Comparing multivariate regression and artificial neural network to predict barley production from soil characteristics in northern Iran

In this study artificial neural network (ANN) models were designed to predict the biomass and grain yield of barley from soil properties; and the performance of ANN models was compared with earlier tested statistical models based on multivariate regression. Barley yield data and surface soil samples (0–30 cm depth) were collected from 1 m² plots at 112 selected points in the arid region of northern Iran. ANN yield models gave higher coefficient of determination and lower root mean square error compared to the multivariate regression, indicating that ANN is a more powerful tool than multivariate regression. Sensitivity analysis showed that soil electrical conductivity, sodium absorption ratio, pH, total nitrogen, available phosphorus, and organic matter consistently influenced barley biomass and grain yield. A comparison of the two methods to identify the most important factors indicated that while in the ANN analysis, soil organic matter (SOM) was included among the most important factors; SOM was excluded from the most important factors in the multivariate analysis. This significant discrepancy between the two methods was apparently a consequence ofthe non-linear relationships of SOM with other soil properties. Overall, our results indicated that the ANN models could explain 93 and 89% of the total variability in barley biomass and grain yield, respectively. The performance of the ANN models as compared to multivariate regression has better chance for predicting yield, especially when complex non-linear relationships exist among thefactors. We suggest that for further potential improvement in predicting thebarley yield, factors other than the soil properties considered such as soil micronutrient status and soil and crop management practices followed during the growing season, need to be included in the models.     

Simulating the Production of Rice Genotypes by Flood Management and End-Season Water Stress Conditions Using AquaCrop Model

ABSTRACT

The present study was conducted to assess the ability of AquaCrop model in predicting of grain and biological yield of rice genotypes in water management. A two-year field experiment was conducted at the experimental farm of the Iranian Rice Research Institute in Rasht, Iran from 2016 to 2017. The experiment was established in a split-plot design with two irrigation management (continuous submergence and end season water stress) as the main plot, fourth rice genotypes as the sub-plot and three replications. The goodness-of-fit between observed and simulated grain yield and final biomass was assessed by means of the coefficient of determination (R ²), the absolute and normalized root mean square errors (RMSE). The RMSE_n of predicting grain yield at calibration and evaluation stages was in the range of 6–12% and 6–8% for biological yield. The results indicated that AquaCrop model is suitable to predict grain yield and biological yield of rice genotypes in northern Iran. AquaCrop model can be used to determine optimization strategies to improve the water consumption of rice genotypes.     

水稻叶片氮素及籽粒蛋白质含量的高光谱估测模型

研究水稻叶片氮素和籽粒蛋白质含量的高光谱快速、无损监测方法,对于水稻营养诊断、籽粒品质监测及氮肥高效利用具有重要意义。本文通过水稻盆栽试验,测定水稻叶片氮素、籽粒蛋白质含量和冠层光谱,采用不同的光谱建模方法来提高氮素、籽粒蛋白质含量的估测精度。先用主成分分析(PCA)方法进行特征波段的提取,再用多元线性回归(MLR)、人工神经网络(ANN)和偏最小二乘回归(PLSR)进行建模。结果表明,水稻叶片氮素和籽粒蛋白质含量与特征光谱存在很好的模型关系,3种模型预测的决定系数(R2_p)均在0.847以上,并以PLSR模型的预测效果为最好,可以实现水稻氮素营养和籽粒品质的高光谱估测。     

Artificial neural network for estimating monthly reference evapotransiration under arid and semi arid environments

The objective of this study is to investigate the potential of artificial neural networks (ANNs) for estimating reference monthly evapotranspiration under arid and semi-arid environments. A simple leave one out data analysis was carried out; one neural network solution on six inputs and another six network solutions on five inputs for each monitoring station were done. Comparison of the results showed that the accuracy of ANNs is decreased when relative humidity, wind speed and solar or extraterrestrial radiation are excluded as input variables. The results also showed that monthly evapotranspiration could be computed with relatively good accuracy compared with local calibrated Hargreaves equation based on air temperature using trained ANNs at another location. We conclude, based on our overall results, that temperature-based method ANNs can be used with relatively good accuracy for water resource management, irrigation scheduling and management, and environmental assessment when data are not enough using trained ANNs from another location.     

