Precision Agriculture Research Research Articles

A discussion on the significance associated with Pearson’s correlation in precision agriculture studies

Pearson’s correlation is a commonly used descriptive statistic in many published precision agriculture studies, not only in the Precision Agriculture Journal, but also in other journals that publish in this domain. Very few of these articles take into consideration auto-correlation in data when performing correlation analysis, despite a statistical solution being available. A brief discussion on the need to consider auto-correlation and the effective sample size when using Pearson’s correlation in precision agriculture research is presented. The discussion is supported by an example using spatial data on vine size and canopy vigour in a juice-grape vineyard. The example data demonstrated that the p-value of the correlation between vine size and canopy vigour increased when auto-correlation was accounted for, potentially to a non-significant level depending on the desired α-level. The example data also demonstrated that the method by which data are processed (interpolated) to achieve co-located data will also affect the amount of auto-correlation and the effective sample size. The results showed that for the same variables, with different approaches to data co-location, a lower r-value may have a lower p-value and potentially hold more statistical significance.

Quantitative Evaluation of Surface Color of Tomato Fruits Cultivated in Remote Farm Using Digital Camera Images

To measure the quantitative surface color information of agricultural products with the ambient information during cultivation, a color calibration method for digital camera images and a remote monitoring system of color imaging using the Web were developed. Single-lens reflex and web digital cameras were used for the image acquisitions. The tomato images through the post-ripening process were taken by the digital camera in both the standard image acquisition system and in the field conditions from the morning to evening. Several kinds of images were acquired with the standard RGB color chart set up just behind the tomato fruit on a black matte, and a color calibration was carried out. The influence of the sunlight could be experimentally eliminated, and the calibrated color information consistently agreed with the standard ones acquired in the system through the post-ripening process. Furthermore, the surface color change of the tomato on the tree in a greenhouse was remotely monitored during maturation using the digital cameras equipped with the Field Server. The acquired digital color images were sent from the Farm Station to the BIFE Laboratory of Mie University via VPN. The time behavior of the tomato surface color change during the maturing process could be measured using the color parameter calculated based on the obtained and calibrated color images along with the ambient atmospheric record. This study is a very important step in developing the surface color analysis for both the simple and rapid evaluation of the crop vigor in the field and to construct an ambient and networked remote monitoring system for food security, precision agriculture, and agricultural research.

Delineating Site-Specific Quality-Based Management Zones for a Tobacco Field

Recent precision agriculture research has focused on the use of management zones (MZ) as a method for variable-rate fertilizer application. A 4-ha tobacco-planted field was selected to define MZ using tobacco quality as basis. To achieve this objective, we collected 111 soil samples from the top 20 cm on an approximately 20-m grid of the study area in March 2009. Classical statistics showed a large variability of all soil nutrients. Results of geostatistics analysis indicated that the eight soil properties examined were best represented through spherical models. Four quality-limiting factors for tobacco were found by stepwise multiple regression analysis. These factors were selected as the variables for delineating MZ. Management Zone Analyst was used to delineate MZ as well as to provide the fuzzy performance index and normalized classification entropy to determine the optimum cluster number. Results showed that the optimum number of MZ in the study area was 4. Furthermore, analysis of variance indicated the heterogeneity of soil variables among the four MZ. Based on the clearly defined MZ, we conclude that tobacco-planted fields can be managed in a site-specific manner by these methods.

Evaluation of Two Ground‐Based Active Crop Canopy Sensors in Maize: Growth Stage, Row Spacing, and Sensor Movement Speed

Precision agriculture research has been directed toward enhancing the efficiency of N inputs by quantifying in‐field variability. Remotely sensed indices such as normalized difference vegetation index (NDVI) can determine in‐field N variability in maize (Zea mays L.). One method of determining NDVI is through the use of ground‐based active crop canopy sensors. Several crop canopy sensors determine NDVI, however, climatic and management variables may affect NDVI readings. Our objectives were to compare two ground‐based active crop canopy sensors (Crop Circle amber and GreenSeeker red) across plant growth stage, wind, crop row spacing, sensor movement speed, and N fertilizer rate under greenhouse conditions. Results show that wind had no effect on the NDVI readings of either sensor. Nitrogen rate and growth stage did affect the NDVI of both sensors with NDVI values generally increasing with increased N rate and advancing growth stage. For both sensors the V8 NDVI r2 with N rate were lower than those observed at V10 and V12. However, the GreenSeeker (red sensor) had much lower r2 values at V8 than the Crop Circle (amber sensor). Sensor speed had an effect on red sensor NDVI values while the amber sensor was not affected by sensor movement speed. The amber and red sensors distinguish plant N status and growth stage differences in maize in a greenhouse environment. However, the red sensor had more variability in NDVI readings and was affected by movement speed. The amber sensor shows no such limitations and therefore performed best under greenhouse conditions.

