REMOTE SENSING BASED ESTIMATION SYSTEM FOR WINTER WHEAT YIELD IN NORTH CHINA PLAIN

This paper presents the applications of Landsat Thematic Mapper (TM) data and Advanced Very High Resolution Radiometer (AVHRR) time series data for winter wheat production estimation in North China Plain. The keytechniques are described systematically about winter wheat yield estimation system, including automatically extractingwheat area, simulating and monitoring wheat growth situation, building wheat unit yield model of large area and forecasting wheat production. Pattern recognition technique was applied to extract sown area using TM data. Temporal NDVI(Normal Division Vegetation Index) profiles were produced from 8 - 12 times AVHRR data during wheat growth dynamically. A remote sensing yield model for large area was developed based on greenness accumulation, temperature andgreenness change rate. On the basis of the solution of key problems, an operational system for winter wheat yield estimation in North China Plain using remotely sensed data was established and has operated since 1993, which consists of 4 subsystems, namely databases management, image processing, models bank management and production prediction system.The accuracy of wheat production prediction exceeded 96 per cent compared with on the spot measurement.

Integrating a novel irrigation approximation method with a process-based remote sensing model to estimate multi-years'winter wheat yield over the North China Plain

Accurate estimation of regional winter wheat yields is essential for understanding the food production status and ensuring national food security.However,using the existing remote sensing-based crop yield models to accurately reproduce the inter-annual and spatial variations in winter wheat yields remains challenging due to the limited ability to acquire irrigation information in water-limited regions.Thus,we proposed a new approach to approximating irrigations of winter wheat over the North China Plain(NCP),where irrigation occurs extensively during the winter wheat growing season.This approach used irrigation pattern parameters(IPPs)to define the irrigation frequency and timing.Then,they were incorporated into a newly-developed process-based and remote sensing-driven crop yield model for winter wheat(PRYM–Wheat),to improve the regional estimates of winter wheat over the NCP.The IPPs were determined using statistical yield data of reference years(2010–2015)over the NCP.Our findings showed that PRYM–Wheat with the optimal IPPs could improve the regional estimate of winter wheat yield,with an increase and decrease in the correlation coefficient(R)and root mean square error(RMSE)of 0.15(about 37%)and 0.90 t ha–1(about 41%),respectively.The data in validation years(2001–2009 and 2016–2019)were used to validate PRYM–Wheat.In addition,our findings also showed R(RMSE)of 0.80(0.62 t ha–1)on a site level,0.61(0.91 t ha–1)for Hebei Province on a county level,0.73(0.97 t ha–1)for Henan Province on a county level,and 0.55(0.75 t ha–1)for Shandong Province on a city level.Overall,PRYM–Wheat can offer a stable and robust approach to estimating regional winter wheat yield across multiple years,providing a scientific basis for ensuring regional food security.

Improved simulation of winter wheat yield in North China Plain by using PRYM-Wheat integrated dry matter distribution coefficient

The accurate simulation of regional-scale winter wheat yield is important for national food security and the balance of grain supply and demand in China.Presently,most remote sensing process models use the“biomass×harvest index(HI)”method to simulate regional-scale winter wheat yield.However,spatiotemporal differences in HI contribute to inaccuracies in yield simulation at the regional scale.Time-series dry matter partition coefficients(Fr)can dynamically reflect the dry matter partition of winter wheat.In this study,Fr equations were fitted for each organ of winter wheat using site-scale data.These equations were then coupled into a process-based and remote sensingdriven crop yield model for wheat(PRYM-Wheat)to improve the regional simulation of winter wheat yield over the North China Plain(NCP).The improved PRYM-Wheat model integrated with the fitted Fr equations(PRYM-Wheat-Fr)was validated using data obtained from provincial yearbooks.A 3-year(2000-2002)averaged validation showed that PRYM-Wheat-Fr had a higher coefficient of determination(R^(2)=0.55)and lower root mean square error(RMSE=0.94 t ha^(-1))than PRYM-Wheat with a stable HI(abbreviated as PRYM-Wheat-HI),which had R^(2) and RMSE values of 0.30 and 1.62 t ha^(-1),respectively.The PRYM-Wheat-Fr model also performed better than PRYM-Wheat-HI for simulating yield in verification years(2013-2015).In conclusion,the PRYM-Wheat-Fr model exhibited a better accuracy than the original PRYM-Wheat model,making it a useful tool for the simulation of regional winter wheat yield.

