Airborne multi-spectral imagery for mapping cruciferous weeds in cereal and legume crops

Cruciferous weeds are competitive broad-leaved species that cause losses in winter crops. In the present study, research on remote sensing was conducted on seven naturally infested fields located in Córdoba and Seville, southern Spain. Multi-spectral aerial images (four bands, including blue (B), green (G), red (R) and near-infrared bands) taken in April 2007 were used to evaluate the feasibility of mapping cruciferous patches (Diplotaxis spp. and Sinapis spp.) in winter crops (wheat, broad bean and pea) and compare the accuracy of different supervised classification methods (vegetation indices, maximum likelihood and spectral angle mapper). The best classification method was selected to develop site-specific cruciferous treatment maps. Cruciferous patches were efficiently discriminated with red/blue (R/B) and blue/green (B/G) vegetation indices and the maximum likelihood classifier. At all of the locations, the accuracy of the results obtained from the spectral angler mapper was relatively low. The cruciferous weed-classified imagery of each location were created according to the method that provided the best discrimination results and were used to obtain site-specific treatment maps for in-season post-emergence control measures or herbicide applications for subsequent years. By applying the site-specific treatment maps, herbicide savings from 71.7 to 95.4% for the no-treatment areas and from 4.3 to 12% for the low-dose herbicide were obtained.     

Site-Specific Weed Control in Maize,Sugar Beet,Winter Wheat,and Winter Barley

A four-year study of site-specific weed control is presented in this paper. Weed seedling distribution was sampled and mapped prior to and after post-emergence herbicide application in four fields planted with maize, sugar beets, winter wheat and winter barley, rotating on one site. Herbicides and other weed control strategies were applied site-specifically according to the spatial variation of weed populations. Different decision algorithms were used for chemical weed control methods in each crop. A weed treatment map was created to direct location and dosage of herbicide application. The sprayer was coupled with a differential Global Positioning System (DGPS). The solenoid valves of the sprayer were opened automatically when the tractor entered a weed patch characterized in the weed treatment map. For site specific herbicide application, a patch sprayer was developed that allowed variable rate application and the selective control of each section of the spray boom.     

The Economic Impact of Site-Specific Weed Control

In order to evaluate economic and ecological benefits, site-specific weed control was realised in a 4-year experiment on five fields with a GPS-guided sprayer. An average of 54% of the herbicides could be saved. Savings were strongly dependent on crop and year. For grass weed herbicides, the savings were 90% in winter cereals, 78% in maize, and 36% in sugar beet. For herbicides against broadleaf weeds, 60% were saved in winter cereals, 11% in maize, and 41% in sugar beet. The monetary savings resulting from the reduction in herbicide use varied between the crops, depending on the amount of herbicides saved and the price of the herbicides. In maize, savings of 42euro/ha were realised, in winter wheat of 32euro/ha, in winter barley of 27euro/ha and in sugar beet of 20euro/ha. Large sections of the fields needed herbicide treatment significantly less frequently. In those areas where weed density remained below the weed control threshold, flora and fauna were allowed to establish largely without disturbance.     

An on-farm approach to quantify yield variation and to derive decision rules for site-specific weed management

Grain yield often varies within agricultural fields as a result of the variation in soil characteristics, competition from weeds, management practices and their causal interactions. To implement appropriate management decisions, yield variability needs to be explained and quantified. A new experimental design was established and tested in a field experiment to detect yield variation in relation to the variation in soil quality, the heterogeneity of weed distribution and weed control within a field. Weed seedling distribution and density, apparent soil electrical conductivity (EC_a) and grain yield were recorded and mapped in a 3.5 ha winter wheat field during 2005 and 2006. A linear mixed model with an anisotropic spatial correlation structure was used to estimate the effect of soil characteristics, weed competition and herbicide treatment on crop yield. The results showed that all properties had a strong effect on grain yield. By adding herbicide costs and current grain price into the model, thresholds of weed density were derived for site-specific weed control. This experimental approach enables the variation of yield within agricultural fields to be explained, and an understanding of the effects on yield of the factors that affect it and their causal interactions to be gained. The approach can be applied to improve decision algorithms for the patch spraying of weeds.     

