Modelling carbon dynamics in mixed grass–red clover swards

Growth and carbon dynamics in mixed grass–red clover leys were simulated using a growth model for pure stands based on radiation use, allocation within plant and loss of biomass. The model and its parameter values were taken from previous applications of the model to pure swards of grass and red clover grown in the same experiment at the same sites and years.     

Macroscopic scale soil moisture dynamics model for a wheat crop

Summary A numerical soil moisture dynamics model was developed for; wheat crop using either observed or generated root length densities with root sink incorporating diminishing rate of water uptake by plant roots due to decreasing soil moisture in drying cycles and loss of absorptive power of roots due to ageing. The simulated soil moisture contents were overestimated by 6.0 and 9.6% on an overall basis by the model when observed and generated root length densities were used, respectively, in comparison to observed moisture contents. The model using generated root length densities simulated less water uptake in comparison with the model which utilized observed root length densities.     

香根草叶片铅含量的近红外光谱快速检测

提出了一种应用近红外光谱技术快速检测香根草叶内重金属铅含量的方法,采用多种预处理方法建立偏最小二乘法(PLS)模型并对建模效果对比分析,得出最优预处理方法。结合不同波段选择方法优化PLS模型参数,建立了香根草叶内重金属铅含量定量分析模型,预测决定系数R2为0.87,预测均方根误差RMSEP为0.18。研究结果表明,利用近红外光谱技术快速定量检测香根草叶内重金属铅含量具有可行性。     

基于SegNet与三维点云聚类的大田杨树苗叶片分割方法

准确分割单个杨树叶是无接触提取杨树苗叶表型参数的前提,针对大田杨树苗的复杂种植环境,本文提出一种基于SegNet与三维点云聚类的大田杨树苗叶片分割方法。首先对Kinect V2相机进行标定,对齐RGB与深度数据,滤除背景,获得RGB与深度数据融合数据;然后针对RGB与深度融合数据采用语义分割算法SegNet对杨树苗叶与杨树干进行分割;为了更好地分割出单个杨树叶,对分割的杨树叶区域重构出三维点云,采用基于几何距离的kd-tree对单个树叶进行分类。对采集的单株树苗与多株树苗数据进行了实验分析,采用SegNet与FCN分别对杨树苗叶区域与茎区域进行分割,结果表明,SegNet对叶、茎检测准确率分别为94.4%、97.5%,交并比分别为75.9%、67.9%,优于FCN;对叶区域采用不同距离阈值的kd-tree算法进行单叶分割分析,确定了适合杨树叶的分割阈值。实验结果表明,本文提出的分割算法不仅能分割出单株杨树苗的叶片,也能分割出多株杨树苗的单个叶片。     

基于SegNet与三维点云聚类的大田杨树苗叶片分割方法

准确分割单个杨树叶是无接触提取杨树苗叶表型参数的前提,针对大田杨树苗的复杂种植环境,本文提出一种基于SegNet与三维点云聚类的大田杨树苗叶片分割方法。首先对Kinect V2相机进行标定,对齐RGB与深度数据,滤除背景,获得RGB与深度数据融合数据;然后针对RGB与深度融合数据采用语义分割算法SegNet对杨树苗叶与杨树干进行分割;为了更好地分割出单个杨树叶,对分割的杨树叶区域重构出三维点云,采用基于几何距离的kd-tree对单个树叶进行分类。对采集的单株树苗与多株树苗数据进行了实验分析,采用SegNet与FCN分别对杨树苗叶区域与茎区域进行分割,结果表明,SegNet对叶、茎检测准确率分别为94.4%、97.5%,交并比分别为75.9%、67.9%,优于FCN;对叶区域采用不同距离阈值的kd-tree算法进行单叶分割分析,确定了适合杨树叶的分割阈值。实验结果表明,本文提出的分割算法不仅能分割出单株杨树苗的叶片,也能分割出多株杨树苗的单个叶片。     

