Stuart J. Birrell

Statistical and Neural Methods for Site-Specific Yield Prediction

Understanding the relationships between yield and soil properties and topographic characteristics is of critical importance in precision agriculture. A necessary first step is to identify techniques to reliably quantify the relationships between soil and topographic characteristics and crop yield. Stepwise multiple linear regression (SMLR), projection pursuit regression (PPR), and several types of supervised feed-forward neural networks were investigated in an attempt to identify methods able to relate soil properties and grain yields on a point-by-point basis within ten individual site-years. To avoid overfitting, evaluations were based on predictive ability using a 5-fold cross-validation technique. The neural techniques consistently outperformed both SMLR and PPR and provided minimal prediction errors in every site-year. However, in site-years with relatively fewer observations and in site-years where a single, overriding factor was not apparent, the improvements achieved by neural networks over both SMLR and PPR were small. A second phase of the experiment involved estimation of crop yield across multiple site-years by including climatological data. The ten site-years of data were appended with climatological variables, and prediction errors were computed. The results showed that significant overfitting had occurred and indicated that a much larger number of climatologically unique site-years would be required in this type of analysis.

Real-time multi ISFET/FIA soil analysis system with automatic sample extraction

Successful implementation of site-specific crop management relies on accurate quantification of spatial variation of important factors. Therefore, there is a tremendous need for the development of sensing technologies that will allow automated collection of soil, crop and pest data, to more accurately characterize within-field variability. The objective of this work was to develop an integrated multi-sensor soil analysis system. Ion-selective field effect transistor (ISFET) technology was coupled with flow injection analysis (FIA) to produce a real-time soil analysis system. Testing of the ISFET/ FIA system for soil analysis was carried out in two stages: (1) using manually extracted samples, and (2) the soil to be analysed was placed in the automated soil extraction system, and the extracted solution fed directly into the FIA system. The sensor was successful in measuring soil nitrates in manually extracted soil solutions (r2>0.9). The rapid response of the system allowed a sample to be analysed in 1.25 s, which is satisfactory for real-time soil sensing. Precision and accuracy of the system were highly dependent on maintaining precise, repetitive injection times and maintaining constant flow parameters during the calibration and testing cycle. The progress toward an automated soil extraction system was notable, but considerable effort will be necessary before commercialization can be realized. However, the concept of using ISFETs for the real-time analysis of soil nitrates is sound. The rapid response and low sample volumes required by the multi-sensor ISFET/FIA system make it a viable candidate for use in real-time soil nutrient sensing.

Membrane Selection and ISFET Configuration Evaluation for Soil Nitrate Sensing

Successful implementation of site-specific crop management relies on accurate quantification of spatial variation of important factors. Data collection on a finer spatial resolution than is feasible with manual and/or laboratory methods is often required but cost prohibitive. Therefore, there is a need for the development of sensors to more accurately characterize within-field variability. The objective of this research was to investigate matrix membranes produced from different combinations of ligand and plasticizer materials using ion-selective electrode (ISE) technology, and to use selected membranes to develop a nitrate ion-selective field effect transistor (ISFET) which might be integrated with a flow injection analysis (FIA) system for real-time soil analysis. Several ion-selective membranes were tested, and all of the evaluated membranes proved to be viable candidates for the development of a nitrate ISFET. Membranes using methyltridodecylammonium chloride (MTDA) as the ligand showed a better response to nitrates at low concentrations while those using tetradodecylammonium nitrate (TDDA) ligand showed superior selectivity for the nitrate ion. A multi-ISFET nitrate sensor was successfully developed. The electrical responses of the ISFETs were consistent and predictable. While significant difficulty was found in preparing a multi-ISFET chip with all four sensors operational, once prepared, the multi-ISFET chips were reliable and performed through extensive tests without failure. The sensitivities of the nitrate ISFETs (38-46 mV/decade) were lower than the theoretical Nernst sensitivity. The nitrate ISFETs proved to be viable sensors for the development of a real-time soil nitrate analysis system, under the conditions of our tests.

EVALUATION OF NITRATE AND POTASSIUM ION-SELECTIVE MEMBRANES FOR SOIL MACRONUTRIENT SENSING

On-the-go, real-time soil nutrient analysis would be useful in site-specific management of soil fertility. The rapid response and low sample volume associated with ion-selective field-effect transistors (ISFETs) make them good soil fertility sensor candidates. Ion-selective microelectrode technology requires an ion-selective membrane that responds selectively to one analyte in the presence of other ions in a solution. This article describes: (1) the evaluation of nitrate and potassium ion-selective membranes, and (2) the investigation of the interaction between the ion-selective membranes and soil extractants to identify membranes and extracting solutions that are compatible for use with a real-time ISFET sensor to measure nitrate and potassium ions in soil. The responses of the nitrate membranes with tetradodecylammonium nitrate (TDDA) or methlytridodecylammonium chloride (MTDA) and potassium membranes with valinomycin were affected by both membrane type and soil extractant. A TDDA-based nitrate membrane would be capable of detecting low concentrations in soils to about 10 −5 mole/L NO3 − . The valinomycin-based potassium membranes showed satisfactory selectivity performance in measuring potassium in the presence of interfering cations such as Na + , Mg 2+ , Ca 2+ , Al 3+ , and Li + as well as provided a consistent sensitivity when DI water, Kelowna, or Bray P1 solutions were used as base solutions. The TDDA-based nitrate membrane and the valinomycin-based potassium membrane, used in conjunction with Kelowna extractant, would allow determination of nitrate and potassium levels, respectively, for site-specific control of fertilizer application.

