The notion that agriculturally productive land may be treated as a relatively homogeneous resource at thewithin-field scale is not sound. This assumption and the subsequent uniform application of planting material,chemicals and/or tillage effort may result in zones within a field being under- or over-treated. Arising fromthese are problems associated with the inefficient use of input resources, economically significant yield losses,excessive energy costs, gaseous or percolatory release of chemicals into the environment, unacceptable long-term retention of chemicals and a less-than-optimal growing environment. The environmental impact of cropproduction systems is substantial. In this millennium, three important issues for scientists and agrariancommunities to address are the need to efficiently manage agricultural land for sustainable production, themaintenance of soil and water resources and the environmental quality of agricultural land.Precision agriculture (PA) aims to identify soil and crop attribute variability, and manage it in an accurate andtimely manner for near-optimal crop production. Unlike conventional agricultural management where anaveraged whole-field analytical result is employed for decision-making, management in PA is based on site-specific soil and crop information. That is, resource application and agronomic practices are matched withvariation in soil attributes and crop requirements across a field or management unit. Conceptually PA makeseconomic and environmental sense, optimising gross margins and minimising the environmental impact ofcrop production systems. Although the economic justification for PA can be readily calculated, concepts suchas environmental containment and the safety of agrochemicals in soil are more difficult to estimate. However,it may be argued that if PA lessens the overall agrochemical load in agricultural and non-agriculturalenvironments, then its value as a management system for agriculture increases substantially.Management using PA requires detailed information of the spatial and temporal variation in crop yieldcomponents, weeds, soil-borne pests and attributes of physical, chemical and biological soil fertility. However,detailed descriptions of fine scale variation in soil properties have always been difficult and costly to perform.Sensing and scanning technologies need to be developed to more efficiently and economically obtain accurateinformation on the extent and variability of soil attributes that affect crop growth and yield. The primary aimof this work is to conduct research towards the development of an �on-the-go� proximal soil pH and limerequirement sensing system for real-time continuous management of acid soil. It is divided into four sections.Section one consists of two chapters; the first describes global and historical events that converged into thedevelopment of precision agriculture, while chapter two provides reviews of statistical and geostatisticaltechniques that are used for the quantification of soil spatial variability and of topics that are integral to theconcept of precision agriculture. The review then focuses on technologies that are used for the completeenumeration of soil, namely remote and proximal sensing.Section two comprises three chapters that deal with sampling and mapping methods. Chapter three provides ageneral description of the environment in the experimental field. It provides descriptions of the field site,topography, soil condition at the time of sampling, and the spatial variability of surface soil chemicalproperties. It also described the methods of sampling and laboratory analyses. Chapter four discusses some ofthe implications of soil sampling on analytical results and presents a review that quantifies the accuracy,precision and cost of current laboratory techniques. The chapter also presents analytical results that show theloss of information in kriged maps of lime requirement resulting from decreases in sample size. The messageof chapter four is that the evolution of precision agriculture calls for the development of �on-the-go� proximalsoil sensing systems to characterise soil spatial variability rapidly, economically, accurately and in a timelymanner. Chapter five suggests that for sparsely sampled data the choice of spatial modelling and mappingtechniques is important for reliable results and accurate representations of field soil variability. It assesses anumber of geostatistical methodologies that may be used to model and map non-stationary soil data, in thisinstance soil pH and organic carbon. Intrinsic random functions of order k produced the most accurate andparsimonious predictions of all of the methods tested.Section three consists of two chapters whose theme pertains to sustainable and efficient management of acidagricultural soil. Chapter six discusses soil acidity, its causes, consequences and current management practices.It also reports the global extent of soil acidity and that which occurs in Australia. The chapter closes byproposing a real-time continuous management system for the management of acid soil. Chapter seven reportsresults from experiments conducted towards the development of an �on-the-go� proximal soil pH and limerequirement sensing system that may be used for the real-time continuous management of acid soil.Assessment of four potentiometric sensors showed that the pH Ion Sensitive Field Effect Transistor (ISFET)was most suitable for inclusion in the proposed sensing system. It is accurate and precise, drift and hysteresisare low, and most importantly it�s response time is small. A design for the analytical system was presentedbased on flow injection analysis (FIA) and sequential injection analysis (SIA) concepts. Two different modesof operation were described. Kinetic experiments were conducted to characterise soil:0.01M CaCl2 pH(pHCaCl2) and soil:lime requirement buffer (pHbuffer) reactions. Modelling of the pHbuffer reactions describedtheir sequential, biphasic nature. A statistical methodology was devised to predict pHbuffer measurements usingonly initial reaction measurements at 0.5s, 1s, 2s and 3s measurements. The accuracy of the technique was 0.1pHbuffer units and the bias was low. Finally, the chapter describes a framework for the development of aprototype soil pH and lime requirement sensing system and the creative design of the system.The final section relates to the management of acid soil by liming. Chapter eight describes the development ofempirical deterministic models for rapid predictions of lime requirement. The response surface models arebased on soil:lime incubations, pHbuffer measurements and the selection of target pH values. These models aremore accurate and more practical than more conventional techniques, and may be more suitably incorporatedinto the spatial decision-support system of the proposed real-time continuous system for the management ofacid soil. Chapter nine presents a glasshouse liming experiment that was used to authenticate the limerequirement model derived in the previous chapter. It also presents soil property interactions and soil-plantrelationships in acid and ameliorated soil, to compare the effects of no lime applications, single-rate andvariable-rate liming. Chapter X presents a methodology for modelling crop yields in the presence ofuncertainty. The local uncertainty about soil properties and the uncertainty about model parameters wereaccounted for by using indicator kriging and Latin Hypercube Sampling for the propagation of uncertaintiesthrough two regression functions; a yield response function and one that equates resultant pH after theapplication of lime. Under the assumptions and constraints of the analysis, single-rate liming was found to bethe best management option.