基于深度学习的大豆生长期叶片缺素症状检测方法

为了检测作物叶片缺素,提出了一种基于神经网络的大豆叶片缺素视觉检测方法。在对大豆缺素叶片进行特征分析后,采用深度学习技术,利用Mask R-CNN模型对固定摄像头采集的叶片图像进行分割,以去除背景特征,并利用VGG16模型进行缺素分类。首先通过摄像头采集水培大豆叶片图像,对大豆叶片图像进行人工标记,建立大豆叶片图像分割任务的训练集和测试集,通过预训练确定模型的初始参数,并使用较低的学习率训练Mask RCNN模型,训练后的模型在测试集上对背景遮挡的大豆单叶片和多叶片分割的马修斯相关系数分别达到了0.847和0.788。通过预训练确定模型的初始参数,使用训练全连接层的方法训练VGG16模型,训练的模型在测试集上的分类准确率为89.42%。通过将特征明显的叶片归类为两类缺氮特征和4类缺磷特征,分析讨论了模型的不足之处。本文算法检测一幅100万像素的图像平均运行时间为0.8 s,且对复杂背景下大豆叶片缺素分类有较好的检测效果,可为农业自动化生产中植株缺素情况估计提供技术支持。     

不同镇压强度对燕麦生长特性的影响

为探究不同镇压强度对燕麦生长特性的影响,设计了一种镇压强度可调的镇压试验装置,分析了镇压轮与土壤之间的相互作用关系,通过受力分析和公式推导得出镇压强度与配重的关系,通过田间标定试验求解不同镇压强度下试验配重。通过田间试验对燕麦的出苗率、苗期根茎叶干物质积累量、产量等农艺学特性进行了追踪分析,试验结果表明,不同的镇压强度影响燕麦的出苗、苗期生长发育及产量,适宜的镇压强度有助于燕麦生长。当镇压强度为45 kPa时,燕麦生长发育情况表现最佳,燕麦出苗率、苗期根干物质积累量、茎叶干物质积累量、产量分别为78.33%、0.088 g/株、0.546 g/株、2 501.25 kg/hm²,相比不镇压时,分别增加了30.09%、33.33%、32.52%、16.58%。     

不同镇压强度对燕麦生长特性影响的试验研究

为探究不同镇压强度对燕麦生长特性的影响，设计了一种镇压强度可调的镇压试验装置，分析了镇压轮与土壤之间的相互作用关系，通过受力分析和公式推导，得出了镇压强度与配重的关系；通过田间标定试验，求解了不同镇压强度下试验配重。通过田间试验对燕麦的出苗率、苗期根茎叶干物质积累量、产量等农艺学特性进行了追踪分析，试验结果表明，不同的镇压强度影响燕麦的出苗、苗期生长发育及产量，适宜的镇压强度有助于燕麦生长，当镇压强度为45.71kPa时，燕麦产量最大，为2439.3 kg/hm2，比不镇压时增产13.69%，燕麦的出苗率、苗期根的干物质积累量、茎叶的干物质积累量分别为75.66%、0.90g/10株、5.33g/10株，相比与不镇压时，分别增加了25.66%、36.36%、29.37%。     

基于运动恢复结构的无规则植物叶片面积三维测量方法

接触式测量植物叶片面积的方法会对叶片造成一定程度的伤害,为此本文提出一种仅利用智能手机的非接触式多类别无规则叶片面积三维测量方法.首先,采用运动恢复结构方法获取植株的三维重建点云,在HSV颜色特征空间去除叶片三维噪点;然后,利用模糊C均值聚类算法分割单个叶片,重建叶片表面三角网格;最后,通过网格法计算叶片面积.对5种不...     