Evaluation of Ceres-Rice,Aquacrop and Oryza2000 Models in Simulation of Rice Yield Response to Different Irrigation and Nitrogen Management Strategies

This study evaluated CERES-Rice, AquaCrop, and ORYZA2000 models performance in simulation of biological and grain yield of rice in response to different irrigation intervals and nitrogen levels. These models were calibrated and validated by using three years (2005 to 2007) field experiments. Three levels of irrigation interval included pond treatment, five days interval, and eight days interval, and consisted of four levels of nitrogen. The study results showed that there were significant differences among study crop models in simulation of grain and biological yield in response to different irrigation intervals. As results showed, study models performed more accurate in estimation of rice yield under irrigation intervals than nitrogen levels. All models illustrated high performance in estimation of rice yield under different irrigation intervals. CERES-Rice and AquaCrop models showed highest accuracy in simulation of grain and biological yield of rice under different levels of nitrogen, respectively. In addition, CERES-Rice model indicated highest performance in simulation of grain yield (rRMSE = 16). However, AquaCrop model estimated biological yield more accurate compared to other models (rRMSE = 15). ORYZA2000 showed less accurate in simulating grain (rRMSE = 23) and biological (rRMSE = 21) yield of rice in comparison with other models.     

Prediction of Cooked Rice Texture Using an Extrusion Test in Combination with Partial Least Squares Regression and Artificial Neural Networks

Spectral stress strain analysis was used in combination with partial least squares (PLS) regression and artificial neural networks (ANN) to predict nine sensory texture attributes of cooked rice. The models calculated with ANN were significantly more accurate in predicting most of the sensory texture characteristics evaluated than the PLS models. Furthermore, ANN models were more robust and discriminative than PLS models.     

水氮互作对冷凉区域稻田氮素特征的影响

[目的]研究寒地稻田不同水肥管理模式下的土壤供氮特征,为筛选环境友好型寒地稻作灌溉施肥模式提供支撑。[方法]在大田试验条件下,设置间歇灌溉、淹灌2种水分管理模式及4个供氮水平(0,75,105,135kg/hm~2),以龙庆稻2号为材料,研究水肥互作模式对水稻产量、土壤供氮特征及氮素利用率的影响。[结果]灌溉模式和供氮水平对水稻产量、地上部氮素积累量、水稻氮素利用率均有显著(p0.05)或极显著影响(p0.01)。间歇灌溉模式下,增加氮肥施用量有利于提高单位面积水稻有效穗数、籽粒产量、生物产量、籽粒氮素累积量,均以施氮105kg/hm~2处理最高;水肥互作对氮素利用率影响明显,水稻的氮肥利用率在21.4%~59.1%;氮肥生理利用率、氮肥农学效率及氮肥偏生产力均随着施氮量的提高而降低,施氮量75kg/hm~2处理的氮素吸收利用各项指标均高于其他处理。相关分析表明,水肥因素是影响氮素积累及氮素吸收利用效率的重要因子。[结论]综合考虑水稻产量及氮素利用率的矛盾,间歇灌溉配合适宜减氮模式应予以高度重视。     

Predicting Cationic Exchange Capacity in Calcareous Soils of East-Azerbaijan Province,Northwest Iran

The aim of this research is to study the efficiency of pedotransfer functions (PTFs) and artificial neural networks (ANNs) for cationic exchange capacity (CEC) prediction using readily available soil properties. Here, 417 soil samples were collected from the calcareous soils located in East-Azerbaijan province, northwest Iran and readily available soil properties, such as particle size distribution (PSD), organic matter (OM) and calcium carbonate equivalent (CCE), were measured. The entire 417 soil samples were divided into two groups, a training data set (83 soil samples) and test data set (334 soil samples). The performances of several published and derived PTFs and developed neural network algorithms using multilayer perceptron were compared, using a test data set. Results showed that, based on statistics of RMSE and R², PTFs and ANNs had a similar performance, and there was no significant difference in the accuracy of the model results. The result of the sensitivity analysis showed that the ANN models were very sensitive to the clay variable (due to the high variability of the clay). Finally, the models tested in this study could account for 85% of the variations in cationic exchange capacity (CEC) of soils in the studied area.