The notion that agriculturally productive land may be treated as a relatively homogeneous resource at thewithin-field scale is not sound. This assumption and the subsequent uniform application of planting material,chemicals and/or tillage effort may result in zones within a field being under- or over-treated. Arising fromthese are problems associated with the inefficient use of input resources, economically significant yield losses,excessive energy costs, gaseous or percolatory release of chemicals into the environment, unacceptable long-term retention of chemicals and a less-than-optimal growing environment. The environmental impact of cropproduction systems is substantial. In this millennium, three important issues for scientists and agrariancommunities to address are the need to efficiently manage agricultural land for sustainable production, themaintenance of soil and water resources and the environmental quality of agricultural land.Precision agriculture (PA) aims to identify soil and crop attribute variability, and manage it in an accurate andtimely manner for near-optimal crop production. Unlike conventional agricultural management where anaveraged whole-field analytical result is employed for decision-making, management in PA is based on site-specific soil and crop information. That is, resource application and agronomic practices are matched withvariation in soil attributes and crop requirements across a field or management unit. Conceptually PA makeseconomic and environmental sense, optimising gross margins and minimising the environmental impact ofcrop production systems. Although the economic justification for PA can be readily calculated, concepts suchas environmental containment and the safety of agrochemicals in soil are more difficult to estimate. However,it may be argued that if PA lessens the overall agrochemical load in agricultural and non-agriculturalenvironments, then its value as a management system for agriculture increases substantially.Management using PA requires detailed information of the spatial and temporal variation in crop yieldcomponents, weeds, soil-borne pests and attributes of physical, chemical and biological soil fertility. However,detailed descriptions of fine scale variation in soil properties have always been difficult and costly to perform.Sensing and scanning technologies need to be developed to more efficiently and economically obtain accurateinformation on the extent and variability of soil attributes that affect crop growth and yield. The primary aimof this work is to conduct research towards the development of an �on-the-go� proximal soil pH and limerequirement sensing system for real-time continuous management of acid soil. It is divided into four sections.Section one consists of two chapters; the first describes global and historical events that converged into thedevelopment of precision agriculture, while chapter two provides reviews of statistical and geostatisticaltechniques that are used for the quantification of soil spatial variability and of topics that are integral to theconcept of precision agriculture. The review then focuses on technologies that are used for the completeenumeration of soil, namely remote and proximal sensing.Section two comprises three chapters that deal with sampling and mapping methods. Chapter three provides ageneral description of the environment in the experimental field. It provides descriptions of the field site,topography, soil condition at the time of sampling, and the spatial variability of surface soil chemicalproperties. It also described the methods of sampling and laboratory analyses. Chapter four discusses some ofthe implications of soil sampling on analytical results and presents a review that quantifies the accuracy,precision and cost of current laboratory techniques. The chapter also presents analytical results that show theloss of information in kriged maps of lime requirement resulting from decreases in sample size. The messageof chapter four is that the evolution of precision agriculture calls for the development of �on-the-go� proximalsoil sensing systems to characterise soil spatial variability rapidly, economically, accurately and in a timelymanner. Chapter five suggests that for sparsely sampled data the choice of spatial modelling and mappingtechniques is important for reliable results and accurate representations of field soil variability. It assesses anumber of geostatistical methodologies that may be used to model and map non-stationary soil data, in thisinstance soil pH and organic carbon. Intrinsic random functions of order k produced the most accurate andparsimonious predictions of all of the methods tested.Section three consists of two chapters whose theme pertains to sustainable and efficient management of acidagricultural soil. Chapter six discusses soil acidity, its causes, consequences and current management practices.It also reports the global extent of soil acidity and that which occurs in Australia. The chapter closes byproposing a real-time continuous management system for the management of acid soil. Chapter seven reportsresults from experiments conducted towards the development of an �on-the-go� proximal soil pH and limerequirement sensing system that may be used for the real-time continuous management of acid soil.Assessment of four potentiometric sensors showed that the pH Ion Sensitive Field Effect Transistor (ISFET)was most suitable for inclusion in the proposed sensing system. It is accurate and precise, drift and hysteresisare low, and most importantly it�s response time is small. A design for the analytical system was