A growth-irrigation scheduling model for wastewater use in forest production

Applying wastewater and sludge to land for remediation has been recommended by the Environmental Protection Agency (EPA) as a method to recycle nutrient and organic matter and conserve water resources. The level of sewage treatment can range from simple primary treatment using a lagoon to tertiary treatment using a standard wastewater treatment plant. Small communities are selecting primary treatment and land application as the most cost-effective way of treating municipal wastewater.Wastewater was used to irrigate an Eucalyptus camaldulensis plantation in Ojinaga, Chihuahua, Mexico. The overall objective of the research was to develop a daily growth-irrigation scheduling model (GISM) for Eucalyptus tree plantations based on the trees’ water needs with the source of irrigation water being wastewater from a lagoon sewage treatment system. A second objective was to check this model against measured growth data to determine the limitations of using a simple irrigation-scheduling model to manage the irrigation system to maximize tree growth and wood production.The GISM calculated the evapotranspiration (Et) from the volume balance soil water model and a tree biomass sub-model, based on a water use efficiency (WUE) (biomass/Et) that partitions biomass determined from Et into the component parts of leaves and stems plus branches. The water balance portion of the model computes the Et for grass growing between the trees until a closed canopy system is reached. Weather data and a soil water stress function were used to calculate Et based on calculated reference Et and crop coefficients (K_cs) for both the trees and grass scaled to nonstressed Et.The GISM model accurately modeled height and diameter growth, although, it slightly overestimated the height growth of Eucalyptus for the high irrigation treatment in the second and the third years. The GISM model was successful in predicting height and diameter growth within a 95% confidence level of the measured height and diameter of the trees under all irrigation treatments.Based on the modeled and measured data analysis, the GISM model can be a useful tool to predict tree growth and schedule irrigations for Eucalyptus tree plantations, understand the trees response to environmental and water stress, and to provide better analyses for future research efforts. The climate-driving variables (temperature and rainfall) needed by the model are readily available for any location in the world from the National Climatic Data Center (NCDC).     

多种生态模式下的IPM植保技术研究

河道防护林往往以单一的杨树为主,容易受到病虫害侵蚀,采用有害生物综合治理的方法,治理病虫害,可实现防护林生态平衡.首先,对目标林场进行病虫害分析,找出影响树木健康的关键病虫害为天牛;其次,建立树木病虫害损伤等级,采用株均羽化孔数建立损伤等级模型;最后,建立防护林健康阈值,株均羽化孔数为3.当株均羽化孔数大于3时,表明防...     

竹叶片氮含量高光谱估测方法对比研究

为实现快速无损检测竹叶片氮含量,采用波长范围为350~2500nm的地物光谱仪获取竹叶片光谱数据,以金镶玉竹叶片为样本,对其进行高光谱分析。将高光谱原始反射率及其一阶微分、对数一阶微分和二阶微分值,与化学法测量的竹叶片氮含量值进行了相关性分析,分别获得了不同微分变化下的特征波段;基于微分变换后的高光谱反射率数据,分别采用二元线性回归、多元逐步回归、偏最小二乘回归和基于主成分分析的BP神经网络方法,建立了4种金镶玉竹叶片的氮含量高光谱估测模型。对比4种估测模型的校验结果表明,在光谱反射率的对数一阶微分变换下,采用拓扑结构为6-10-1的基于主成分分析的BP神经网络估测模型,校验环节决定系数为0.838,均方根误差RMSE为0.0452,具备较好的竹叶片氮含量估测效果。     

Development of a simple potato growth model for use in crop-pest management

A simple whole-plant level potato growth model was developed that accumulates and partitions dry matter into four state variables—leaves, stems, roots and tubers. Daily growth is computed from a function of the total solar radiation, the proportion of total radiation intercepted by the crop, temperature and soil water status. Dry matter is partitioned using modified Michaelis-Menten equations. At initialization, the seed piece size, plant and row spacing, and the values for the dry matter partitioning parameters are required. After initialization, the inputs required are the daily minimum and maximum temperatures and the daily soil water potential. Daily site-specific, seasonal radiation is presently estimated from a sine function. Model parameterization and output plotted against field data are presented for two cultivars, each planted a total of three times in two seasons. The model is intended for use in analyzing the dynamics of yield under variable pest pressures and environmental conditions. Model sensitivity, behavior and use in relation to modeling yield loss due to pests are discussed.     