Winter wheat identification by integrating spectral and temporal information derived from multi-resolution remote sensing data

Timely crop acreage and distribution information are the basic data which drive many agriculture related applications.For identifying crop types based on remote sensing,methods using only a single image type have significant limitations.Current research that integrates fine and coarser spatial resolution images,using techniques such as unmixing methods,regression models,and others,usually results in coarse resolution abundance without sufficient detail within pixels,and limited attention has been paid to the spatial relationship between the pixels from these two kinds of images.Here we propose a new solution to identify winter wheat by integrating spectral and temporal information derived from multi-resolution remote sensing data and determine the spatial distribution of sub-pixels within the coarse resolution pixels.Firstly,the membership of pixels which belong to winter wheat is calculated using a 25-m resolution resampled Landsat Thematic Mapper(TM)image based on the Bayesian equation.Then,the winter wheat abundance(acreage fraction in a pixel)is assessed by using a multiple regression model based on the unique temporal change features from moderate resolution imaging spectroradiometer(MODIS)time series data.Finally,winter wheat is identified by the proposed Abundance-Membership(AM)model based on the spatial relationship between the two types of pixels.Specifically,winter wheat is identified by comparing the spatially corresponding 10×10 membership pixels of each abundance pixel.In other words,this method takes advantage of the relative size of membership in a local space,rather than the absolute size in the entire study area.This method is tested in the major agricultural area of Yiluo Basin,China,and the results show that acreage accuracy(Aa)is 93.01%and sampling accuracy(As)is 91.40%.Confusion matrix shows that overall accuracy(OA)is 91.4%and the kappa coefficient(Kappa)is 0.755.These values are significantly improved compared to the traditional Maximum Likelihood classification(MLC)and Random Forest classification(RFC)which rely on spectral features.The results demonstrate that the identification accuracy can be improved by integrating spectral and temporal information.Since the identification of winter wheat is performed in the space corresponding to each MODIS pixel,the influence of differences of environmental conditions is greatly reduced.This advantage allows the proposed method to be effectively applied in other places.

Establishment of Winter Wheat Regional Simulation Model Based on Remote Sensing Data and Its Application

Accurate crop growth monitoring and yield forecasting are significant to the food security and the sustainable development of agriculture. Crop yield estimation by remote sensing and crop growth simulation models have highly potential application in crop growth monitoring and yield forecasting. However, both of them have limitations in mechanism and regional application, respectively. Therefore, approach and methodology study on the combination of remote sensing data and crop growth simulation models are concerned by many researchers. In this paper, adjusted and regionalized WOFOST （World Food Study） in North China and Scattering by Arbitrarily Inclined Leaves-a model of leaf optical PROperties SPECTra （SAIL-PROSFPECT） were coupled through LAI to simulate Soil Adjusted Vegetation Index （SAVI） of crop canopy, by which crop model was re-initialized by minimizing differences between simulated and synthesized SAVI from remote sensing data using an optimization software （FSEOPT）. Thus, a regional remote-sensingcrop-simulation-framework-model （WSPFRS） was established under potential production level （optimal soil water condition）. The results were as follows： after re-initializing regional emergence date by using remote sensing data, anthesis, and maturity dates simulated by WSPFRS model were more close to measured values than simulated results of WOFOST; by re-initializing regional biomass weight at turn-green stage, the spatial distribution of simulated storage organ weight was more consistent with measured yields and the area with high values was nearly consistent with actual high yield area. This research is a basis for developing regional crop model in water stress production level based on remote sensing data.