Early season weed mapping in sunflower using UAV technology: variability of herbicide treatment maps against weed thresholds

Site-specific weed management is defined as the application of customised control treatments only where weeds are located within the crop-field by using adequate herbicide according to weed emergence. The aim of the study was to generate georeferenced weed seedling infestation maps in two sunflower fields by analysing overlapping aerial images of the visible and near-infrared spectrum (using visible or multi-spectral cameras) collected by an unmanned aerial vehicle (UAV) flying at 30 and 60 m altitudes. The main tasks focused on the configuration and evaluation of the UAV and its sensors for image acquisition and ortho-mosaicking, as well as the development of an automatic and robust image analysis procedure for weed seedling mapping used to design a site-specific weed management program. The control strategy was based on seven weed thresholds with 2.5 steps of increasing ratio from 0 % (herbicide must be applied just when there is presence or absence of weed) to 15 % (herbicide applied when weed coverage >15 %). As a first step of the imagery analysis, sunflower rows were correctly matched to the ortho-mosaicked imagery, which allowed accurate image analysis using object-based image analysis [object-based-image-analysis (OBIA) methods]. The OBIA algorithm developed for weed seedling mapping with ortho-mosaicked imagery successfully classified the sunflower-rows with 100 % accuracy in both fields for all flight altitudes and camera types, indicating the computational and analytical robustness of OBIA. Regarding weed discrimination, high accuracies were observed using the multi-spectral camera at any flight altitude, with the highest (approximately 100 %) being those recorded for the 15 % weed threshold, although satisfactory results from 2.5 to 5 % thresholds were also observed, with accuracies higher than 85 % for both field 1 and field 2. The lowest accuracies (ranging from 50 to 60 %) were achieved with the visible camera at all flight altitudes and 0 % weed threshold. Herbicide savings were relevant in both fields, although they were higher in field 2 due to less weed infestation. These herbicide savings varied according to the different scenarios studied. For example, in field 2 and at 30 m flight altitude and using the multi-spectral camera, a range of 23–3 % of the field (i.e., 77 and 97 % of area) could be treated for 0–15 % weed thresholds. The OBIA procedure computed multiple data which permitted calculation of herbicide requirements for timely and site-specific post-emergence weed seedling management.     

Patchy weed distribution and site-specific weed control in winter cereals

Site-specific weed control in winter cereals was performed on the same fields every year over a 5-year period (1999–2003). The most common weeds (Apera spica-venti, Galium aparine, Veronica hederifolia, Viola arvensis) were counted by species, at grid points which were georeferenced and the data were analysed spatially. For weed control, weeds were grouped into three classes: grass, broad-leaved weeds (without Galium aparine), and Galium aparine. Based on weed distribution maps generated by the spatial analyses, herbicide application maps were created and site-specific herbicide application was carried out for grouped and or single weed species. This resulted in a significant reduction in herbicide use. Averaging the results for all fields and years, the total field area treated with herbicides was 39% for grass weeds, 44% for broad-leaved weeds (without Galium aparine) and 49% for Galium aparine. Therefore, site-specific weed control has the potential to reduce herbicide use compared to broadcast application, thus giving environmental and economic benefits.     

