加剧土壤盐分表聚,降雨量和大气温度的互作效应增加会对土壤盐分累积产生正效应。3)地表覆盖(包括PC和SM)显著地改变了气候因子对土壤盐分动态变化的影响,累积降雨量和大气平均温度与土壤盐分无显著相关性,且大量秸秆覆盖对滩涂表层土壤脱盐具有更明显的效果。因此,在沿海气候向暖湿方向发展的趋势下,综合考虑脱盐及控盐作用,选择适量秸秆覆盖(如覆盖量15 t·hm-2)或适量秸秆覆盖结合植被种植覆盖,同时充分利用沿海地区降雨量集中的特点,可能是未来滩涂盐碱盐渍土快速脱盐和土壤改良的重要措施。     

Application of VNIR and machine learning technologies to predict heavy metals in soil and pollution indices in mining areas

Purpose

Soil pollution indices are an effective tool in the computation of metal contamination in soil. They monitor soil quality and ensure future sustainability in agricultural systems. However, calculating a soil pollution index requires laboratory measurements of multiple soil heavy metals, which increases the cost and complexity of evaluating soil heavy metal pollution. Visible and near-infrared spectroscopy (VNIR, 350–2500 nm) has been widely used in predicting soil properties due to its advantages of a rapid analysis, non-destructiveness, and a low cost.

Methods

In this study, we evaluated the ability of the VNIR to predict soil heavy metals (As, Cu, Pb, Zn, and Cr) and two commonly used soil pollution indices (Nemerow integrated pollution index, NIPI; potential ecological risk index, RI). Three nonlinear machine learning techniques, including cubist regression tree (Cubist), Gaussian process regression (GPR), and support vector machine (SVM), were compared with partial least squares regression (PLSR) to determine the most suitable model for predicting the soil heavy metals and pollution indices.

Results

The results showed that the nonlinear machine learning models performed significantly better than the PLSR model in most cases. Overall, the SVM model showed a higher prediction accuracy and a stronger generalization for Zn (R²_V?=?0.95, RMSE_V?=?6.75 mg kg^?1), Cu (R²_V?=?0.95, RMSE_V?=?8.04 mg kg^?1), Cr (R²_V?=?0.90, RMSE_V?=?6.57 mg kg^?1), Pb (R²_V?=?0.86, RMSE_V?=?4.14 mg kg^?1), NIPI (R²_V?=?0.93, RMSE_V?=?0.31), and RI (R²_V?=?0.90, RMSE_V 3.88). In addition, the research results proved that the high prediction accuracy of the three heavy metal elements Cu, Pb, and Zn and their significant positive correlations with the soil pollution indices were the reason for the accurate prediction of NIPI and RI.

Conclusion

Using VNIR to obtain soil pollution indices quickly and accurately is of great significance for the comprehensive evaluation, prevention, and control of soil heavy metal pollution.

     

N2O and CH4 fluxes from a volcanic grassland soil in Nasu,Japan: Comparison between manure plus fertilizer plot and fertilizer-only plot