Monitoring of winter wheat distribution and phenological phases based on MODIS time-series： A case study in the Yellow River Delta, China

Accurate winter wheat identification and phenology extraction are essential for field management and agricultural policy making. Here, we present mechanisms of winter wheat discrimination and phenological detection in the Yellow River Delta（YRD） region using moderate resolution imaging spectroradiometer（MODIS） time-series data. The normalized difference vegetation index（NDVI） was obtained by calculating the surface reflectance in red and infrared. We used the Savitzky-Golay filter to smooth time series NDVI curves. We adopted a two-step classification to identify winter wheat. The first step was designed to mask out non-vegetation classes, and the second step aimed to identify winter wheat from other vegetation based on its phenological features. We used the double Gaussian model and the maximum curvature method to extract phenology. Due to the characteristics of the time-series profiles for winter wheat, a double Gaussian function method was selected to fit the temporal profile. A maximum curvature method was performed to extract phenological phases. Phenological phases such as the green-up, heading and harvesting phases were detected when the NDVI curvature exhibited local maximum values. The extracted phenological dates then were validated with records of the ground observations. The spatial patterns of phenological phases were investigated. This study concluded that, for winter wheat, the accuracy of classification is 87.07%, and the accuracy of planting acreage is 90.09%. The phenological result was comparable to the ground observation at the municipal level. The average green-up date for the whole region occurred on March 5, the average heading date occurred on May 9, and the average harvesting date occurred on June 5. The spatial distribution of the phenology for winter wheat showed a significant gradual delay from the southwest to the northeast. This study demonstrates the effectiveness of our proposed method for winter wheat classification and phenology detection.

Application of EOS/MODIS-NDVI at Different Time Sequences on Monitoring Winter Wheat Acreage in Henan Province

[Objective] Calculation of winter wheat acreage in Henan Province using EOS/MODIS-NDVI data at different time sequences. [Method] After process of EOS/MODIS images, geographical adjustment, wave band combination, normal difference vegetation index (NDVI) was obtained. Based on the wide spectrum measurement, the processed data were supervisedly classified, thus the acreage of winter wheat in Henan Province in 2005 was acquired. [Result] Total 92208 pixels were observed for the winter wheat in Henan Province, and the plantation acreage was 5 760 thousand hm2. Compared with the data from statistical department, the error of this method was about 9.66%. [Conclusion] The method introduced in the present study could be applied in monitoring winter wheat acreage.

Rapid determination of leaf water content for monitoring waterlogging in winter wheat based on hyperspectral parameters

Waterlogging is becoming an obvious constraint on food production due to the frequent occurrence of extremely high-level rainfall events.Leaf water content(LWC)is an important waterlogging indicator,and hyperspectral remote sensing provides a non-destructive,real-time and reliable method to determine LWC.Thus,based on a pot experiment,winter wheat was subjected to different gradients of waterlogging stress at the jointing stage.Leaf hyperspectral data and LWC were collected every 7 days after waterlogging treatment until the winter wheat was mature.Combined with methods such as vegetation index construction,correlation analysis,regression analysis,BP neural network(BPNN),etc.,we found that the effect of waterlogging stress on LWC had the characteristics of hysteresis and all waterlogging stress led to the decrease of LWC.LWC decreased faster under severe stress than under slight stress,but the effect of long-term slight stress was greater than that of short-term severe stress.The sensitive spectral bands of LWC were located in the visible(VIS,400–780 nm)and short-wave infrared(SWIR,1400–2500 nm)regions.The BPNN Model with the original spectrum at 648 nm,the first derivative spectrum at 500 nm,the red edge position(λr),the new vegetation index RVI(437,466),NDVI(437,466)and NDVI´(747,1956)as independent variables was the best model for inverting the LWC of waterlogging in winter wheat(modeling set:R^(2)=0.889,RMSE=0.138;validation set:R^(2)=0.891,RMSE=0.518).These results have important theoretical significance and practical application value for the precise control of waterlogging stress.