Variable-rate nitrogen fertilization of winter wheat under high spatial resolution

Variable-rate application (VRA) addresses in-field variation in soil nitrogen (N) availability and crop response, and as such is a tool for more effective site-specific management. This study assessed the performance of a VRA system for on-the-go delivery of granular fertilizer in 7-m wide and 200-m long strips of a 2.4-ha wheat field. A randomized complete block design consisted of three treatment strips (a preplant uniform application of 100 kg N/ha, a preplant + in-season uniform farmer rate of 212 kg N/ha and a preplant + in-season VRA) within four blocks. The VRA prototype consisted of Crop Circle ACS-430 active canopy sensors, a GeoScout X data logger that processed the geospatial data to convey a real-time N rate signal (1 Hz) to a Gandy Orbit Air 66FSC spreader through a Raven SCS 660 controller. Crop monitoring included analysis of in-season soil and plant samples, water balance and grain yield. VRA delivered an economic optimum N rate using 72% less in-season N or 38% less total N (131 kg N/ha) than that applied by the farmer (212 kg N/ha). The reduction of total N inputs came about without any yield losses and translated to 58% N-use efficiency in comparison to 44% of the farmer practice and 52% of the preplant control. VRA also provided a much higher revenue over fertilizer costs, €68/ha and €118/ha higher than the preplant control and the farmer practice, respectively. The return of VRA per unit of N was equal to that of the large preplant application due to low leaching losses. Overall, the high-resolution VRA was superior in terms of environmental benefits and profitability with the least uncertainty to the farmer.     

Developing a new Crop Circle active canopy sensor-based precision nitrogen management strategy for winter wheat in North China Plain

Active canopy sensor (ACS)—based precision nitrogen (N) management (PNM) is a promising strategy to improve crop N use efficiency (NUE). The GreenSeeker (GS) sensor with two fixed bands has been applied to improve winter wheat (Triticum aestivum L.) N management in North China Plain (NCP). The Crop Circle (CC) ACS-470 active sensor is user configurable with three wavebands. The objective of this study was to develop a CC ACS-470 sensor-based PNM strategy for winter wheat in NCP and compare it with GS sensor-based N management strategy, soil N_min test-based in-season N management strategy and conventional farmer’s practice. Four site-years of field N rate experiments were conducted from 2009 to 2013 to identify optimum CC vegetation indices for estimating early season winter wheat plant N uptake (PNU) and grain yield in Quzhou Experiment Station of China Agricultural University located in Hebei province of NCP. Another nine on-farm experiments were conducted at three different villages in Quzhou County in 2012/2013 to evaluate the performance of the developed N management strategy. The results indicated that the CC ACS-470 sensor could significantly improve estimation of early season PNU (R² = 0.78) and grain yield (R² = 0.62) of winter wheat over GS sensor (R² = 0.60 and 0.33, respectively). All three in-season N management strategies achieved similar grain yield as compared with farmer’s practice. The three PNM strategies all significantly reduced N application rates and increased N partial factor productivity (PFP) by an average of 61–67 %. It is concluded that the CC sensor can improve estimation of early season winter wheat PNU and grain yield as compared to the GS sensor, but the PNM strategies based on these two sensors perform equally well for improving winter wheat NUE in NCP. More studies are needed to further develop and evaluate these active sensor-based PNM strategies under more diverse on-farm conditions.     

Beyond patch spraying: site-specific weed management with several herbicides

Site-specific weed management can include both limiting herbicide application to areas of the field where weed pressure is above the economic threshold (patch spraying) and varying the choice of herbicide for most cost-effective weed control of local populations. The benefits of patch spraying with several, postemergence herbicides in irrigated corn were evaluated in simulation studies using weed counts from 16 fields. Patch spraying with one, two or the number of herbicides that maximized net return for a field was simulated. With patch spraying of one herbicide, the average area of a field left untreated is 34.5%. Net return increases by $3.09 ha⁻¹ compared to a uniform application without decreasing crop yield. Additional herbicides increase the average benefits with just 4% more of the field treated. With two herbicides, the increase in net return is almost tripled and herbicide use is reduced nearly 10-fold compared to patch spraying with one herbicide, and weed control is better than the uniform application in 10 fields. Using more than two herbicides for patch spraying further reduces weed escapes, but herbicide use is greater than a uniform application in 10 fields. Growers might be more willing to adopt patch spraying if more than one herbicide is used in a field.     

