Scholarship & Research
Permanent URI for this communityhttps://scholarworks.montana.edu/handle/1/1
Browse
9 results
Search Results
Item Precision organic agriculture(Montana State University - Bozeman, College of Agriculture, 2023) Loewen, Royden Alexander Sasha; Chairperson, Graduate Committee: Bruce D. Maxwell; This is a manuscript style paper that includes co-authored chapters.Organic agriculture addresses some of the shortcomings of industrialized conventional agriculture, but is prevented from more mainstream uptake by reduced yields. Organic agriculture relies on knowledge of intricate biological interactions in place of synthetic inputs used in other forms of agriculture, and in this way reflects an older way of practicing agriculture. Precision agriculture (PA) conversely is a technologically driven method of farming and combines guidance and data collection via remote sensing technologies to bring new efficiencies to farm operations. In this dissertation PA tools were used to explore the potential of improving organic production through site-specific management. By conducting on farm precision experiments (OFPE) with PA farmers can learn quickly about spatial variability across fields enabling well defined management templates. In organic systems this experimentation can be conducted with varied seeding rate inputs of both cover and cash crops. Here, we explored the relevancy of PA in organic settings, first broadly laying the philosophical foundation for the paradigm shift from production-oriented agriculture to precision agroecology. Secondly, a greenhouse experiment was used to develop the first-principle relationship between cover crop and cash crop seeding rates to maximize net return, establishing the basis for field experiments. Field scale experiments on five organic grain farms across the northern great plains deployed OFPE to optimize net returns, or suppress weeds, with varied seeding rates of cover and cash crops. Based on OFPE data, simulations across all sites found net returns could be improved on average by $45.82 ha-1 if economically optimum variable seeding rates were used. While seeding rates were found to have variable effects on weeds across fields, an optimized site-specific seeding strategy to balance net return and weed minimization improved net return and weed suppression compared to farmer-chosen seeding rates in every field tested. Overall, these results reveal the relevancy of precision agriculture to be deployed in organic systems to improve management for increased farmer net returns, and as a weed management method. In this way modern tools can be used to augment farmer knowledge about their local spaces to enable greater understanding and improved management of complex agroecosystems.Item Optimizing site-specific nitrogen fertilizer management based on maximized profit and minimized pollution(Montana State University - Bozeman, College of Agriculture, 2022) Hegedus, Paul Briggs; Chairperson, Graduate Committee: Bruce D. Maxwell and Stephanie A. Ewing (co-chair); This is a manuscript style paper that includes co-authored chapters.Application of nitrogen fertilizers beyond crop needs contributes to nitrate pollution and soil acidification. Excess nitrogen applications are most prevalent when synthetic fertilizers are applied at uniform rates across fields. Precision agroecology harnesses the tools and technology of variable rate precision agriculture, a common but underutilized management strategy, to make ecologically conscious decisions about field management that promote economic and environmental sustainability. On-farm precision experimentation provides the basis for making data driven ecological management decisions through the field-specific assessment of crop responses. This dissertation work used on-farm experimentation with variable nitrogen fertilizer rates, combined with intensive data collection and data science, to address the main objective of this dissertation: development and evaluation of optimized nitrogen fertilizer management on a subfield scale, based on maximization of farmer net-returns and nitrogen use efficiency. The response of winter wheat yield and grain protein concentration to rates of nitrogen fertilizer application varied among fields, and across time, which influenced the model form used to characterize the relationships of grain yield and quality to fertilizer within a field. Machine learning approaches, such as random forest regression, tended to provide the lowest degree of error when forecasting future crop responses. Machine learning also demonstrated its utility for use in agronomic applications, as a support vector regression model provided the most accurate predictions of nitrogen use efficiency on a subfield scale. Crop response and nitrogen use efficiency models were integrated into a decision-making framework for optimized site-specific based nitrogen fertilizer management based on between maximized profits and minimized potential of nitrogen loss. Simulations of optimized site-specific nitrogen fertilizer management compared to farmer's status quo management showed a 100% probability across all fields tested that that mean net-return from the site-specific approaches were more profitable than applications of farmer selected nitrogen fertilizer rates. However, even while considering minimization of the potential for nitrogen loss when identifying optimum nitrogen fertilizer rates, there was field specific variation in the probability that site-specific, compared to farmer selected, nitrogen fertilizer management reduced the total amount of nitrogen applied across a field.Item Improving a precision agriculture on-farm experimentation workflow through