Effects of meteorological factors on different grades of winter wheat growth in the Huang-Huai-Hai Plain, China

The sown area of winter wheat in the Huang-Huai-Hai（HHH） Plain accounts for over 65% of the total sown area of winter wheat in China. Thus, it is important to monitor the winter wheat growth condition and reveal the main factors that influence its dynamics. This study assessed the winter wheat growth condition based on remote sensing data, and investigated the correlations between different grades of winter wheat growth and major meteorological factors corresponding. First, winter wheat growth condition from sowing until maturity stage during 2011–2012 were assessed based on moderate-resolution imaging spectroradiometer（MODIS） normalized difference vegetation index（NDVI） time-series dataset. Next, correlation analysis and geographical information system（GIS） spatial analysis methods were used to analyze the lag correlations between different grades of winter wheat growth in each phenophase and the meteorological factors that corresponded to the phenophases. The results showed that the winter wheat growth conditions varied over time and space in the study area. Irrespective of the grades of winter wheat growth, the correlation coefficients between the winter wheat growth condition and the cumulative precipitation were higher than zero lag（synchronous precipitation） and one lag（pre-phenophase precipitation） based on the average values of seven phenophases. This showed that the cumulative precipitation during the entire growing season had a greater effect on winter wheat growth than the synchronous precipitation and the pre-phenophase precipitation. The effects of temperature on winter wheat growth varied according to different grades of winter wheat growth based on the average values of seven phenophases. Winter wheat with a better-than-average growth condition had a stronger correlation with synchronous temperature, winter wheat with a normal growth condition had a stronger correlation with the cumulative temperature, and winter wheat with a worse-than-average growth condition had a stronger correlation with the pre-phenophase temperature. This study may facilitate a better understanding of the quantitative correlations between different grades of crop growth and meteorological factors, and the adjustment of field management measures to ensure a high crop yield.

Change of winter wheat planting area and its impacts on groundwater depletion in the North China Plain

The North China Plain is one of the most water-stressed areas in China. Irrigation of winter wheat mainly utilizes groundwater resources, which has resulted in severe environmental problems. Accurate estimation of crop water consumption and net irrigation water consumption is crucial to guarantee the management of agricultural water resources. An actual crop evapotranspiration(ET) estimation model was proposed, by combining FAO Penman-Monteith method with remote sensing data. The planting area of winter wheat has a significant impact on water consumption; therefore, the planting area was also retrieved. The estimated ET showed good agreement with field-observed ET at four stations. The average relative bias and root mean square error(RMSE) for ET estimation were –2.2% and 25.5 mm, respectively. The results showed the planting area and water consumption of winter wheat had a decreasing trend in the Northern Hebei Plain(N-HBP) and Southern Hebei Plain(S-HBP). Moreover, in these two regions, there was a significant negative correlation between accumulated net irrigation water consumption and groundwater table. The total net irrigation water consumption in the N-HBP and S-HBP accounted for 12.9×10~9 m^3 and 31.9×10~9 m^3 during 2001–2016, respectively. Before and after 2001, the decline rate of groundwater table had a decreasing trend, as did the planting area of winter wheat in the N-HBP and S-HBP. The decrease of winter wheat planting area alleviated the decline of groundwater table in these two regions while the total net irrigation water consumption was both up to 28.5×10~9 m^3 during 2001–2016 in the Northwestern Shandong Plain(NW-SDP) and Northern Henan Plain(N-HNP). In these two regions, there was no significant correlation between accumulated net irrigation water consumption and groundwater table. The Yellow River was able to supply irrigation and the groundwater table had no significant declining trend.

Comparison of three models for winter wheat yield prediction based on UAV hyperspectral images

Predicting crop yield timely can considerably accelerate agricultural production management and food policy-making,which are also important requirements for precise agricultural development.Given the development of hyperspectral imaging technology,a simple and efficient modeling method is convenient for predicting crop yield by using airborne hyperspectral images.In this study,the Unmanned Aerial Vehicle(UAV)hyperspectral and maturity yield data in 2014-2015 and 2017-2018 were collected.The winter wheat yield prediction model was established by optimizing Vegetation Indices(VIs)feature scales and sample scales,incorporating Partial Least Squares Regression(PLSR),Random Forest algorithm(RF),and Back Propagation Neural Network algorithm(BPN).Results showed that PLSR stands out as the optimal wheat yield prediction model considering stability and accuracy(RMSE=948.88 kg/hm2).Contrary to the belief that more input features result in higher accuracy,PLSR,RF,and BPN models performed best when trained with the top 3,8,and 4 VIs with the highest correlation,respectively.With an increase in training samples,model accuracy improves,reaching stability when the training samples reach 70.Using PLSR and optimal feature scales,UAV yield prediction maps were generated,holding significant value for field management in precision agriculture.