Two-Year Weed Seedling Population Responses to a Post-Emergent Method of Site-Specific Weed Management

Field experiments were conducted to determine how a site-specific weed management practice in Zea mays L. influenced the numerical and spatial distribution of a naturally occurring weed infestation in Z. mays and the succeeding Beta vulgaris L. crop. Compared to conventional broadcast herbicide applications, site-specific herbicide applications reduced herbicide load by 11.5 and 98.0% in two separate Z. mays fields. The broad range in outcomes was attributed to the spatial aggregation and density of target weed populations. While herbicide use was successfully reduced at field locations with low weed density, most survivors of multiple control tactics were in locations with the highest initial density. A greater understanding of interactions between weed management and weed density would increase the likelihood that site-specific weed management offers long-term improvements over conventional approaches.     

Development of an in-season estimate of yield potential utilizing optical crop sensors and soil moisture data for winter wheat

When utilizing optical sensors to make in-season agronomic recommendations in winter wheat, one parameter often required is the in-season grain yield potential at the time of sensing. Current estimates use an estimate of biomass, such as normalized difference vegetation index (NDVI), and growing degree days (GDDs) from planting to NDVI data collection. The objective of this study was to incorporate soil moisture data to improve the ability to predict final grain yield in-season. Crop NDVI, GDDs that were adjusted based upon if there was adequate water for crop growth, and the amount of soil profile (0–0.80 m) water were incorporated into a multiple linear regression model to predict final grain yield. Twenty-two site-years of N fertility trials with in-season grain yield predictions for growth stages ranging from Feekes 3 to 10 were utilized to calibrate the model. Three models were developed: one for all soil types, one for loamy soil textured sites, and one for coarse soil textured sites. The models were validated with 11 independent site-years of NDVI and weather data. The results indicated there was no added benefit to having separate models based upon soil types. Typically, the models that included soil moisture, more accurately predicted final grain yield. Across all site years and growth stages, yield prediction estimates that included soil moisture had an R² = 0.49, while the current model without a soil moisture adjustment had an R² = 0.40.     

Weed Mapping with Co-Kriging Using Soil Properties

Our aim is to build reliable weed maps to control weeds in patches. Weed sampling is time consuming but there are some shortcuts. If an intensively sampled variable (e.g. soil property) can be used to improve estimation of a sparsely sampled variable (e.g. weed distribution), one can reduce weed sampling. The geostatistical estimation method co-kriging uses two or more sampled variables, which are correlated, to improve the estimation of one of the variables at locations where it was not sampled. We did an experiment on a 2.1ha winter wheat field to compare co-kriging using soil properties, with kriging based only on one variable. The results showed that co-kriging Lamium spp. from 96 0.25m2 sample plots ha–1 with silt content improved the prediction variance by 11 % compared to kriging. With 51 or 18 sample plots ha–1 the prediction variance was improved by 21 and 15 %.     

Mapping wheat nitrogen uptake from RapidEye vegetation indices

Mapping wheat nitrogen (N) uptake at 5 m spatial resolution could provide growers with new insights regarding nitrogen-use efficiency at the field scale. This study explored the use of spectral information from high resolution (5 × 5 m) RapidEye satellite data at peak leaf area index (LAI) to estimate end-of-season cumulative N uptake of wheat (Triticum spp.) in a heterogeneous, rainfed system. The primary objectives were to evaluate the usefulness of simple, widely used vegetation indices (VIs) from RapidEye as a tool to map crop N uptake over three growing seasons, farms and growing conditions, and to examine the usefulness of remotely sensed N uptake maps for precision agriculture applications. Data on harvested wheat N was collected at twelve plots over three seasons at four farms in the Palouse region of Northern Idaho and Eastern Washington. Seventeen commonly used spectral VIs were computed for images collected during ‘peak greenness’ (maximum LAI) to determine which VIs would be most appropriate for estimating wheat N uptake at harvest. The normalized difference red-edge index was the top performing VI, explaining 81 % of the variance in wheat N uptake with a regression slope of 1.06 and RMSE of 15.94 kg/ha. Model performance was strong across all farms over all three seasons regardless of crop variety, allowing the creation of high accuracy wheat N uptake maps. In conclusion, for this particular agro-ecosystem, mid-season VIs that incorporate the use of the NIR and red-edge bands are generally better predictors of end-of-season crop N uptake than VIs that do not include these bands, thereby further enabling their use in precision agriculture applications.     