The arable farmer as the assessor of within-field soil variation

Feasible, fast and reliable methods of mapping within-field variation are required for precision agriculture. Within precision agriculture research much emphasis has been put on technology, whereas the knowledge that farmers have and ways to explore it have received little attention. This research characterizes and examines the spatial knowledge arable farmers have of their fields and explores whether it is a suitable starting point to map the within-field variation of soil properties. A case study was performed in the Hoeksche Waard, the Netherlands, at four arable farms. A combination of semi-structured interviews and fieldwork was used to map spatially explicit knowledge of within-field variation. At each farm, a field was divided into internally homogeneous units as directed by the farmer, the soil of the units was sampled and the data were analysed statistically. The results show that the farmers have considerable spatial knowledge of their fields. Furthermore, they apply this knowledge intuitively during various field management activities such as fertilizer application, soil tillage and herbicide application. The sample data on soil organic matter content, clay content and fertility show that in general the farmers’ knowledge formed a suitable starting point for mapping within-field variation in the soil. Therefore, it should also be considered as an important information source for highly automated precision agriculture systems.

Special issue on 6th European Conference on Precision Agriculture and 2nd Precision Crop Protection Conference

This special issue of the Journal concludes publication of enhanced papers from the 6th European Conference on Precision Agriculture (3 papers) and the 2nd Precision Crop Protection Conference (5 papers) both held in 2007. Some 22 good papers from the former and 10 from the latter have now been published bringing together an important body of research from several disciplines to the attention of the precision agriculture research community. Technology development was a prime driver for precision agriculture in its early phases. With the maturing of the concept, other disciplines (soil science, crop science and many others) have taken their place alongside technology. Technology continues to develop and the developments are being applied to PA as illustrated in the three 6ECPA papers; they describe RFID tags for produce tracing, laser rangefinding applied to field crops and wireless soil sensors. Precision crop protection includes the investigation of weed, insect and fungal pests, the latter two being the topic of 3 of the 2PCPC papers. Technology features in the last two papers, both dealing with precision spray application.

Lessons from nearly 20 years of Precision Agriculture research, development, and adoption as a guide to its appropriate application

Precision Agriculture (PA) is an all-encompassing term given to the use of a suite of technologies that promote improved management of agricultural production through recognition that the potential productivity of agricultural land can vary considerably, even over very short distances (a few m). It can be regarded as a means of increasing the chance that the right crop management strategies are implemented in the right place at the right time. Numerous examples exist of the successful application of PA to various cropping systems around the world, in many cases supported by a burgeoning PA literature. However, the rate of adoption by growers of many crops remains low and, in some industries, is negligible. One such example is the Australian sugar industry, in spite of its relatively high rate of adoption of controlled traffic and the ready access that growers have to supporting infrastructure such as local GPS base stations. However, the Australian sugar industry is now seeking an informed basis from which to make decisions as to appropriate investment in PA, whether these be in terms of pragmatic application by growers, the level of involvement (if any) by millers, or with respect to research to facilitate such adoption. A part of acquiring this informed view of PA is to look at its application in other cropping systems. This review therefore examines PA research and application in a range of cropping systems from around the world and considers the key drivers of variability in these production systems. Constraints to the adoption of PA and its likely economic benefits are also considered in light of experiences from around the world. It is concluded that sugarcane production is ideally suited to the adoption of PA. Like other broadacre systems, such as cereal production, the opportunity exists to target the management of inputs to production. However, the vertically integrated nature of the sugar industry and existence of a potentially significant crop quality imperative also present opportunities for targeted strategies such as selective harvesting, as used in the wine industry. Thus, to get the best result from adoption of PA, the sugar industry will need to consider it as a tool for optimising management of the production of sugar, as opposed to solely an avenue for improving the agronomic management of sugarcane. Several recommendations are made as to how this adoption might be supported.