Abbreviations: ANN: arti?cial neural networks; MLP: multilayer perceptron; MLR: multiple linear regression; PTFs: Pedotransfer Functions; RBF: Radial Basis Function; MAE: mean absolute error; MSE: mean square error; CEC: cationic exchange capacity     

干湿交替灌溉与施氮量对水稻叶片光合性状的耦合效应

【目的】探讨干湿交替灌溉与施氮量耦合对水稻光合性状及其效应的影响,从光合源及光合质方面阐明不同水氮组合处理在光合性状上的差异。【方法】以新稻20为材料进行土培试验,设置浅水层灌溉 (0 kPa)、轻度干湿交替灌溉 (–20 kPa) 和重度干湿交替灌溉 (–40 kPa) 3种灌溉方式及不施氮 (N0)、中氮 (MN, 240 kg/hm2) 和高氮 (HN, 360 kg/hm2) 3种氮水平,研究不同水氮耦合处理对水稻产量、叶片叶绿素含量、叶面积指数、叶片氮含量、净光合速率、光合氮素利用率、PSⅡ的潜在活性和最大光化学效率的影响。【结果】灌溉方式与施氮量存在显著的互作效应,轻度干湿交替灌溉增加了主要生育期叶片叶绿素含量、氮含量、净光合速率、光合氮素利用率、PSⅡ的潜在活性和最大光化学效率,提高抽穗后群体叶面积指数,且与MN耦合后产量最高,为本试验最佳的水氮耦合运筹模式;重度干湿交替灌溉则显著降低主要生育期叶片叶绿素含量、叶片氮含量、叶面积指数、净光合速率、光合氮素利用率、PSⅡ的潜在活性和最大光化学效率;在同一灌溉方式下,中氮处理提高叶片净光合速率、光合氮素利用率、PSⅡ的潜在活性和最大光化学效率,有利于穗后叶片叶绿素含量及叶面积指数提高,重施氮肥反而降低叶片光合及荧光效率。水稻叶片叶面积指数、光合速率、光合氮素利用率、PSⅡ潜在活力及最大光化学效率与产量均呈显著或极显著的正相关关系。水稻主要生育期光合性状指标的供氮效应均为正效应,轻度干湿交替灌溉下主要生育期叶片叶绿素含量、氮含量、净光合速率、光合氮素利用率、PSⅡ的潜在活性和最大光化学效率的供水效应及耦合效应均为正效应,而重度干湿交替灌溉的控水及耦合效应则为负效应。【结论】轻度干湿交替灌溉耦合中氮处理水稻叶片净光合速率、光合氮素利用率、PSⅡ的潜在活性和最大光化学效率提高,有利于穗后叶片叶绿素含量及叶面积指数形成,表明通过适宜的肥水调控发挥水氮耦合效应,可以创造良好的光合性状,提高水稻光合生产能力,从而促进水稻高产。     

Modeling of maximum dry density and optimum moisture content of stabilized soil using artificial neural networks

This study considers the use of artificial neural networks (ANNs) to predict the maximum dry density (MDD) and optimum moisture content (OMC) of soil‐stabilizer mix. Multilayer perceptron (MLP), one of the most widely used ANN architectures in the literature, is utilized to construct comprehensive and accurate models relating the MDD and OMC of stabilized soil to the properties of natural soil such as particle‐size distribution, plasticity, linear shrinkage, and the type and quantity of stabilizing additives. Five ANN models are constructed using different combinations of the input parameters. Two separate sets of ANN prediction models, one for MDD and the other for OMC, and also a combined ANN model for multiple outputs are developed using the potentially influential input parameters. Relative‐importance values of various inputs of the models are calculated to determine the significance of each of the predictor variables to MDD and OMC. Inferring the most relevant input parameters based on Garson's algorithm, modified ANN models are separately developed for MDD and OMC. The modified ANN models are utilized to introduce explicit formulations of MDD and OMC. A parametric study is also conducted to evaluate the sensitivity of MDD and OMC due to the variation of the most influencing input parameters. A comprehensive set of data including a wide range of soil types obtained from the previously published stabilization test results is used for training and testing the prediction models. The performance of ANN‐based models is subsequently analyzed and compared in detail. The results demonstrate that the accuracy of the proposed models is satisfactory as compared to the experimental results.     