machine learning(Montana State University - Bozeman, College of Engineering, 2019) Peerlinck, Amy; Chairperson, Graduate Committee: John SheppardReducing environmental impact while simultaneously improving net return of crops is one of the key goals of Precision Agriculture (PA). To this end, an on-farm experimentation workflow was created that focuses on reducing the applied nitrogen (N) rate through variable rate application (VRA). The first step in the process, after gathering initial data from the farmers, creates experimental randomly stratified N prescription maps. One of the main concerns that arises for farmers within these maps is the large jumps in N rate between consecutive cells. To this end we successfully develop and apply a Genetic Algorithm to minimize rate jumps while maintaining stratification across yield and protein bins. The ultimate goal of the on-farm experiments is to determine the final N rate to be applied. This is accomplished by optimizing a net return function based on yield and protein prediction. Currently, these predictions are often done with simple linear and non-linear regression models. Our work introduces six different machine learning (ML) models for improving this task: a single layer feed-forward neural network (FFNN), a stacked auto-encoder (SAE), three different AdaBoost ensembles, and a bagging ensemble. The AdaBoost and bagging methods each use a single layer FFNN as its weak model. Furthermore, a simple spatial analysis is performed to create spatial data sets, to better represent the inherent spatial nature of the field data. These methods are applied to four actual fields' yield and protein data. The spatial data is shown to improve accuracy for most yield models. It does not perform as well on the protein data, possibly due to the small size of these data sets, resulting in a sparse data set and potential overfitting of the models. When comparing the predictive models, the deep network performed better than the shallow network, and the ensemble methods outperformed both the SAE and a single FFNN. Out of the four different ensemble methods, bagging had the most consistent performance across the yield and protein data sets. Overall, spatial bagging using FFNNs as the weak learner has the best performance for both yield and protein prediction.Item Resilience of Montana's agroecosystems to economic and climatic change(Montana State University - Bozeman, College of Agriculture, 2015) Lawrence, Patrick Glenn; Chairperson, Graduate Committee: Bruce D. Maxwell; Bruce D. Maxwell and Lisa J. Rew were co-authors of the article, 'A probabilistic bayesian framework for progressively updating site-specific recommendations' in the journal 'Precision agriculture' which is contained within this thesis.; Bruce D. Maxwell, Lisa J. Rew, Anton Bekkerman, Clain Jones and Perry Miller were co-authors of the article, 'Managing uncertainty in semiarid dryland agriculture: a data-driven approach to optimize inputs and crop rotations based on farmer risk preferences' submitted to the journal 'Agricultural systems' which is contained within this thesis.; Bruce D. Maxwell, Lisa J. Rew, Colter Ellis and Anton Bekkerman were co-authors of the article, 'Vulnerability of dryland agricultural regimes to economic and climatic change' submitted to the journal 'Climatic change' which is contained within this thesis.Semiarid dryland agricultural systems in the western United States are faced with a highly uncertain production environment that complicates decision-making and makes static agronomic prescriptions unreliable for maintaining sustainability. The primary sources of uncertainty for farmers are weather, fluctuations in prices, and site-specific environmental and ecological variability, some of which may be amplified by climate change. To effectively respond to the risks posed by these uncertainties requires knowledge of the vulnerability of these agricultural systems. The aim of this dissertation was to meet this need for Montana by analyzing the economic resilience of the state dryland agricultural systems at site-specific and county-wide scales. To begin, a framework was created to integrate weather, prices, nitrogen inputs, and spatial soil variability within a statistical model for site-specific crop responses and net returns. Simulations suggest that six crop years of simulated data collection and parameter tuning were required to derive an accurate model, suggesting that an extended period of observation and targeted nitrogen rate experimentation was required to optimize spatial fertilizer management. The framework was subsequently applied to a spatiotemporal precision agricultural dataset from a farm near Great Falls, MT, and was modified to account for several crop rotations and different farmer risk preferences. Regardless of farmers' level of risk aversion, winter wheat-pea rotations resulted in higher value (utility) for the farmer than winter wheat-fallow and continuous winter wheat rotations. For most levels of risk adversity, it was also optimal to apply no nitrogen fertilizer. Net returns at the field site were always threatened by drought. Subsequently, a qualitative analysis of farmer adaptability in Montana based on survey and interview data determined that farmers had few options for responding to drought but were more adaptable to high input prices. On-farm experimentation and crop rotations could greatly increase adaptability in the future. Finally, simulations of alternative price, precipitation, and crop rotation scenarios were completed. The most resilient agricultural systems were located in northeastern Montana where pulses have been more widely adopted; systems in north-central Montana were