RAPID NITRATE ANALYSIS OF SOIL CORES USING ISFETS

An intact core extraction procedure was tested that might be used in the field for real–time prediction of soil nitrates. An extraction solution was pushed through a soil core held between two filters, and an ion–selective field–effect transistor/flow injection analysis (ISFET/FIA) system was used to sense soil nitrates in real time. Laboratory tests were conducted using four soil types and two levels of nitrate concentration, soil moisture, core density, core length, core diameter, and extraction solution flow rate. The extraction solution flow was sampled at the exit face of the core and routed to the ISFET/FIA system. The ISFET output voltage was sampled at 100 Hz. Results of the test indicate that nitrate extraction of the soil cores was successful, and that data descriptors based on response curve peak and slope of the ISFET nitrate response curve might be used in tandem in a real–time prediction system.

Evaluation of Phosphate Ion-Selective Membranes and Cobalt-Based Electrodes for Soil Nutrient Sensing

A real-time soil nutrient sensor would allow efficient collection of data with a fine spatial resolution to accurately characterize within-field variability for site-specific nutrient application. Ion-selective electrodes are promising candidates because they have rapid response, directly measure the analyte, and are small and portable. Our goal was to investigate the ability of three different phosphate ion-selective electrodes (two fabricated with organotin compound-based PVC membranes, and one fabricated from a cobalt rod) used in conjunction with Kelowna soil extractant to determine phosphorus over the typical range of soil concentrations. Electrodes using organotin compound-based PVC membranes containing bis(p-chlorobenzyl)tin dichloride as an ionophore exhibited sensitive responses to HPO42- over a range of 10-4 to 10-1 mol/L in Tris buffer at pH 7. They were nearly insensitive to phosphate when using Kelowna soil extractant as the base solution, perhaps because of the high concentration of fluoride (0.015 mol/L) in the Kelowna solution. In addition, the life of the membranes was less than 14 days. Electrodes using another tin-compound-based PVC membrane containing tributyltin chloride as an ionophore also provided unsatisfactory results, showing much less sensitivity to H2PO4- than previously reported. The cobalt rod-based electrodes exhibited sensitive responses to H2PO4- over a range from 10-5 to 10-1 mol/L total phosphate concentration with a detection limit of 10-5 mol/L in the Kelowna solution. This detection range would encompass the typical range of soil phosphorus concentrations measured in agricultural fields. The selectivity of the cobalt electrodes was satisfactory for measuring phosphates in the presence of each of six interfering ions, i.e., HCO3-, Cl -, Br -, NO3-, Ac -, and F -, with the electrodes being 47 to 1072 times more responsive to phosphate than to the tested interfering ions.

Evaluation of GPS for Applications in Precision Agriculture

Location coordinate information is needed in precision agriculture to map in-field variability, and to serve as a control input for variable rate application. Differential global positioning system (DGPS) measurement techniques were compared with other independent data sources for sample point location and combine yield mapping operations. Sample point location could be determined to within 1 m (3 ft) 2dRMS using C/A code processing techniques and data from a high-performance GPS receiver. Higher accuracies could be obtained with carrier phase kinematic positioning methods, but this required more time and was a less robust technique with a greater potential for data acquisition problems. Data from a DGPS C/A code receiver was accurate enough to provide combine position information in yield mapping. However, distance data from another source, such as a ground-speed radar or shaft speed sensor, was needed to provide sufficient accuracy in the travel distance measurements used to calculate yield on an area basis.

Drought effects on composition and yield for corn stover, mixed grasses, and Miscanthus as bioenergy feedstocks

Drought conditions in 2012 were some of the most severe in recent history. The purpose of this study is to examine the impact of drought on quality, quantity, and theoretical ethanol yield (TEY) of three bioenergy feedstocks, corn stover, mixed grasses from Conservation Reserve Program lands, and Miscanthus × giganteus. To assess drought effects on these feedstocks, samples from 2010 (minimal to no drought) and 2012 (severe drought) were compared from multiple locations in the US. In all feedstocks, drought significantly increased extractives and reduced structural sugars and lignin; subsequently, TEYs were reduced 10–15%. Biomass yields were significantly reduced for M. × giganteus and mixed grasses. When reduction in quality and quantity were combined, TEYs decreased 26–59%. Drought negatively affected biomass quality and quantity that resulted in significant TEY reductions. Such fluctuations in biomass quality and yield may have significant consequences for developing lignocellulosic biorefineries.

Midwest vision for sustainable fuel production.

This article charts the progress of CenUSA Bioenergy, a USDA-NIFA-AFRI coordinated agricultural project focused on the North Central region of the US. CenUSA’s vision is to develop a regional system for producing fuels and other products from perennial grass crops grown on marginally productive land or land that is otherwise unsuitable for annual cropping. This article focuses on contributions CenUSA has made to nine primary systems needed to make this vision a reality: feedstock improvement; feedstock production on marginal land; feedstock logistics; modeling system performance; feedstock conversion into biofuels and other products; marketing; health and safety; education, and outreach. The final section, Future Perspectives, sets forth a roadmap of additional research, technology development and education required to realize commercialization.

Ultrasonic Sensing for Corn Plant Canopy Characterization

Non-destructive measurement of crop growth stage, canopy development, and height may be useful for more efficient crop management practices. In this study, ultrasonic sensing technology was investigated as one approach for corn plant canopy characterization. Ultrasonic echo signals from corn plant canopies were collected using a lab-based sensor platform. Echo signal peak features were extracted from multiple scans of plant canopies. These features included peak amplitude, scan number, and time of flight. Feature vectors with similarities were clustered together to identify individual leaves of the canopy. The mean height of the clustered data of individual leaves was estimated. The growth stage of each plant was estimated based on the number of leaves detected. Regression analysis was used to describe the relationship between manually measured leaf heights and ultrasonic estimates. A leaf-signal interaction model was developed to predict which