This thesis examines the use of two specific types of precision agriculture technologies: cost maps and lightbar. Cost maps visually depict spatial differences in production costs. The visual depictions of these costs are represented using ArcGIS in an attempt to aide farmers in further decision making. Results will show that cost maps have great possibilities in their addition to the set of tools that farmers use in decision making. This thesis will expand the understanding of lightbar from a partial budget study to a whole farm model incorporating competition across different enterprises for labor and capital. The results from the study of cost maps indicate that inaccuracy of machinery movement, whether in the application stage or the harvesting stage is very costly. As a result, the suggestion of lightbar as a guidance aide to improve farm profitability is recommended under the conditions analyzed and shows a net farm return increase in just over 6%.

At no point in history have we demanded so much from our agricultural land whilst simultaneously leaving so little room for management error. Of the many possible environmental impacts from agriculture, soil and water salinisation has some of the most long-lived and deleterious effects. Despite its importance, however, land managers are often unable to make informed decisions of how to manage the risk of salinisation due to a lack of data. Furthermore, there remains no universally agreed method for salinity risk mapping. This thesis addresses these issues by investigating new methods for producing high-resolution predictions of soil salinity, soil physical properties and groundwater depth using a variety of traditional and emerging ancillary data sources. The results show that the methodologies produce accurate predictions yielding natural resource information at a scale and resolution not previously possible. Further to this, a new methodology using fuzzy logic is developed that exploits this information to produce high-resolution salinity risk maps designed to aid both agricultural and natural resource management decisions. The methodology developed represents a new and effective way of presenting salinity risk and has numerous advantages over conventional risk models. The incorporation of fuzzy logic provides a meaningful continuum of salinity risk and allows for the incorporation of uncertainty. The method also allows salinity risk to be calculated relative to any vegetation community and shows where the risk is coming from (root-zone or groundwater) allowing more appropriate management decisions to be made. The development of this methodology takes us a step closer to closing what some have called our greatest gap in agricultural knowledge. That is, our ability to manage the salinity risk at the subcatchment scale.

At no point in history have we demanded so much from our agricultural land whilst simultaneously leaving so little room for management error. Of the many possible environmental impacts from agriculture, soil and water salinisation has some of the most long-lived and deleterious effects. Despite its importance, however, land managers are often unable to make informed decisions of how to manage the risk of salinisation due to a lack of data. Furthermore, there remains no universally agreed method for salinity risk mapping. This thesis addresses these issues by investigating new methods for producing high-resolution predictions of soil salinity, soil physical properties and groundwater depth using a variety of traditional and emerging ancillary data sources. The results show that the methodologies produce accurate predictions yielding natural resource information at a scale and resolution not previously possible. Further to this, a new methodology using fuzzy logic is developed that exploits this information to produce high-resolution salinity risk maps designed to aid both agricultural and natural resource management decisions. The methodology developed represents a new and effective way of presenting salinity risk and has numerous advantages over conventional risk models. The incorporation of fuzzy logic provides a meaningful continuum of salinity risk and allows for the incorporation of uncertainty. The method also allows salinity risk to be calculated relative to any vegetation community and shows where the risk is coming from (root-zone or groundwater) allowing more appropriate management decisions to be made. The development of this methodology takes us a step closer to closing what some have called our greatest gap in agricultural knowledge. That is, our ability to manage the salinity risk at the subcatchment scale.

At no point in history have we demanded so much from our agricultural land whilst simultaneously leaving so little room for management error. Of the many possible environmental impacts from agriculture, soil and water salinisation has some of the most long-lived and deleterious effects. Despite its importance, however, land managers are often unable to make informed decisions of how to manage the risk of salinisation due to a lack of data. Furthermore, there remains no universally agreed method for salinity risk mapping. This thesis addresses these issues by investigating new methods for producing high-resolution predictions of soil salinity, soil physical properties and groundwater depth using a variety of traditional and emerging ancillary data sources. The results show that the methodologies produce accurate predictions yielding natural resource information at a scale and resolution not previously possible. Further to this, a new methodology using fuzzy logic is developed that exploits this information to produce high-resolution salinity risk maps designed to aid both agricultural and natural resource management decisions. The methodology developed represents a new and effective way of presenting salinity risk and has numerous advantages over conventional risk models. The incorporation of fuzzy logic provides a meaningful continuum of salinity risk and allows for the incorporation of uncertainty. The method also allows salinity risk to be calculated relative to any vegetation community and shows where the risk is coming from (root-zone or groundwater) allowing more appropriate management decisions to be made. The development of this methodology takes us a step closer to closing what some have called our greatest gap in agricultural knowledge. That is, our ability to manage the salinity risk at the subcatchment scale.