Comparison of three remotely sensed drought indices for assessing the impact of drought on winter wheat yield

Agricultural drought threatens food security.Numerous remote-sensing drought indices have been developed,but their different principles,assumptions and physical quantities make it necessary to compare their suitability for drought monitoring over large areas.Here,we analyzed the performance of three typical remote sensing-based drought indices for monitoring agricultural drought in two major agricultural production regions in Shaanxi and Henan provinces,northern China(predominantly rain-fed and irrigated agriculture,respectively):vegetation health index(VHI),temperature vegetation dryness index(TVDI)and drought severity index(DSI).We compared the agreement between these indices and the standardized precipitation index(SPI),soil moisture,winter wheat yield and National Meteorological Drought Monitoring(NMDM)maps.On average,DSI outperformed the other indices,with stronger correlations with SPI and soil moisture.DSI also corresponded better with soil moisture and NMDM maps.The jointing and grain-filling stages of winter wheat are more sensitive to water stress,indicating that winter wheat required more water during these stages.Moreover,the correlations between the drought indices and SPI,soil moisture,and winter wheat yield were generally stronger in Shaanxi province than in Henan province,suggesting that remote-sensing drought indices provide more accurate predictions of the impacts of drought in predominantly rain-fed agricultural areas.

Estimation model of winter wheat disease based on meteorological factors and spectral information

Wheat scab(WS,Fusarium head blight),one of the most severe diseases of winter wheat in Yangtze-Huaihe river region,whose monitoring and timely forecasting at large scale would help to optimize pesticide spraying and achieve the purpose of reducing yield loss.In the present study,remote sensing monitoring on WS was conducted in 4 counties in Yangtze-Huaihe river region.Sensitive factors of WS were selected to establish the remote sensing estimation model of winter wheat scab index(WSI)based on interactions between spectral information and meteorological factors.The results showed that:1)Correlations between the daily average temperature(DAT)and daily average relative humidity(DAH)at different time scales and WSI were significant.2)There were positive linear correlations between winter wheat biomass,leaf area index(LAI),leaf chlorophyll content(LCC)and WSI.3)NDVI(normalized difference vegetation index),RVI(ratio vegetation index)and DVI(difference vegetation index)which had a good correlation with LAI,biomass and LCC,respectively,and could be used to replace them in modeling.4)The estimated values of the model were consistent with the measured values(RMSE=5.3%,estimation accuracy=90.46%).Estimation results showed that the model could efficiently estimate WS in Yangtze-Huaihe river region.

基于遥感多参数和CNN-Transformer的冬小麦单产估测

为了提高冬小麦单产估测精度,改善估产模型存在的高产低估和低产高估等现象,以陕西省关中平原为研究区域,选取旬尺度条件植被温度指数(VTCI)、叶面积指数(LAI)和光合有效辐射吸收比率(FPAR)为遥感特征参数,结合卷积神经网络(CNN)局部特征提取能力和基于自注意力机制的Transformer网络的全局信息提取能力,构建CNN-Transformer深度学习模型,用于估测关中平原冬小麦产量。与Transformer模型(R^(2)为0.64,RMSE为465.40 kg/hm^(2),MAPE为8.04%)相比,CNN-Transformer模型具有更高的冬小麦单产估测精度(R^(2)为0.70,RMSE为420.39 kg/hm^(2),MAPE为7.65%),能够从遥感多参数中提取更多与产量相关的信息,且对于Transformer模型存在的高产低估和低产高估现象均有所改善。基于5折交叉验证法和留一法进一步验证了CNN-Transformer模型的鲁棒性和泛化能力。此外,基于CNN-Transformer模型捕获冬小麦生长过程的累积效应,分析逐步累积旬尺度输入参数对产量估测的影响,评估模型对于冬小麦不同生长阶段的累积过程的表征能力。结果表明,模型能有效捕捉冬小麦生长的关键时期,3月下旬至5月上旬是冬小麦生长的关键时期。