控释氮肥的氮素释放特征及其对水稻的增产效应

采用浸水释放试验和水稻田间试验方法研究了疏水性材料有机高分子聚合物包膜控释氮肥的释放特征,并以普通氮肥作对照研究了包膜控释氮肥对水稻的增产效应。研究结果表明,包膜控释氮肥养分释放达到了控释效果;与普通氮肥相比,施用控释氮肥有效提高了稻田土壤中碱解氮和铵态氮含量,促进了水稻植株和根系对氮素的吸收利用;两种处理的水稻总生物量差异不显著,但是在籽粒产量上,控释氮肥各处理两年的试验结果均比普通氮肥增产10%~40%,尤其是控释氮肥施用量为普通氮肥1/3时仍有增产效果,2003年高达15·1%;在当地稻田条件下,2002年控释氮肥氮素利用率最高达到了51·2%。采用疏水性包膜材料有机高分子聚合物包膜的控释氮肥比普通氮肥具有控制释放、促进水稻吸收氮素养分、提高籽粒产量和肥料的当季利用率与降低施肥量的优越性。     

氮素和水分处理对稻米植酸含量和蛋白组分的影响

以直立密穗型粳稻品种秀水110和弯曲散穗型粳稻品种春江15为材料,对不同氮素水平处理和水分管理方式下稻米植酸含量和蛋白组分的影响及其互作效应进行了研究。结果表明,旱作栽培处理会导致水稻籽粒中植酸含量上升,而施氮处理对籽粒植酸含量的影响效应与水稻的水分管理方式有关,在常规水作条件下,高氮处理（N3）的籽粒植酸含量有所提高,但在旱作栽培方式下,中氮（N2）和低氮处理（N1,不施N）的籽粒植酸含量却略高于高氮处理（N3）,氮素水平与水分管理方式间的互作效应明显; 水稻籽粒植酸含量与粗蛋白总量、4种蛋白组分（清蛋白、球蛋白、醇溶蛋白和谷蛋白）及有关产量性状指标（有效穗数、每穗粒数、千粒重和结实率）在不同水肥处理间的相关性不显著,但氮肥用量过高不仅会导致水稻的产量水平下降,而且也不利于稻米营养品质的改良; 水肥处理对水稻籽粒植酸含量、蛋白总量和4种蛋白组分穗内粒位分布也存在一定影响,着生在稻穗下部的弱势粒,其稻米植酸含量高于稻穗上部或中部的强势粒,因此改善弱势粒灌浆的水肥管理措施将有利于稻米植酸含量的降低。     

Simulating the Impact of Nitrogen Management on Rice Yield and Nitrogen Uptake in Irrigated Lowland by ORYZA2000 Model

This paper considers the implications of ORYZA2000 model in simulating physiological traits of rice at different nitrogen concentrations. The experiment was conducted over the course of the growing season in 2012 and 2013 in Rice Research Institute, Deputy of Mazandaran, Iran. The variety used was Shiroudi as a high yielding variety. The considered factors were the amount of nitrogen at four levels (40, 80, 120, 160 kg N ha^?1 and control) and nitrogen splitting in four levels. We compared simulated and measured grain yield, biomass, grain nitrogen, total plant nitrogen uptake, and leaf area index (LAI) by Student’s t-test of means and by absolute and normalized root mean square errors (RMSE). Results showed that grain yield was simulated with an RMSE of 411–423 kg ha^?1 and a normalized RMSE of 6%. RMSE was 671–910 kg ha^?1 for biomass on harvesting date. RMSE were 7–11 for grain nitrogen, and 10–13 for total plant nitrogen uptake. LAI was simulated with a normalized RMSE of 17–23%. Generally the model simulated LAI, an exceeded measured value for different nitrogen treatments. The most obvious finding that emerged from this study was that ORYZA2000 model can be applied as a supportive research tool for selecting the most appropriate strategies for rice yield improvement at various nitrogen fertilization concentrations.     