Measurements and simulations of heat and water balance components in a clay soil cropped with winter wheat under drought stress or daily irrigation and fertilization

Soil water and temperature dynamics were measured in a field experiment with winter wheat on a clay soil. There were four treatments: Control (C), receiving natural precipitation, drought (D), protected from rain by plastic screens during the growing season, daily irrigation (I) and daily irrigation and fertilization (IF). Treatments C, D and I received the nitrogen fertilizer as a single application of solid fertilizer in spring. In IF daily dressings of nutrients were supplied in the irrigation water. All treatments received 20 g Nm^–2. An associated experiment with a newly sown grass ley (L) that was irrigated and fertilized daily (total 5.6 g Nm ^–2) was also performed. Standard meteorological variables (air temperature and humidity, wind speed, precipitation, global radiation, and relative cloudiness) and crop development data (green area index, crop height, relative root distribution in depth) above and below ground were used as driving variables within a physically based dynamic model (SOIL) for simulating water and heat fluxes. Measured soil temperature and water content from one treatment (I) were used to tune the model parameters, tentatively set from literature data. Thereafter, water and heat fluxes in the other treatments were simulated using the same parameter values but with different crop-related measurements as driving variables for each treatment. Measured soil temperature and water content in C, D, IF and L could thus be used for validation of the simulations. The theory formulated in the model could accurately explain measured treatment differences in soil water and temperature dynamics. Since the soil-related parameters were identical in all treatments, the model was shown to be applicable over a wide range of moisture conditions.     

太阳能草捆干燥设备设计与试验

针对饲草在田间干燥干物质损失大和营养成分保持率低的问题,提出了牧草湿法收获工艺的技术原理、工艺路线,为太阳能饲草干燥实现规模机械化提供技术支撑,设计了可直接进行整捆饲草干燥的太阳能草捆干燥设备。该设备能实现太阳能的自动采集并对含水率在40%左右的整捆饲草进行干燥处理,成品草捆可直接进行贮藏。设备的性能试验和牧草干法收获与湿法收获对比试验结果表明：太阳能干燥饲草损失率小于等于2％,成品草捆含水率小于等于17%,太阳能空气集热器白天平均热效率大于等于0.5,处理能力大于等于1t/h。     

四川盆地丘陵区土壤水分的动态及其随机模拟

土壤水分动态的研究是定量理解植被对水分胁迫响应、土壤养分循环的水文控制、植物水分竞争等生态系统动态的关键,是目前国内外的研究热点。利用2004年-2007年每天的土壤水分监测数据,结合laio土壤水分动态随机模型,研究了四川盆地丘陵区(重庆铜梁虎峰)土壤水分的动态特征及其laio模型在亚热带气候条件推求土壤随机动态特征的适用性。结果表明：监测年内各层土壤水分无论在枯水年还是平水年均差异显著,其中连续平水年土壤水分的含量和变异系数均高于枯水年,枯水年后的平水年低于枯水年;土壤水分的季节变化可分为稳定期、消耗期、波动期;土壤水分的垂直变化来看,土壤水分含量并非完全随着土壤深度的增加而增加。用laio概率随机模型导出的土壤水分概率密度图表明：各层土壤水分的峰度出现在=0.5左右,变化幅度较宽。用laio模型导出的土壤水分峰度和变幅与观测的概率密度函数结果基本一致,相对误差在5%,laio模型可用于分析亚热带气候下土壤水分动态随机特征。     

Cotton root and shoot responses to subsurface drip irrigation and partial wetting of the upper soil profile

The ability of cotton roots to grow downwards through a partially-wetted soil (Calcic Haploxeralf) profile toward a water source located beneath them was investigated. Plants were grown in 60-cm-high soil columms (diameter 10 cm), the bottom 15 cm of which was kept wet by frequent drip irrigation, while the upper 45 cm was wetted three times per week up to 20, 40, 60, 80 or 100% of pot capacity. Pot capacity was defined as the water content which gave uniform distribution within the pot and was at a soil matric potential ( _m ) of –0.01 MPa. Plants were harvested 42 and 70 days after emergence (DAE). Root length density was reduced by decreased soil moisture content. At 42 DAE, density was reduced in the soil profile down to 36 cm. The density in the middle segment of the cylinder (24–36 cm) increased at the second harvest, from 0.1 to 0.35 cm · cm^–3 at 40% and from 0.2 to 0.5 cm · cm^–1 at 60% of pot capacity, respectively. A significant rise in root length density was found at all moisture contents above 20% in the two deepest soil segments. It was most marked at 40% where the rise was from 0.2 to 0.8 cm · cm^–3, due to the development of secondary roots at the wetted bottom of the column. When only 20% of pot capacity was maintained in the top 45 cm of the profile, almost no roots reached the wetted soil volume, and root length density was very low. Hydrotropism, namely root growth through dry soil layers toward a wet soil layer was thus not apparent. Root dry weight per unit length decreased with increasing depth in the column at all moisture levels. However, the only significant decrease was, found between the top and the second soil segments and was due to thicker primary roots in the top segment. There was no clear relationship between length and dry weight of roots. Total plant dry weight and transpiration were reduced significantly only at 20% of pot capacity. Dry matter production by roots was less severely inhibited than that by shoots, under decreased moisture content in the soil profile. Leaf water potential decreased when the soil moisture content of the top 45 cm of the profile was reduced below 60% of pot capacity. It was concluded that even at soil moisture content equivalent to a _m of 0.1 MPa, the rate of root growth was sufficient to reach a wetted soil layer at the bottom of the soil column, where the plant roots then proliferated. This implies that as long as the soil above the subsurface dripper is not very dry there is no real need for early surface irrigation.     