A synthesis of multi-disciplinary research in precision agriculture: site-specific management zones in the semi-arid western Great Plains of the USA

Researchers from Colorado State University, in collaboration with scientists from the United States Department of Agriculture (USDA), initiated a long-term multi-disciplinary study in precision agriculture in 1997. Site-specific management zones (SSMZ) were investigated as a means of improving nitrogen management in irrigated maize cropping systems. The objective was to develop precise nutrient management strategies for semi-arid irrigated cropping systems. This study was conducted in five fields in northeastern Colorado, USA. Two techniques for delineating management zones were developed and compared: SSMZ and yield-based management zones (YBMZ). Nitrogen uptake and grain yield differences among SSMZs were compared as were soil properties. Both management zone techniques were used to divide fields into smaller units that were different with regard to productivity potential (e.g., high zones had high productivity potential while low zones had low productivity potential). Economic analysis was also performed. Based on grain yield productivity, the SSMZs performed better than the YBMZ technique in most cases. Grain yield and N uptake between the low and high productivity management zones were statistically different for most site-years and N fertilizer rates (p < 0.05). Soil properties helped to explain the productivity potential of the management zones. The low SSMZ was markedly different from the high SSMZ based on bulk density, organic carbon, sand, silt, porosity and soil moisture. Net returns ranged from 188 to 679 USD ha−1. In two out of three site-years the variable yield goal strategy resulted in the largest net returns. In this study, the SSMZ approach delineates areas of different productivity accurately across the agricultural fields. The SSMZs are different with regard to soil properties as well as grain yield and N uptake. Site-specific management zones are an inexpensive and pragmatic approach to precise N management in irrigated maize.

Separating spatial and temporal sources of variation for model testing in precision agriculture

The application of crop simulation models in precision agriculture research appears to require only the specification of some input parameters and then running the model for each desired location in a field. Reports in the extensive literature on modeling have described independent tests for different cultivars, soil types and weather, and these have been presumed to validate the model performance in general. However, most of these tests have evaluated model performance for simulating mean yields for multiple plots in yield trials or in other large-area studies. Precision agriculture requires models to simulate not only the mean, but also the spatial variation in yield. No consensus has emerged about how to test model performance rigorously, or what level of performance is sufficient. In addition, many measures of goodness of fit between the observed and simulated data (i.e., model performance) depend on the range of variation in the observed data. If, for example, inter-annual and spatial sources of variation are combined in a test, poor performance in one might be masked by good performance in the other. Our objectives are to: (1) examine research aims that are common in precision agriculture, (2) discuss expectations of model performance, and (3) compare several traditional and some alternative measures of model performance. These measures of performance are explained with examples that illustrate their limitations and strengths. The risk of relying on a test that combines spatial and temporal data was shown with data where the overall fit was good (r 2 > 0.8), but the fit within any year was zero. Information gained using these methods can both guide and help to build confidence in future modeling efforts in precision agriculture.

Empowering Farmer Leaders to Acquire and Practice Site-Specific Nutrient Management Technology

ABSTRACT Precision agriculture, using high-tech equipment and information science software, is not appropriate for the farmers of small parcels of land in the tropics. Our hypothesis, however, is that the concepts of precision agriculture and participatory action research have a similar philosophical basis and complement each other. We have adopted an approach of both empowering farmers and simplifying nutrient management technology to enable farmers to use the concepts. The simplified technology includes a visual tool to identify soil series, a soil test kit that brings the laboratory to the field, and a decision-aid that enables farmer leaders to interpret the soil test kit data in light of the soil series. The results have demonstrated increased yields with reduced fertilizer inputs over a three-year period in Central Thailand.

Delineation of site-specific management zones using fuzzy clustering analysis in a coastal saline land

Recent research in precision agriculture has focused on use of management zones as a method for variable application of inputs. In this paper, five soil and landscape attributes, including a NDVI image, soil electrical conductivity, total nitrogen, organic matter and cation exchange capacity acquired for a coastal saline land were selected as data sources, and their spatial variability were analyzed and spatial distribution maps constructed with geostatistics technique. Principal component analysis and fuzzy c-means clustering algorithm were then performed to delineate the management zones, fuzzy performance index (FPI) and normalized classification entropy (NCE) was used to determine the optimal cluster number. To assess whether the defined three management zones can be used to characterize spatial variability of soil chemical properties and crop productivity, 139 georeferenced soil and yield sampling points across each management zone was examined by using variance analysis. It was found that the optimal number of management zones for the study area was three and there existed significantly statistical differences between the chemical properties of soil samples and yield data in each defined management zone, management zone 3 presented highest nutrient level and potential crop productivity, whereas management zone 1 lowest. The results revealed that the given five variables could be aggregated into management zones that characterize spatial variability in soil chemical properties and crop productivity. The defined management zones not only can direct soil sampling design, but also provide valuable information for site-specific management in precision agriculture.