ROCHA JÚNIOR, Odílio Coimbra. Desenvolvimento de um aplicativo para o manejo de irrigação utilizando a ferramenta “App inventor” na plataforma “Android”. 2015. 67 f. Dissertação (Mestrado em engenharia agrícola)- Universidade Federal do Ceará, Fortaleza-CE, 2015. / Submitted by Elineudson Ribeiro (elineudsonr@gmail.com) on 2016-06-28T19:44:32Z No. of bitstreams: 1 2015_dis_ocrochajunior.pdf: 3121689 bytes, checksum: e4efd8e1e745a4bd1938851b481c80e4 (MD5) / Approved for entry into archive by José Jairo Viana de Sousa (jairo@ufc.br) on 2016-06-30T22:47:34Z (GMT) No. of bitstreams: 1 2015_dis_ocrochajunior.pdf: 3121689 bytes, checksum: e4efd8e1e745a4bd1938851b481c80e4 (MD5) / Made available in DSpace on 2016-06-30T22:47:34Z (GMT). No. of bitstreams: 1 2015_dis_ocrochajunior.pdf: 3121689 bytes, checksum: e4efd8e1e745a4bd1938851b481c80e4 (MD5) Previous issue date: 2015 / The lack of a proper irrigation management by the majority of producers in irrigated public areas in the state of Ceará is considered the main factor with direct influence on productivity, cost of production, quality of products, and especially in the environmental issue, where the inappropriate use of water from our springs could seriously compromise our future generations. The initiative in the development of tools to assist the producer to carry out a satisfactory irrigation management that is accessible and of easy operation is a goal that should always be set as a priority. This work aims to develop an application for mobile devices, able to use its own database in the realization of the daily irrigation time calculation for each crop on a certain date. Techniques and software development tools were used specificly for the Android mobile platform. The application intends to assist in conducting a management with rational use of water in irrigated areas in the state of Ceará, semi-arid northeast, mitigating the lack of technical assistance specialized in precision farming, using databases with information necessary for the calculation of the need water and subsequent irrigation time recommendation. Among the installed databases, it has historical series of "potential reference evapotranspiration" for the past ten years in representative regions where the eight main irrigation schemes of the Ceará state: Curu-Paraipaba, Pentecostes, Baixo Acaraú, Araras Norte, Jaguaribe Apodi, Morada Nova, Tabuleiros de Russas e Icó-Lima Campos are located and also databases with information on the "crop coefficient" and "coverage percentage" of major crops exploited on the perimeters. This application was developed to present a new approach to irrigation management, making use of the available mobile phones in order to allow users to perform the calculation of the daily irrigation time in any of the irrigated perimeters of the Ceará state. The use of the application in simulation for the Irrigated Perimeter Baixo Acaraú demonstrated a reduction of approximately 26% over the current use of water in irrigation, allowing its use be concluded that provides good irrigation management and could result in an economy of water required to reduce the impact of recurring long and repetitive periods of drought on water resources of the state of Ceará. / A falta de um manejo adequado da irrigação por parte da maioria dos produtores instalados nos perímetros públicos irrigados no estado do Ceará é considerado como o principal fator com interferência direta na produtividade, custo de produção, qualidade dos produtos comercializados e principalmente na questão ambiental, onde o uso inadequado da água de nossos mananciais poderá comprometer seriamente nossas gerações futuras. A iniciativa de desenvolvimento de ferramentas para auxiliar o produtor a realizar um manejo de irrigação satisfatório, que seja acessível e de fácil operação é uma meta que devemos sempre definir como prioritária. Esse trabalho visa o desenvolvimento de um aplicativo para dispositivos móveis, capaz de utilizar uma base de dados própria na realização do cálculo do tempo diário de irrigação para cada cultura em uma determinada data. Foram utilizadas técnicas e ferramentas de desenvolvimento de software, específicas para a plataforma móvel Android. O aplicativo visa auxiliar na condução de um manejo com uso racional de água nos perímetros irrigados localizados no estado de Ceará, semiárido nordestino, mitigando a falta da assistência técnica especializada em agricultura de precisão, utilizando bancos de dados com informações necessárias para o cálculo da necessidade hídrica e posterior recomendação do tempo de irrigação. Entre os bancos de dados instalados, conta-se com as series históricas da “evapotranspiração potencial de referência” dos últimos dez anos nas regiões representativas onde se encontram implantados os oito principais perímetros irrigados do estado do Ceará: Curu-Paraipaba, Pentecostes, Baixo Acaraú, Araras Norte, Jaguaribe Apodi, Morada Nova, Tabuleiros de Russas e Icó-Lima Campos, além dos bancos de dados com informações relativas ao “coeficiente de cultivo” e “Percentual de cobertura” dos principais cultivos explorados nos perímetros. O sistema desenvolvido apresenta uma nova abordagem para o manejo da irrigação, fazendo uso da mobilidade disponível em celulares para permitir que os usuários realizem o cálculo do tempo de irrigação diário em qualquer um dos perímetros irrigados do estado do Ceará. A utilização do aplicativo em simulação para o Perímetro Irrigado Baixo Acaraú, demonstrou uma redução de aproximadamente 26 % em relação a utilização atual de água na irrigação, permitindo concluir-se que o seu uso proporciona um bom manejo de irrigação e poderá resultar em uma economia de água necessária para redução dos impactos recorrentes dos longos e repetitivos períodos de estiagem sobre os recursos hídricos do estado do Ceará.