less resilient. State-wide, over 50% of dryland farmers may not be resilient to future economic or climatic variability.Item An industry needs assessment of competencies in precision agriculture(Montana State University - Bozeman, College of Agriculture, 1999) Perleberg, Rick A.Item Terrain analysis in support of precision farming(Montana State University - Bozeman, College of Letters & Science, 1995) Spangrud, Damian JeremiahItem Prototype of 3D visualization tool for precision agriculture analysis(Montana State University - Bozeman, College of Engineering, 2003) Sanchez, Paula DoloresItem Wheat yield prediction modeling for localized optimization of fertilizer and herbicide application(Montana State University - Bozeman, College of Agriculture, 2004) Wagner, Nicole Catherine; Chairperson, Graduate Committee: Bruce D. Maxwell.The specific goal of this thesis was the development of a five-variable dryland wheat yield prediction model for the optimal localized variable-rate management of fertilizer and herbicide considering varying levels of available water and weed infestation. The motivation for this work was to increase on-farm net return and reduce off-target chemical effects. The five most influential predictor variables of wheat yield were investigated: wheat density, wild oat density, nitrogen fertilizer rate, herbicide rate, and water level Previously collected field data sets that included dryland wheat yield as the dependent variable and at least one of the predictors were investigated for linear and nonlinear trends. The best-fit nonlinear yield model to the combined field data set included crop and wild oat density, and growing season precipitation; nitrogen and herbicide rates were not significant factors in this model. These results illustrated the large amount of unexplored variation in wheat yield, and the lack of ecological first principles upon which farmers base input management decisions, especially when weed infestation causes competition for limited nitrogen and water. To take an initial step towards elucidating the biological mechanisms of wheatwild oat competition with varying combinatorial levels of resources, a five-variable greenhouse experiment was conducted. The best-fitting yield model to the greenhouse data set was a nonlinear equation including all five variables. This predictive model was used to demonstrate how such an equation would help farmers make localized variable rate input decisions within a decision support framework. Monte Carlo simulation was used to produce net return prediction probabilities for site-specific variable-rate management, low level input management, and high level input management of nitrogen and herbicide based on the two sources of parameter estimates-field data and greenhouse data. The variable rate scenario resulted in larger net returns over the broadcast management scenarios in at least 48%, and at most 66%, of the simulations. This initial exploration provided considerable support for future on-farm experiments and yield prediction modeling. In addition, it established a first principle model to be parameterized for use in different dryland spring wheat growing regions.Item Modeling soil water content for precision range management(Montana State University - Bozeman, College of Agriculture, 2005) Sankey, Joel Brown; Chairperson, Graduate Committee: Rick Lawrence.I developed site-specific empirical models to predict spring soil water content for two Montana ranches. The models used publicly available Landsat TM 5, USGS DEM, and soil survey-derived data as predictor variables. The goal of the project was to test whether ranchers could collect a limited size soil water content data set, build sitespecific regression models based on the data set, and construct soil water content maps based on the models. The response variable for models consisted of 100 and 82 average soil profile mass water content samples for each ranch, respectively. Half the samples were used for model calibration and half for model validation. Multiple regression models had calibration R2 of 0.64 and 0.43 for each ranch, respectively. Validation showed that the multiple regression models predicted the validation data sets with average error (RMSD) within 0.04 mass water content and regression tree models predicted within 0.055 mass water content. The majority of the validation MSD for all models was accounted for by a lack of correlation between predicted and observed values along a 1:1 line. Models were then constructed with decreased sample sizes. Regression tree models and multiple regression models constructed with 20 to 30 samples predicted soil water content with similar, though still limited, accuracy and precision to full sample models. Site-specific field and lab soil characterization data developed with diffuse reflectance spectroscopy modeling was used to assess the suitability of the soil survey based predictor variables. The multiple regression models with the site-specific data predicted soil water content with average prediction errors (RMSD) of 0.035 and 0.036 mass water content for the two ranches, respectively. Soil survey model predictions were statistically significantly different than site-specific model predictions for one ranch but not the other. Especially dry conditions were a factor contributing to the difficulty in accurately modeling and predicting soil water content encountered at both study sites. Landsat imagery from the peak of the previous growing season, DEM-derived slope and aspect variables, and soil survey attribute data each showed promise as significant predictors of spring soil water content, particularly considering the dry conditions of the data collection period.