氮肥品种和用量对水稻产量和镉吸收的影响研究

采用盆栽试验,研究了Cd污染土壤上,不同氮肥品种和用量对水稻产量和Cd吸收的影响。结果表明,与磷钾配施的4个氮肥处理中,施用尿素处理水稻产量最高,其次为施(NH4)2SO4和NH4Cl处理,施NH4NO3处理水稻产量最低。与施(NH4)2SO4、NH4NO3和尿素处理相比,施NH4Cl处理可显著增加水稻对Cd的吸收,并促进Cd由秸秆向籽粒的转移;而其他3种氮肥对水稻秸秆和籽粒中Cd含量的影响效应相当。适量尿素[0.2g(N)·kg-1]处理能显著降低水稻籽粒Cd含量,而不施尿素和高量尿素处理都显著提高了水稻籽粒中的Cd含量。研究表明,在Cd污染的水稻土上,采用抗Cd污染的水稻品种和优化肥、水管理措施,可使稻米中Cd含量低于国家无公害大米的限量指标。     

稻-麦轮作条件下两种施肥模式对作物产量 和农田氮磷径流流失的比较研究

稻-麦轮作是太湖流域典型的集约化粮食作物种植体系,化肥用量大,氮磷流失控制广为关注。本文采用大区田间对比试验研究了习惯施肥（FP）和优化控制施肥（CM）两种施肥模式对作物产量及氮磷肥料偏生产力的影响,同时探讨了两种施肥模式下农田径流水中各形态氮、磷的特征和径流氮、磷损失的差异。结果表明优化控制施肥水稻和小麦地上部总生物量、籽粒产量、植株地上各部位养分（氮磷钾）含量及积累量与习惯施肥差异不显著（P>0.05）;优化控制施肥水稻和小麦的氮肥偏生产力显著大于习惯施肥（P<0.05）,磷肥偏生产力也相似。稻季和麦季优化控制施肥径流水中各形态氮、磷浓度小于习惯施肥,甚至达到显著水平（P<0.05）;稻季、麦季和完整轮作期优化控制施肥总氮、总磷的累积流失量显著小于习惯施肥（P<0.05）。优化控制施肥模式不仅能保持水稻和小麦的籽粒产量,而且能显著减少稻-麦轮作体系的氮磷流失,可以在实际农业生产中加以推广和利用。     

太湖地区黄泥土水稻适宜施氮量研究——长期定位试验

在长期定位试验的基础上,研究了黄泥土(太湖地区主要土壤类型)不同肥料配施土壤N供应、植株N吸收及产量之间的关系。结果表明,土壤N供应和植株N吸收、植株N吸收和产量之间呈显著正相关关系。在本试验条件下,稻季N施入161.00~241.00kg/hm2,产量达7285~8172kg/hm2,与该地区大面积产量基本一致;长期不施肥,能维持一定的产量;长期仅施入有机肥,不能满足水稻对N的需要,产量较低;长期不施入P、K,对产量影响不大。     

减施氮肥和增氧灌溉对水稻氮代谢关键酶活性及氮素利用的影响

该研究旨在分析减施氮肥和增氧灌溉对水稻氮代谢关键酶活性及氮素利用的影响。2020年以中旱221（旱稻）、中浙优8号（水稻）和IR45765-3B（深水稻）共3个品种为材料,设常规淹水灌溉（Conventional Flood Irrigation, WL）、微纳米气泡水增氧灌溉（Micro-nano Bubble Water Oxygenation Irrigation, MBWI）2个灌溉模式和常规施氮（195.0 kg/hm2）、减施氮肥（157.5 kg/hm2）2个氮水平,研究了减施氮肥和增氧灌溉对水稻关键生育时期氮代谢相关酶活性、植株含氮量、氮素积累量以及产量和氮肥利用率的影响。结果表明,与常规施氮处理相比,减施氮肥降低了氮代谢酶活性,而增氧灌溉有助于提高硝酸还原酶、谷氨酰胺合成酶、谷氨酸合酶、谷丙转氨酶活性。增氧灌溉和氮肥一定程度上有助于水稻氮素积累,增氧灌溉下减施氮肥处理比淹水灌溉常规施氮量的当季氮肥利用率分别提高15.6%、36.2%、21.5%（P<0.05）。增氧灌溉和增施氮肥显著增加水稻产量,中旱221增氧灌溉下减施氮肥比淹水灌溉常规施氮量处理增产3.5%（P<0.05）,而中浙优8号和IR45765-3B增氧灌溉下减施氮肥与淹水灌溉常规施氮量差异不显著（P>0.05）。相关分析表明,氮代谢酶活性与同时期叶片含氮量及氮素积累量大多呈显著或极显著的正相关。可见,增氧灌溉可以显著提高水稻氮素代谢相关酶活性,从而显著提高水稻氮素积累量、产量和当季氮肥利用效率,水稻氮肥减施条件下采用增氧灌溉有助于水稻维持较高氮肥吸收和利用效率,而谷氨酰胺合成酶活性可以用于预测水稻各时期氮素积累量。研究结果可为水稻氮肥减施和提高水稻氮肥利用效率提供理论和技术支持。     