Made available in DSpace on 2014-06-11T19:32:18Z (GMT). No. of bitstreams: 0 Previous issue date: 2008-09-05Bitstream added on 2014-06-13T19:21:49Z : No. of bitstreams: 1 nascimento_gf_dr_rcla.pdf: 1247848 bytes, checksum: 89c8f2950afff4ac2857fc4e72433047 (MD5) / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / O presente trabalho teve por objetivo verificar a variabilidade espacial de atributos químicos do solo, referentes à macronutrientes utilizando métodos estatísticos e geoestatísticos, em uma área considerada homogênea, contendo campo com pastagem e área de floresta. A área escolhida foi a da Fazenda Mata Verde, situada no município de Candeias do Jamari – RO, a cerca de 30 km da capital. Na fase inicial da pesquisa foi elaborado um plano amostral, a partir de uma préamostragem georreferenciada, que resultou na determinação do tamanho da amostra para cada uma das áreas estudadas, cuja opção foi coletar 130 amostras no perfil 00,20 m distribuídas em malha quadrada para cada uma das áreas. Os dados foram analisados em duas fases: análise química e análise estatística. A análise química das amostras foi realizada para identificar os teores de alumínio (Al), cálcio (Ca), matéria orgânica (MO) e potencial de hidrogênio (pH). Para a realização da análise quantitativa, os valores resultantes da análise química foram armazenados em planilhas eletrônicas com os respectivos posicionamentos geográficos. Na análise descritiva e inferências estatísticas clássicas, foram efetuados os cálculos da média, variância, desvio padrão, coeficiente de variação, coeficiente de assimetria e coeficiente de curtose, dos dados amostrados. Foram ajustados modelos teóricos de variogramas aos dados experimentais e realizado o teste de validação cruzada, bem como a construção de mapas de atributos de solo, fazendo uso de krigagem ordinária. A partir dos resultados obtidos, foi possível concluir que a ocorrência de disparidades nos mapas dos atributos Al, Ca, MO e pH, nas áreas estudadas, é um indicativo da necessidade de adoção de metodologias diferenciadas para uso da área de pastagem, considerando o manejo adotado pelo pecuarista. / The purpose of this work is to verify spatial variability of soil’s macro nutrients chemical attributes using statistical and geostatistical tools, in an area considered homogeneous containing pasture and forest. The chosen area was the Mata Verde Farm situated in Candeias do Jamari City – Rondonia State, distant 30 km from the State capital. In the initial stage a sampling plan was elaborated based upon a georeferencing presampling. This resulted in sample sizes for each one of studied areas. The chosen option was the collect of 130 samples in soil profile 00.20 m grid distributed for each one of the areas. The data was analyzed in two phases: chemical analysis and statistical analysis. The samples’ chemicals analysis were done to identify macronutrients contents Al, Ca, organic matter (MO) and pH and the values of the chemical analysis were recorded in spreadsheets with respective geographic positions. To descriptive analysis and classic statistical inferences were carried out calculating means, variance, standard deviation, variation coefficient, asymmetry coefficient and kurtosis coefficient of sampled data. The theoretic model of variograms were adjusted to experimental data and carried out a cross validation test, as well as the maps construction of soil attributes using ordinary kriging. From results was possible to conclude that the occurrence of disparities in spatial distribution maps of aluminum, calcium, organic matter and pH in studied areas is an indicative of necessity of adoption of differentiated methodologies for use of the pasture area, considering the management adopted by the cattle farmers.