Ground-level hyperspectral imagery for detecting weeds in wheat fields

Site-specific weed management can allow more efficient weed control from both an environmental and an economic perspective. Spectral differences between plant species may lead to the ability to separate wheat from weeds. The study used ground-level image spectroscopy data, with high spectral and spatial resolutions, for detecting annual grasses and broadleaf weeds in wheat fields. The image pixels were used to cross-validate partial least squares discriminant analysis classification models. The best model was chosen by comparing the cross-validation confusion matrices in terms of their variances and Cohen’s Kappa values. This best model used four classes: broadleaf, grass weeds, soil and wheat and resulted in Kappa of 0.79 and total accuracy of 85 %. Each of the classes contains both sunlit and shaded data. The variable importance in projection method was applied in order to locate the most important spectral regions for each of the classes. It was found that the red-edge is the most important region for the vegetation classes. Ground truth pixels were randomly selected and their confusion matrix resulted in a Kappa of 0.63 and total accuracy of 72 %. The results obtained were reasonable although the model used wheat and weeds from different growth stages, acquisition dates and fields. It was concluded that high spectral and spatial resolutions can provide separation between wheat and weeds based on their spectral data. The results show feasibility for up-scaling the spectral methods to air or spaceborne sensors as well as developing ground-level application.     

Improved remote sensing detection of wheat powdery mildew using dual-green vegetation indices

In this study, we investigated the possibility of using ground-based remote sensing technology to estimate powdery mildew disease severity in winter wheat. Using artificially inoculated fields, potted plants, and disease nursery tests, we measured the powdery mildew canopy spectra of varieties of wheat at different levels of incidence and growth stages to investigate the disease severity. The results showed that the powdery mildew sensitive bands were between 580 and 710 nm. The best two-band vegetation index that correlated with wheat powdery mildew between 400 and 1000 nm wavelength were the normalized spectrum 570–590 and 536–566 nm bands for the ratio index, and 568–592 and 528–570 nm for the normalized difference index. The coefficients of determination (R ²) for both were almost the same. The optimum dual-green vegetation index was constructed based on a calculation of the ratio and normalized difference between the normalized spectrum within the two green bands. The coefficients of determination (R ²) of DGSR (584, 550) (dual-green simple ratio) and DGND (584, 550) (dual-green normalized difference) were both 0.845. The inverse models of disease severity performed well in the test process at the canopy scale, and indicated that, compared with the traditional vegetation indices of Lwidth, mND705, ND (SDr, SDb), SIPI, and GNDVI, the novel dual-green indices greatly improved the remote sensing detection of wheat powdery mildew disease. Following these results, combined disease severity and canopy spectra were shown to be of enormous value when applied to the accurate monitoring, prevention, and control of crop diseases.     

阔世玛3.6% WG防除冬小麦田杂草试验及对四川省不同类型的冬小麦品种的安全性研究

2000~2002年,在四川省进行了阔世玛3.6%WG防除冬小麦田杂草田间药效试验,除草效果优良。2002~2004年,在四川省进行了阔世玛3.6%WG对四川省不同类型的冬小麦品种的安全性研究,结果表明:在推荐剂量条件下对四川省不同类型的生产小麦品种基本安全,可以在四川冬小麦生产上推广使用。