不同耕作方式下水分管理对水稻氮素吸收利用的影响?

2008年早季和晚季,以桂旱1号和金优253为试验材料,盆栽条件下利用15N示踪技术研究了湿润灌溉、交替灌溉、水层灌溉3种水分管理模式下和免耕、常耕2种土壤耕作方式下水稻氮素与吸收利用的差异。结果表明：（1）湿润灌溉栽培降低了2种耕作方式水稻的氮肥吸收量,其中对基肥和分蘖肥吸收量的降低达显著水平;（2）湿润灌溉栽培时施入稻田中的氮肥被当季水稻吸收利用的比率下降,在土壤中的残留比率增加;（3）常耕条件下,湿润灌溉水稻氮素吸收总量明显低于水层灌溉和交替灌溉水稻,但免耕条件下湿润灌溉对水稻氮素吸收总量的影响比较小;（4）湿润灌溉时水稻产量和氮素的生产效率下降。     

施氮量和时期运筹对超级杂交稻植株氮含量与籽粒产量的影响研究

以当前长江中下游稻区具有代表性的高产超级稻品种Y两优1号为材料,采用不同施N量与施N时期分配的田间试验,研究施N对超级稻植株N含量与籽粒产量的影响,从而明确超级稻高产的穗肥N施用量和比例。结果表明：随着施N量的增加,超级稻茎叶和籽粒N含量均呈线性升高,而且施N量对生殖生长阶段茎叶N含量的影响更为显著;与总施N量相比,穗肥N对生殖生长阶段各时期茎叶N含量的影响斜率要高1.88、1.02、1.86和2.28倍,而对成熟期籽粒N含量的影响斜率要高3.59倍。分蘖盛期至抽穗期的茎叶N含量与籽粒产量呈显著线性相关,而成熟期茎叶和籽粒N均与籽粒产量表现为二次抛物线关系;通过模拟方程和不同施N比例处理的叶片外观特征均推算出超级杂交稻达到最高产量的穗肥N用量为59.0 kg/hm2,占总施肥量32.8%。对4个时期施N比例进一步分析表明,各处理籽粒产量与由方程计算的最高产量接近,10-50-25-15处理的籽粒产量要高1.7%,说明合理分配部分N作为粒肥更有利于提高超级杂交稻产量。超级杂交稻合理穗肥（和粒肥）N有利于优化成熟期茎叶和籽粒N含量,从而构建超级稻最大库容量,使其达到最高产量水平。     

南太湖流域控释包膜尿素对水稻产量及稻田氮素流失的影响

通过研究控释包膜尿素(硫磺加树脂包膜尿素SPCU、树脂包膜尿素PCU、硫磺包膜尿素SCU)和普通尿素PU对水稻产量和稻田水体氮含量的影响,为水稻氮素减控施肥技术提供依据。结果表明,等氮条件下,不同氮肥品种对稻田田面水和地下水氮影响较大,与普通尿素(PU)处理相比,SPCU处理可以显著降低水稻灌浆期之前田面水和地下水总氮、氨氮及硝氮的浓度;三种控释肥处理下水稻的产量均比PU处理高,其中SPCU处理最高,比PU处理增产19.7%;SPCU处理的氮肥利用率高达40%,比PU处理高75.4个百分点。因此,从氮肥利用率及环境安全角度出发,SPCU是水稻生产中比较理想的氮素肥料。