Apparent soil electrical conductivity measurements in agriculture

The field-scale application of apparent soil electrical conductivity (EC a) to agriculture has its origin in the measurement of soil salinity, which is an arid-zone problem associated with irrigated agricultural land and with areas having shallow water tables. Apparent soil electrical conductivity is influenced by a combination of physico-chemical properties including soluble salts, clay content and mineralogy, soil water content, bulk density, organic matter, and soil temperature; consequently, measurements of EC a have been used at field scales to map the spatial variation of several edaphic properties: soil salinity, clay content or depth to clay-rich layers, soil water content, the depth of flood deposited sands, and organic matter. In addition, EC a has been used at field scales to determine a variety of anthropogenic properties: leaching fraction, irrigation and drainage patterns, and compaction patterns due to farm machinery. Since its early agricultural use as a means of measuring soil salinity, the agricultural application of EC a has evolved into a widely accepted means of establishing the spatial variability of several soil physico-chemical properties that influence the EC a measurement. Apparent soil electrical conductivity is a quick, reliable, easy-to-take soil measurement that often, but not always, relates to crop yield. For these reasons, the measurement of EC a is among the most frequently used tools in precision agriculture research for the spatio-temporal characterization of edaphic and anthropogenic properties that influence crop yield. It is the objective of this paper to provide a review of the development and use of EC a measurements for agricultural purposes, particularly from a perspective of precision agriculture applications. Background information is presented to provide the reader with (i) an understanding of the basic theories and principles of the EC a measurement, (ii) an overview of various EC a measurement techniques, (iii) applications of EC a measurements in agriculture, particularly site-specific crop management, (iv) guidelines for conducting an EC a survey, and (v) current trends and future developments in the application of EC a to precision agriculture. Unquestionably, EC a is an invaluable agricultural tool that provides spatial information for soil quality assessment and precision agriculture applications including the delineation of site-specific management units. Technologies such as geo-referenced EC a measurement techniques have brought precision agriculture from a 1980's concept to a promising tool for achieving sustainable agriculture.

AN EXAMPLE OF DEVELOPING COVARIATES FOR PROBLEMS IN PRECISION AGRICULTURE

Methodology for precision agriculture is, perhaps, too focused on methods that allow for spatial correlation in the ANOVA error term. While sound inference about differences between local yields can be computed, no understanding of what is driving these differences is achieved. A completely general form for a spatial model can include suitable covariates. Most research in precision agriculture includes gathering a variety of site-specific information. Through the presentation of the analysis of data from a published soybean [Glycine max (L.) Merr.] study, one specific type of covariate is developed - a duration index for soybean canopy light interception over the growing season. The relationship of the index to grain yield is reasonably well determined (R² = 0.82). We, therefore, suggest that the quest for modeling an appropriate covariate or covariates is primary. Treating spatial variation by other methods should only be used when the quest has failed.

A COMPARISON OF GEOSTATISTICAL AND SPATIAL AUTOREGRESSIVE APPROACHES FOR DEALING WITH SPATIALLY CORRELATED RESIDUALS IN REGRESSION ANALYSIS FOR PRECISION AGRICULTURE APPLICATIONS

Regressions such as Grain yield=f(soil,landscape) are frequently reported in precision agriculture research, and are typically computed using conventional OLS methods, implicitly ignoring spatial correlation of the residuals. This oversight can have a marked effect on the final conclusions derived from these regressions. A further issue is, which approach should be used to account for this problem? We investigated this question using a 2 year data set that includes sitespecific soil and topographic information and soybean yields and compare regression results from direct covariance representation and spatial autoregressive approaches. Our results show that the coefficients from both spatial approaches are in many cases significantly different to those from OLS, but the estimates from both spatial approaches appear to show little differences. To provide further insight into the comparison among these approaches we use a simulation of spatial random fields, with a model containing 2 independent explanatory variables and a spatially structured residual term. We then estimated the coefficients for 1000 simulations of this field and assessed their distributional properties. All methods yielded overall unbiased estimates and OLS showed the largest standard errors, while the ‘spatial’ approaches proved to be relatively consistent, although a certain neighborhood specification within the spatial autoregressive model had an evidently lower performance than the rest.

Dynamics of spatial variability of millet growth and yields at three sites in Niger, West Africa and implications for precision agriculture research

This paper focuses on possibilities for using a simple scoring technique to make estimates of millet yield. Estimates are intended to be used for defining application rates of manure in the context of low-tech precision agriculture. Yields from 1995 and 1996 at three locations were related to scoring, soil data and elevation. Kriging was used to interpolate point data to areas. Several procedures for pattern comparison were applied. Because scoring data were available at a much higher density than yield data a sensitivity analysis was made to compare scoring and yield. Correlations between scoring and yield ranged from 0.42 to 0.91. During a separate experiment the optimal time for scoring turned out to be approximately 3 months after seeding for the local varieties of 120 days, but 2 months is convenient for farmers to locally apply chemical fertilizer. R 2 values ranging from 0.15 to 0.60 were observed for soil data and yield, subject to local conditions and changes in weather. We conclude that scoring is a cheap and reliable procedure to identify field patterns which can form the basis for precision agriculture.