Master of Agribusiness / Department of Agricultural Economics / Vincent R. Amanor-Boadu / As the integration of technology and data rises in production agriculture, John Deere dealers in North America are in a constant quest to differentiate themselves and be more than just an equipment provider. Customers with more technologically advanced products are requiring more support from the dealerships. Each dealership has a unique opportunity to provide unprecedented levels of support and each may do it in a slightly different way. This creates a challenge for Deere & Company in providing resources and support to those dealers in their endeavors. This thesis was requested by Deere & Company (John Deere) to provide the company with information on Dealer Customer Support Centers in North America. In order to provide resources and tools for dealers to be successful, it is necessary to understand what they are currently doing with customer support centers and the barriers to implementing more. An online survey was sent out to the Integrated Solutions Manager at every John Deere Dealer organization in North America. From that survey there were a total of 127 responses. The two most common forms of customer support systems that dealers are using are having Integrated Solutions Staff members take calls directly from customers and having someone within the dealership answer the phone and manually route the call to the right person for support. Data also shows that some of the less common but more technologically advanced methods of support have been implemented more in the past 12 months. Survey analysis indicates that only a small percentage of dealers have a true centralized dedicated support center for customers. This subset of dealers is utilizing several different methods to support customers. The two indicative methods are having a 1- 800 number for customers to utilize for support and having a dedicated staff to help customers remotely. Dealers are frequently using several types of tools and resources to help support customers, the most frequently occurring ones include JDLink™ and Data Management Services. Respondents indicated using many other tools to provide value to customers including John Deere Remote Display Access, clinics and optimization sessions and many others. Barriers to implementing more complex forms of customer support are numerous, the biggest of which is the cost of implementation and lack of resources to support a more sophisticated customer support system. With this information, John Deere is better positioned to provide resources and support to our dealer channel facing these challenges.

Master of Science / Department of Agronomy / Antonio R. Asebedo / Sensor technology has become more important in precision agriculture, by real time sensing for site specific management to monitor crops during the season especially nitrogen (N). In Kansas N available in the soils can vary year to year or over a course of a year. The objective of this study was to compare current available passive (PS) and active optical sensor technologies (AOS) performance in regards to sky conditions effects and derive the NDVI (normalized difference vegetation index) relationship to wheat yield, as well as evaluate KSU optical sensor-based N recommendations against KSU soil test N recommendation system and sUAS (small unmanned aircraft systems) based recommendation algorithms with the PS and AOS platforms. Each year (2015-2016 & 2016-2017) five field trails across Kansas were conducted during the winter wheat crop year in cooperation with county ag agents, farmers, and KSU Agronomy Experiment Fields. Treatments consisted of N response curve, 1st and 2nd generation KSU N recommendation algorithms, sUAS based recommendation algorithms, and KSU soil test based N recommendations applied in the spring using N rates ranging from 0 to 140 kg ha⁻¹. Results indicate the Holland Scientific Rapid Scan and MicaSense RedEdge NDVI data was strongly correlated and generated strong relationships with grain yield at 0.60 and 0.57 R² respectively. DJI X3 lacks an NIR band producing uncalibrated false NDVI and no relationship to grain yield at 0.03 R². Calibrated NDVI from both sensors are effective for assessing yield potential and could be utilized for developing N recommendation algorithms. However, sensor based treatments preformed equal to higher yields compared the KSU soil test recommendations, as well as reduced the amount of fertilizer applied compared to the soil test recommendation. The intensive management algorithm was the most effective in determining appropriate N recommendations across locations. This allows farmers to take advantage of potential N mineralization that can occur in the spring. Further research is needed considering on setting the NUE (nitrogen use efficiency) in KSU N rec. algorithms for effects of management practice, weather, and grain protein for continued refinement.