Abstract

We examined the effects of manure + fertilizer application and fertilizer-only application on nitrous oxide (N₂O) and methane (CH₄) fluxes from a volcanic grassland soil in Nasu, Japan. In the manure + fertilizer applied plot (manure plot), the sum of N mineralized from the manure and N applied as ammonium sulfate was adjusted to 210 kg N ha^?1 year^?1. In the fertilizer-only applied plot (fertilizer plot), 210 kg N ha^?1 year^?1 was applied as ammonium sulfate. The manure was applied to the manure plot in November and the fertilizer was applied to both plots in March, May, July and September. From November 2004 to November 2006, we regularly measured N₂O and CH₄ fluxes using closed chambers. Annual N₂O emissions from the manure and fertilizer plots ranged from 7.0 to 11.0 and from 4.7 to 9.1 kg N ha^?1, respectively. Annual N₂O emissions were greater from the manure plot than from the fertilizer plot (P < 0.05). This difference could be attributed to N₂O emissions following manure application. N₂O fluxes were correlated with soil temperature (R = 0.70, P < 0.001), NH⁺ ₄ concentration in the soil (R = 0.67, P < 0.001), soil pH (R = –0.46, P < 0.001) and NO^? ₃ concentration in the soil (R = 0.40, P < 0.001). When included in the multiple regression model (R = 0.72, P < 0.001), however, the following variables were significant: NH⁺ ₄ concentration in the soil (β = 0.52, P < 0.001), soil temperature (β = 0.36, P < 0.001) and soil moisture content (β = 0.26, P < 0.001). Annual CH₄ emissions from the manure and fertilizer plots ranged from –0.74 to –0.16 and from –0.84 to –0.52 kg C ha^?1, respectively. No significant difference was observed in annual CH₄ emissions between the plots. During the third grass-growing period from July to September, however, cumulative CH₄ emissions were greater from the manure plot than from the fertilizer plot (P < 0.05). CH₄ fluxes were correlated with NH⁺ ₄ concentration in the soil (R = 0.21, P < 0.05) and soil moisture content (R = 0.20, P < 0.05). When included in the multiple regression model (R = 0.29, P < 0.05), both NH⁺ ₄ concentration in the soil (β = 0.20, P < 0.05) and soil moisture content (β = 0.20, P < 0.05) were significant.     

Evaluation of the NH4HCO3‐DTPA soil test for assessing boron availability to wheat

Abstract

The NH₄HCO₃‐DTPA (AB‐DTPA), 1 MNH₄HCO₃, 0.005 M DTPA, pH=7.6, was proposed as a multi‐element extractant, for evaluating macro and micronutrients availability to plants. AB‐DTPA was also evaluated as a soil test, for assessing boron availability and toxicity to alfalfa. In a pot experiment, ten soils of Northern Greece were used to assess AB‐DTPA as an extractant of available boron to wheat (Triticum aestivum L., cv. Yecora), in comparison with hot water and saturation extract. Boron (B) was added as borax (Na₂B₄O₇*10H₂O) to the ten soils, at rates equal to 0, 3, and 5 mg B kg^‐1. Wheat was grown in pots containing the boron amended soils to the stage of tillering, and dry aboveground biomass, B concentration and B uptake by wheat were determined. AB‐DTPA extractable B was significantly greater than saturation extract and similar to hot water at each B application rate, and was correlated significantly with hot water (r=0.84), or with saturation extract (r=0.48). Extractable boron by all extractants, boron concentration in wheat and boron uptake were significantly affected by the soil x B application rate interaction. In assessing B availability to wheat using AB‐DTPA as a soil test, CEC should be included in the regression equation for B concentration, or pH for B uptake. However, the corresponding adjusted coefficients of determination for B concentration (adjusted R²=0.46) and B uptake (adjusted R²=0.48) were similar or lower to those of hot water (adjusted R²=0.45 and 0.60, respectively) and the saturation extract (adjusted R²=0.70 and 0.49, respectively), when the latter two soil tests were used in the regression equations without the inclusion of any soil property.     

Prediction of macronutrients in plant leaves using chemometric analysis and wavelength selection

Purpose

Fast and real-time prediction of leaf nutrient concentrations can facilitate decision-making for fertilisation regimes on farms and address issues raised with over-fertilisation. Cacao (Theobroma cacao L.) is an important cash crop and requires nutrient supply to maintain yield. This project aimed to use chemometric analysis and wavelength selection to improve the accuracy of foliar nutrient prediction.

Materials and methods

We used a visible-near infrared (400–1000 nm) hyperspectral imaging (HSI) system to predict foliar calcium (Ca), potassium (K), phosphorus (P) and nitrogen (N) of cacao trees. Images were captured from 95 leaf samples. Partial least square regression (PLSR) models were developed to predict leaf nutrient concentrations and wavelength selection was undertaken.