基于改进HRNet的遥感影像冬小麦语义分割方法

冬小麦在影像中呈现田块碎小且分布零散等空间特征,同时影像包含的复杂地物对冬小麦识别造成干扰,易出现识别精度低且边界分割模糊等问题。为及时准确获取大范围冬小麦空间分布信息,该研究以高分二号卫星影像作为数据源,提出一种CAHRNet(change attention high-resolution Net)语义分割模型。采用HRNet(high-resolution Net)替换ResNet作为模型的主干网络,网络的并行交互方式易获取高分辨率的特征信息;联合OCR(object-contextual representations)模块聚合上下文信息,以增强像素点与目标对象区域的关联性;3)引入坐标注意力(coordinate attention)机制,使网络模型充分利用有效的空间位置信息,以保留分割区域的边缘细节,提高对分布零散、形状多变的冬小麦田块的特征提取能力。试验结果表明,在自制的高分辨率遥感数据集上,CAHRNet模型的平均交并比(mean intersection over union,MIoU)和像素准确率(pixel accuracy, PA)分别达到81.72%和97.08%,MIoU相较U-Net、DeepLabv3+分别提高了9.09、2.44个百分点;PA相较U-Net、DeepLabv3+分别提高6.80、1.59个百分点,说明CAHRNet模型具有较高的分割识别精度,可为进一步准确获取冬小麦作物分布信息提供技术支撑。

基于遥感多参数和IPSO-WNN的冬小麦单产估测

冬小麦是我国的主要粮食作物之一。为进一步准确地估测冬小麦产量,以陕西省关中平原为研究区域,选取冬小麦主要生育期与水分胁迫和光合作用等密切相关的条件植被温度指数(VTCI)、叶面积指数(LAI)和光合有效辐射吸收比率(FPAR)作为遥感特征参数,采用改进的粒子群算法优化小波神经网络(IPSO-WNN)以改善梯度下降方法易陷入局部最优的缺陷,并构建冬小麦产量估测模型。结果表明,IPSO-WNN模型的决定系数R2为0.66,平均绝对百分比误差(MAPE)为7.59%,相比于BPNN(R2=0.46,MAPE为11.80%)与WNN(R2=0.52,MAPE为9.80%),IPSO-WNN能够进一步提高模型的精度、增强模型的鲁棒性。采用灵敏度分析的方法探究对冬小麦产量影响较大的输入参数,结果发现,抽穗-灌浆期的FPAR对冬小麦产量影响最大,其次拔节期的VTCI、抽穗-灌浆期和乳熟期的LAI以及返青期和拔节期的FPAR对冬小麦产量的影响较大。通过IPSO-WNN输出获取冬小麦综合监测指数I,构建I与统计单产之间的估产模型以估测关中平原冬小麦单产,结果显示,估测单产与统计单产之间的R2为0.63,均方根误差(RMSE)为505.50 kg/hm^(2),相比于前人的研究较好地解决了估产模型存在的“低产高估”的问题,因此,本文基于IPSO-WNN构建的估产模型能够较准确地估测关中平原冬小麦产量。

公里网格尺度的陕西冬小麦综合遥感指数保险产品设计

基于遥感时序数据MOD13A2和陕西省99个气象站点的逐日气象数据,使用EVI差值法提取陕西冬小麦种植区,筛选与冬小麦单产相关性最高的遥感指数,结合冬小麦生育期内倒春寒、干旱、连阴雨以及干热风4个农业气象灾害的气象指标,构建综合遥感指数模型,以覆盖冬小麦全生育期的农业气象灾害风险。基于最优产量预测模型设计冬小麦综合遥感指数保险,采用分布拟合和蒙特卡洛模拟方法,计算10770个公里网格冬小麦综合遥感指数保险的理赔门槛值和精算纯费率,绘制理赔门槛地图和精算纯费率地图。结果表明:(1)采用EVI差值法提取冬小麦种植区,不同区域使用不同差值时段和门槛值,可获得较高提取精度,县域提取面积与2020年实际播种面积的相关系数达0.997,平均绝对误差524.9hm²;(2)2000-2020年第65、81天EVI及两者最大值与陕西冬小麦单产相关性较高,县域相关系数年平均值最高达0.692;(3)融合干旱、连阴雨和倒春寒3个农业气象灾害的气象指标,最优综合遥感指数模型模拟单产与实际单产的相关系数达0.837,最优综合遥感指数模型R²为0.602;(4)关中、陕南部分地区冬小麦种植风险较低,平均单产损失率低于2%,渭河与黄河交口处冬小麦种植风险较高,平均单产损失率高于4%,其余地区冬小麦种植风险介于两者之间。冬小麦单产和种植风险在县域以下区域存在较大空间差异,提高测算理赔门槛和精算纯费率的空间精度,能够使冬小麦高产地区和低产地区获得同样的赔付机会,避免赔付超过实际损失导致的道德风险,同时能够使费率与实际种植风险相匹配,增加费率公平性,提高低风险地区参保意愿,减少逆选择。