基于改进遗传算法的棉花异性纤维目标特征选择

为提高基于机器视觉的棉花异性纤维在线分类的精度和速度,提出了一种基于改进遗传算法的特征选择方法.采用分段式染色体管理方案实现对多质特征空间局部化管理;利用分段交叉和变异算子避免出现无效染色体,提高搜索效率;通过自适应调整交叉和变异概率实现强搜索能力和快收敛速度的动态平衡.实验结果表明,该方法比基本遗传算法搜索能力更强、收敛速度更快,所得最优特征子集较小,更适用于棉花异性纤维在线分类.     

Spatial extrapolation of the vine (Vitis vinifera L.) water status: a first step towards a spatial prediction model

The goal of this study is to propose a model that allows for spatial extrapolation of the vine water status over a whole field from a single reference site. The precision of the model was tested using data of spatial plant water status from a commercial vineyard block located in the Languedoc-Roussillon region, France. Observations of plant water status were made on 49 sites (three vines per site) on a regular grid at various times in the growing seasons over two non-irrigated fields planted with Shiraz and Mourvèdre cultivars. Plant water status was determined by measuring predawn leaf water potential (PLWP). Results showed a significant within-field variability of PLWP over space and time, and the existence of significant linear relationship amongst PLWP values measured at different dates. Based on these results, a linear model of spatial extrapolation of PLWP values was proposed. This model was able to predict spatial variability of PLWP with a spatial and temporal mean error less than 0.1 MPa on Shiraz as well as on Mourvèdre. This model provides maps of spatial variability in PLWP at key phenological stages on the basis of one measurement performed on a reference site. The model calibration is, in its current state, based on a significant database of PLWP measurements. This makes unrealistic its application to commercial vineyards. However, the approach constitutes a significant step towards the spatial extrapolation of vine water status. Finally, the study mentions alternative ways to build up such models using auxiliary information such as airborne imagery, apparent soil conductivity and easily measured vine/canopy development parameters.     

玉米叶片净光合速率快速检测方法研究

玉米叶片的净光合速率可以用来表征植物生物量的积累和营养盈亏等健康状态,为探求玉米叶片净光合速率的快速无损检测方法,利用叶绿素荧光光谱分析技术对拔节期玉米叶片净光合速率进行检测。实验选取了吉林省典型种植品种先玉335作为研究对象,通过对80组数据的无量纲化处理和标准化处理,降低光谱噪声引起的样本差异,分析不同光谱波段与叶片净光合速率的相关性,确定500~550nm、675~715nm、715~745nm等3组波段作为光谱检测样本。选择675~715nm波段作为光谱波段的典型参数预测玉米叶片的净光合速率,得出两者之间存在显著线性关系,其决定系数R^2=0.7 9 2 4,表明以6 7 5~7 1 5 nm波段预测玉米叶片的净光合速率是可行的。对回归模型进行验证,得到预测值与真实值之间的决定系数R^2=0.7 9 2 1,表明此回归模型对拔节期玉米叶片净光合速率具有良好预测能力,为植物生理信息快速无损检测提供了新的方法。