Results and discussion

Using all wavelengths, Ca (R²_CV?=?0.76, RMSE_CV?=?0.28), K (R²_CV?=?0.35, RMSE_CV?=?0.46), P (R²_CV?=?0.75, RMSE_CV?=?0.019) and N (R²_CV?=?0.73, RMSE_CV?=?0.17) were predicted. Wavelength selection increased the prediction accuracy of Ca (R²_CV?=?0.79, RMSE_CV?=?0.27) and N (R²_CV?=?0.74, RMSE_CV?=?0.16), while did not affect prediction accuracy of foliar K (R²_CV?=?0.35, RMSE_CV?=?0.46) and P (R²_CV?=?0.75, RMSE_CV?=?0.019).

Conclusions

Visible-near infrared HSI has a good potential to predict Ca, P and N concentrations in cacao leaf samples, but K concentrations could not be predicted reliably. Wavelength selection increased the prediction accuracy of foliar Ca and N leading to a reduced number of wavelengths involved in developed models.

     

Estimating Mo,Cu, Ni,Cd Contents in the Crop Leaves Growing on Small Land Plots Using Satellite Data

ABSTRACT

The main goal of this research was to estimate heavy metals (HMs) (molybdenum (Mo), copper (Cu), nickel (Ni), cadmium (Cd)) contents in crop leaves through multispectral satellite imagery. During the acquisition of a SPOT 7 satellite image (28 July 2017) in situ sampling (38 samples) was done from the leaves of potatoes and beans growing close to the mining town of Kajaran (Armenia). To estimate HMs contents, multivariate regression (multiple linear regression (MLR), partial least squares regression (PLSR)), and artificial neural network (ANN) were used. As input data for the models raw, atmospherically corrected (Dark Object Subtraction (DOS)) and hyperspherical direction cosines (HSDC) normalized values of SPOT 7 spectral data in combination with one or combined log10, multiplicative scatter correction (MSC), standard normal variate transform (SNV) preprocessing methods were utilized. The best results were obtained for Cu using MLR (R² _cal. = 0.79, R² _CV = 0.70, RMSE_cal. = 11.27, RMSE_CV = 13.47) and ANN (R² _Train ≈ 0.80, R² _Test ≈ 0.72, RMSE_Train ≈ 11, RMSE_Test ≈ 13) models in case of bean leaves. The results are quite optimistic, however, further research with the use of high spatial/spectral resolution satellite images is needed to improve the accuracy of models.     

Neural network models to predict soil water retention

Soil water retention curves are needed to describe the availability of soil water to plants and to model movement of water through unsaturated soils. Measuring these characteristics is time-consuming, labour-intensive and therefore expensive. This study was conducted to develop and evaluate functions based on neural networks to predict soil water retention characteristics. Dutch and Scottish data sets were available; they contained data on 178 and 165 soil horizons, respectively. A series of three neural networks (A, B and C) was developed. Neural network A had the following input variables: topsoil, bulk density, organic matter, clay, silt and sand content. In addition neural network B had matric potential as input, and network C included soil structural data expressed as the upper and lower boundary of the ped-size class. Neural network A had three output variables: the volumetric water content at matric potentials of 0, –100 and –15 000 hPa. Both models B and C had volumetric water content, at the matric potential given as input, as output variable. The networks were tested against independent data that were extracted from the original sets of soil profiles. Accuracy of the predictions was quantified by the root of the mean squared difference (RMSE) between the measured and the predicted water contents, and the coefficient of determination (R²). For network A the RMSE varied for the three estimated water contents from 0.0264 to 0.0476 cm³ cm^–3, and R² varied from 0.80 to 0.93 for the individual model outputs. Networks B and C had an RMSE of 0.0435 and 0.0426 cm³ cm^–3, respectively. For both networks, R² was 0.89. The neural networks performed somewhat better than previous regression functions, but the improvements were not significant.     