基于多源遥感数据与模型对比的冬小麦土壤含水量区域监测研究

实时、精准的土壤水分含量监测是农业用水管理的基础,探究冬小麦土壤水分反演的最优模型对于提高农业用水效率和可持续发展均具有重要的意义。本研究以河南省鹤壁市浚县冬小麦种植区域的土壤水分含量为研究对象,采用无人机遥感数据、卫星遥感数据、田间采样数据,分别运用温度植被干旱指数模型、水云模型和改进的水云模型3种方法,进行土壤含水量反演对比分析与最优模型选择。结果表明,3种方法中10 cm深度的反演精度均高于20 cm,且R^(2)均大于0.4。其中采用改进的水云模型方法在10 cm深度的R^(2)为0.7055、RMSE为0.0209,20 cm深度的R^(2)为0.5069、RMSE为0.0271,优于水云模型和温度植被干旱指数的反演效果。因此,改进的水云模型是一种适合用于麦田土壤水分反演的方法,它能够提供较高的反演精度。

基于无人机影像的冬小麦株高提取与LAI估测模型构建

株高和叶面积指数(Leaf Area Index,LAI)反映着作物的生长发育状况。为了探究基于无人机可见光遥感提取冬小麦株高的可靠性,以及利用株高和可见光植被指数估算LAI的精度,本文获取了拔节期、抽穗期、灌浆期的无人机影像,提取了冬小麦株高与可见光植被指数,使用逐步回归、偏最小二乘、随机森林、人工神经网络四种方法建立LAI估测模型,并对株高提取及LAI估测情况进行精度评价。结果显示:(1)株高提取值Hc与实测值Hd高度拟合(R^(2)=0.894,RMSE=6.695,NRMSE=9.63%),株高提取效果好;(2)与仅用可见光植被指数相比,基于株高与可见光植被指数构建的LAI估测模型精度更高,且随机森林为最优建模方法,当其决策树个数为50时模型估测效果最好(R^(2)=0.809,RMSE=0.497,NRMSE=13.85%,RPD=2.336)。利用无人机可见光遥感方法,高效、准确、无损地实现冬小麦株高及LAI提取估测可行性较高,该研究结果可为农情遥感监测提供参考。

基于遥感多参数和VMD-GRU的冬小麦单产估测

为充分挖掘时间序列遥感参数的时序信息和趋势信息,并进一步提升冬小麦估产精度,以陕西省关中平原为研究区域,选取与冬小麦长势密切相关的生育时期尺度的条件植被温度指数(VTCI)、叶面积指数(LAI)和光合有效辐射吸收比率(FPAR)作为遥感参数,构建耦合变分模态分解(VMD)与门控循环单元(GRU)神经网络的估产模型。应用VMD算法将各个时间序列遥感参数分解为多组平稳的本征模态函数(IMF)分量,选取与原始时间序列遥感参数高度相关的IMF分量进行特征重构,并将重构特征作为GRU网络的输入,以构建冬小麦组合估产模型。结果表明,VMD-GRU组合估产模型决定系数为0.63,均方根误差为448.80 kg/hm^(2),平均相对误差为8.14%,相关性达到极显著水平(P<0.01),其精度优于单一估产模型精度,表明该组合估产模型能够提取非平稳时间序列数据的多尺度、多层次特征,并充分挖掘冬小麦各生育时期遥感参数间的内在联系,获得准确单产估测结果的同时提升了估产模型的可解释性。