presentedbased on flow injection analysis (FIA) and sequential injection analysis (SIA) concepts. Two different modesof operation were described. Kinetic experiments were conducted to characterise soil:0.01M CaCl2 pH(pHCaCl2) and soil:lime requirement buffer (pHbuffer) reactions. Modelling of the pHbuffer reactions describedtheir sequential, biphasic nature. A statistical methodology was devised to predict pHbuffer measurements usingonly initial reaction measurements at 0.5s, 1s, 2s and 3s measurements. The accuracy of the technique was 0.1pHbuffer units and the bias was low. Finally, the chapter describes a framework for the development of aprototype soil pH and lime requirement sensing system and the creative design of the system.The final section relates to the management of acid soil by liming. Chapter eight describes the development ofempirical deterministic models for rapid predictions of lime requirement. The response surface models arebased on soil:lime incubations, pHbuffer measurements and the selection of target pH values. These models aremore accurate and more practical than more conventional techniques, and may be more suitably incorporatedinto the spatial decision-support system of the proposed real-time continuous system for the management ofacid soil. Chapter nine presents a glasshouse liming experiment that was used to authenticate the limerequirement model derived in the previous chapter. It also presents soil property interactions and soil-plantrelationships in acid and ameliorated soil, to compare the effects of no lime applications, single-rate andvariable-rate liming. Chapter X presents a methodology for modelling crop yields in the presence ofuncertainty. The local uncertainty about soil properties and the uncertainty about model parameters wereaccounted for by using indicator kriging and Latin Hypercube Sampling for the propagation of uncertaintiesthrough two regression functions; a yield response function and one that equates resultant pH after theapplication of lime. Under the assumptions and constraints of the analysis, single-rate liming was found to bethe best management option.

Thesis (M. Sc. (Soil Science)) --University of Limpopo, 2019 / Many conventional laboratory methods are used to characterise spatial and temporal variation of soil properties in order to understand soil quality for different purposes. Currently there is a high demand for accurate soil information by land users. Therefore there is a need to develop a rapid, inexpensive, non-destructive and accurate technique that could compensate or replace conventional laboratory methodologies. Remote sensing has the potential to serve as an alternative approach to characterise soil properties due to its advantages over conventional laboratory methods such as it is rapid, non-destructive and it has low cost. The objectives of this study were to: (i) evaluate the ability of proximal soil sensing to characterise soil properties namely organic matter, soil moisture content, macronutrients, soil texture, cation exchange capacity (CEC), and pH. (ii) Identify bands of relevance from proximal soil sensing (300-2400 nm) that can provide acceptable reflectance variation for different levels of selected soil properties. (iii) Evaluate the performance of models developed from multispectral space-borne image in characterising selected soil properties. In this study spectroradiometer (proximal sensor) and worldview 2 satellite images (space-borne) were the two remote sensing techniques used to collect information about soil at Syferkuil experimental farm of the University of Limpopo. Visible and near infrared spectral data of 98 soil samples were collected at the study site using Analytical spectral device (ASD) field spectroradiometer. Spectral reflectance from spectroradiometer and those extracted from worldview 2 satellite image were used to develop prediction models of selected soil properties using Partial least square regression (PLSR). Bands of relevance were also identified from PLSR models developed from spectral data acquired by spectroradiometer. The results showed that estimation accuracy of PLSR models developed using spectral data from proximal soil sensing were excellent (Category A) for clay, sand, soil organic matter (SOM), and soil moisture content, while good prediction accuracy (Category B) was observed for other soil properties such as silt, ammonium, nitrate, active acidity (pHw), calcium, magnesium, phosphorus, potassium, sulphur, CEC, and reserve acidity (pHKCl). Then, relevant bands which contributed greatly in the prediction of these soil attributes were selected from the electromagnetic spectrum, the range was from 451 nm to 2400 nm. These bands fall within visible, shortwave infrared and near-infrared x regions of electromagnetic spectrum. In addition all selected soil properties were approximately quantitatively estimated using spectral data from satellite image. Based on the results obtained it can be concluded that proximal soil sensing has the ability to predict selected soil properties with various accuracies and it can be used as an alternative technique to characterise soil properties of South African soils. Soil predicting models developed from proximal soil sensing data also showed that there are bands of relevance within spectral range of 451 nm to 2400 nm. However more work is required for space-borne sensing before it can be used as one of the soil characterisation methods since its prediction accuracy was low as compared to that of hyperspectral proximal soil sensing. Keywords: Space-borne sensing; proximal soil sensing; soil characterisation.