Soil phosphorus availability to millet roots

Abstract

Soil solution P level is believed to be important in determining P uptake rates from soil. The objective of this research was to investigate the relation between initial P concentration in the soil solution and P flux into the root. Millet (Panicum milaceum) was grown on five soils each of which was adjusted to six C_li levels by addition of P. Millet was also grown in solution culture and P influx vs. P concentration in solution measured. There was a curvilinear relation between P influx and relative yield of the C_li levels on each soil (R²=0.74). A P influx of at least 16 pmoles cm^‐1 sec^‐1 was needed to obtain 90% of maximum yield. However, yield response was not correlated with C_li, indicating C_li was not a suitable indicator of P availability on these soils. Influx of P on soils with C_li less than 6 μM was greater than occurred at similar P concentrations in solution culture indicating P influx was increased by the effect of the root on the soil.     

设施番茄单株成果数模型研究

单株成果数是番茄单株产量的构成因子,为了定量分析不同品种设施番茄单株成果数与环境条件之间的关系,以"美国摩尔一号"(B1,偏早熟)、"超世纪番茄大王"(B2,偏晚熟)和"美国903"(B3,中熟)为材料,于2009年、2010年和2011年开展了品种和施肥、品种和水分田间试验。通过分析不同品种、水分和施肥水平番茄坐果数、果实脱落数、开花数及现蕾数与环境因子的关系,建立了设施番茄单株现蕾数、单株花脱落数、单株果脱落数和单株成果数模型。经独立试验资料检验,设施番茄品种B1、B2和B3平均单株累积现蕾数实测值与模拟值的根均方差(RMSE)、平均绝对误差(Xde)和决定系数(R2)分别为2.452个(n=24)、1.851个和0.976,1.820个(n=24)、1.422个和0.948,1.849个(n=24)、1.464个和0.949。单株花脱落数实测值与模拟值的RMSE、Xde和R2分别为0.712个(n=16)、0.662个和0.786,0.730个(n=17)、0.662个和0.965,1.229个(n=16)、1.091个和0.952。单株果实累积脱落数实测值与模拟值的RMSE、Xde和R2分别为0.391个(n=15)、0.342个和0.849,0.439个(n=15)、0.346个和0.966,0.318个(n=15)、0.288个和0.961。单株成果数模拟值与实测值的RMSE、Xde和R2分别为0.839个(n=27)、0.712个和0.934,实测值与模拟值的吻合程度较好,说明模型可较好地模拟不同品种、水分和施肥水平设施番茄单株成果数。     

Soil texture affects the conversion factor of electrical conductivity from 1:5 soil-water to saturated paste extracts

Electrical conductivity (EC) of soil-water extracts is commonly used to assess soil salinity. However, its conversion to the EC of saturated soil paste extracts (EC_e), the standard measure of soil salinity, is currently required for practical applications. Although many regression models can be used to obtain EC_e from the EC of soil-water extracts, the application of a site-specific model to different sites is not straightforward due to confounding soil factors such as soil texture. This study was conducted to develop a universal regression model to estimate a conversion factor (CF) for predicting EC_e from EC of soil-water extracts at a 1:5 ratio (EC_1:5), by employing a site-specific soil texture (i.e., sand content). A regression model, CF=8.910 5e^{0.010 6sand}/1.298 4 (r²=0.97, P < 0.001), was developed based on the results of coastal saline soil surveys (n=173) and laboratory experiments using artificial saline soils with different textures (n=6, sand content=10%-65%) and salinity levels (n=7, salinity=1-24 dS m^-1). Model performance was validated using an independent dataset and demonstrated that EC_e prediction using the developed model is more suitable for highly saline soils than for low saline soils. The feasibility of the regression model should be tested at other sites. Other soil factors affecting EC conversion factor also need to be explored to revise and improve the model through further studies.