Orientador: Paulo Milton Barbosa Landim / Banca: Célia Regina Lopes Zimback / Banca: Dilermando Perecin / Banca: Marcelo Monteiro da Rocha / Banca: Roberto Wagner Lourenço / Resumo: O presente trabalho teve por objetivo verificar a variabilidade espacial de atributos químicos do solo, referentes à macronutrientes utilizando métodos estatísticos e geoestatísticos, em uma área considerada homogênea, contendo campo com pastagem e área de floresta. A área escolhida foi a da Fazenda Mata Verde, situada no município de Candeias do Jamari - RO, a cerca de 30 km da capital. Na fase inicial da pesquisa foi elaborado um plano amostral, a partir de uma préamostragem georreferenciada, que resultou na determinação do tamanho da amostra para cada uma das áreas estudadas, cuja opção foi coletar 130 amostras no perfil 00,20 m distribuídas em malha quadrada para cada uma das áreas. Os dados foram analisados em duas fases: análise química e análise estatística. A análise química das amostras foi realizada para identificar os teores de alumínio (Al), cálcio (Ca), matéria orgânica (MO) e potencial de hidrogênio (pH). Para a realização da análise quantitativa, os valores resultantes da análise química foram armazenados em planilhas eletrônicas com os respectivos posicionamentos geográficos. Na análise descritiva e inferências estatísticas clássicas, foram efetuados os cálculos da média, variância, desvio padrão, coeficiente de variação, coeficiente de assimetria e coeficiente de curtose, dos dados amostrados. Foram ajustados modelos teóricos de variogramas aos dados experimentais e realizado o teste de validação cruzada, bem como a construção de mapas de atributos de solo, fazendo uso de krigagem ordinária. A partir dos resultados obtidos, foi possível concluir que a ocorrência de disparidades nos mapas dos atributos Al, Ca, MO e pH, nas áreas estudadas, é um indicativo da necessidade de adoção de metodologias diferenciadas para uso da área de pastagem, considerando o manejo adotado pelo pecuarista. / Abstract: The purpose of this work is to verify spatial variability of soil's macro nutrients chemical attributes using statistical and geostatistical tools, in an area considered homogeneous containing pasture and forest. The chosen area was the Mata Verde Farm situated in Candeias do Jamari City - Rondonia State, distant 30 km from the State capital. In the initial stage a sampling plan was elaborated based upon a georeferencing presampling. This resulted in sample sizes for each one of studied areas. The chosen option was the collect of 130 samples in soil profile 00.20 m grid distributed for each one of the areas. The data was analyzed in two phases: chemical analysis and statistical analysis. The samples' chemicals analysis were done to identify macronutrients contents Al, Ca, organic matter (MO) and pH and the values of the chemical analysis were recorded in spreadsheets with respective geographic positions. To descriptive analysis and classic statistical inferences were carried out calculating means, variance, standard deviation, variation coefficient, asymmetry coefficient and kurtosis coefficient of sampled data. The theoretic model of variograms were adjusted to experimental data and carried out a cross validation test, as well as the maps construction of soil attributes using ordinary kriging. From results was possible to conclude that the occurrence of disparities in spatial distribution maps of aluminum, calcium, organic matter and pH in studied areas is an indicative of necessity of adoption of differentiated methodologies for use of the pasture area, considering the management adopted by the cattle farmers. / Doutor

Master of Science / Department of Computing and Information Sciences / William H. Hsu / The goal of this research is to present a comparison between different clustering and segmentation techniques, both supervised and unsupervised, to detect plant and crop rows. Aerial images, taken by an Unmanned Aerial Vehicle (UAV), of a corn field at various stages of growth were acquired in RGB format through the Agronomy Department at the Kansas State University. Several segmentation and clustering approaches were applied to these images, namely K-Means clustering, Excessive Green (ExG) Index algorithm, Support Vector Machines (SVM), Gaussian Mixture Models (GMM), and a deep learning approach based on Fully Convolutional Networks (FCN), to detect the plants present in the images. A Hough Transform (HT) approach was used to detect the orientation of the crop rows and rotate the images so that the rows became parallel to the x-axis. The result of applying different segmentation methods to the images was then used in estimating the location of crop rows in the images by using a template creation method based on Green Pixel Accumulation (GPA) that calculates the intensity profile of green pixels present in the images. Connected component analysis was then applied to find the centroids of the detected plants. Each centroid was associated with a crop row, and centroids lying outside the row templates were discarded as being weeds. A comparison between the various segmentation algorithms based on the Dice similarity index and average run-times is presented at the end of the work.