Oliver Geostatistical Applications for Precision Agriculture

1. Auflage 2010
ISBN: 978-90-481-9133-8
Verlag: Springer Netherlands
Format: PDF
Kopierschutz: 1 - PDF Watermark

E-Book, Englisch, 331 Seiten

The aim of this book is to bring together a series of contributions from experts in the field to cover the major aspects of the application of geostatistics in precision agriculture. The focus will not be on theory, although there is a need for some theory to set the methods in their appropriate context. The subject areas identified and the authors selected have applied the methods in a precision agriculture framework. The papers will reflect the wide range of methods available and how they can be applied practically in the context of precision agriculture. This book is likely to have more impact as it becomes increasingly possible to obtain data cheaply and more farmers use onboard digital maps of soil and crops to manage their land. It might also stimulate more software development for geostatistics in PA.

Autoren/Hrsg.

Oliver, M.A.

Oliver, Margaret A.

Inhaltsverzeichnis

1;Preface;6
2;Contents;8
3;Contributors;14
4;Chapter 1: An Overview of Geostatistics and Precision Agriculture;16
4.1;1.1 Introduction;16
4.1.1;1.1.1 A Brief History of Geostatistics;17
4.1.2;1.1.2 A Brief History of Precision Agriculture;18
4.1.3;1.1.3 A Brief History of Geostatistics in Precision Agriculture;21
4.2;1.2 The Theory of Geostatistics;22
4.2.1;1.2.1 Stationarity;23
4.2.1.1;1.2.1.1 Intrinsic Variation and the Variogram;24
4.2.2;1.2.2 The Variogram;24
4.2.2.1;1.2.2.1 Estimating the Variogram;24
4.2.2.2;1.2.2.2 Features of the Variogram;25
4.2.2.3;1.2.2.3 Modelling the Variogram;26
4.2.3;1.2.3 Geostatistical Prediction: Kriging;27
4.2.3.1;1.2.3.1 Ordinary Kriging;28
4.2.3.2;1.2.3.2 Kriging Weights;30
4.2.3.3;1.2.3.3 Other Types of Kriging;30
4.2.3.4;1.2.3.4 Disjunctive Kriging;31
4.3;1.3 Case Study: Football Field;33
4.3.1;1.3.1 Summary Statistics;34
4.3.2;1.3.2 Variography;35
4.3.3;1.3.3 Kriging;41
4.3.3.1;1.3.3.1 Ordinary Kriging;41
4.3.3.2;1.3.3.2 Disjunctive Kriging;44
4.3.3.3;1.3.3.3 Factorial Kriging;45
4.3.4;1.3.4 Conclusions;46
4.4;References;47
5;Chapter 2: Sampling in Precision Agriculture;50
5.1;2.1 Introduction;51
5.1.1;2.1.1 The Importance of Spatial Scale for Sampling;52
5.1.2;2.1.2 How Can Geostatistics Help?;53
5.1.3;2.1.3 How can the Variogram be Used to Guide Sampling?;54
5.2;2.2 Variograms to Guide Sampling;55
5.2.1;2.2.1 Nested Survey and Analysis: Reconnaissance Variogram;55
5.2.1.1;2.2.1.1 Unequal Sampling;55
5.2.2;2.2.2 Variograms from Ancillary Data;58
5.2.2.1;2.2.2.1 Case Study;58
5.3;2.3 Use of the Variogram to Guide Sampling for Bulking;62
5.3.1;2.3.1 Case Study;63
5.4;2.4 The Variogram to Guide Grid-Based Sampling;66
5.4.1;2.4.1 The Variogram and Kriging Equations;66
5.4.1.1;2.4.1.1 Case Study;66
5.4.2;2.4.2 Half the Variogram Range `Rule of Thumb' as a Guide to Sampling Interval;69
5.5;2.5 Variograms to Improve Predictions from Sparse Sampling;70
5.5.1;2.5.1 Residual Maximum Likelihood (REML) Variogram Estimator;70
5.5.1.1;2.5.1.1 Case Study;71
5.5.2;2.5.2 Standardized Variograms;74
5.6;2.6 Conclusions;76
5.7;References;77
6;Chapter 3: Sampling in Precision Agriculture, Optimal Designs from Uncertain Models;79
6.1;3.1 Introduction;79
6.2;3.2 The Linear Mixed Model: Estimation, Predictionsand Uncertainty;81
6.2.1;3.2.1 The Model;81
6.2.2;3.2.2 Estimation;82
6.2.3;3.2.3 Prediction;84
6.2.4;3.2.4 Uncertainty;85
6.3;3.3 Optimizing Sampling Schemes by SpatialSimulated Annealing;86
6.3.1;3.3.1 Spatial Simulated Annealing;86
6.3.2;3.3.2 Objective Functions from the LMM;87
6.3.3;3.3.3 Optimized Sample Scheme for Single Phase Geostatistical Surveys;91
6.3.4;3.3.4 Adaptive Exploratory Surveys to Estimatethe Variogram;92
6.4;3.4 A Case Study in Soil Sampling;95
6.5;3.5 Conclusions;99
6.6;References;100
7;Chapter 4: The Spatial Analysis of Yield Data;102
7.1;4.1 Introduction;102
7.2;4.2 Background of Site-Specific Yield Monitors;103
7.2.1;4.2.1 Concept of a Yield Monitor;106
7.2.2;4.2.2 Calibration and Errors;107
7.2.3;4.2.3 Common Uses of Yield Monitor Data;108
7.2.4;4.2.4 Profitability of Yield Monitors;109
7.2.5;4.2.5 Quantity and Quality of Product;110
7.3;4.3 Managing Yield Monitor Data;110
7.3.1;4.3.1 Quality of Yield Monitor Data;110
7.3.2;4.3.2 Challenges in the Use of Yield Data for Decision Making;113
7.3.3;4.3.3 Aligning Spatially Disparate Spatial Data Layers;113
7.4;4.4 Spatial Statistical Analysis of Yield Monitor Data;114
7.4.1;4.4.1 Explicit Modelling of Spatial Effects;114
7.4.2;4.4.2 Spatial Interaction Structure;116
7.4.3;4.4.3 Empirical Determination of Spatial Neighbourhood Structure;117
7.5;4.5 Case Study: Spatial Analysis of Yield Monitor Data from a Field-Scale Experiment;120
7.5.1;4.5.1 Case Study Data;120
7.5.2;4.5.2 Data Analysis;123
7.5.3;4.5.3 Case Study Results;125
7.5.4;4.5.4 Case Study Summary;125
7.6;4.6 Conclusion;126
7.7;References;126
8;Chapter 5: Space-Time Geostatistics for Precision Agriculture: A Case Study of NDVI Mapping for a Dutch Potato Field;130
8.1;5.1 Introduction;130
8.2;5.2 Description of the Lauwersmeer Study Site and Positional Correction of NDVI Data;132
8.3;5.3 Exploratory Data Analysis of Lauwersmeer Data;133
8.4;5.4 Space--Time Geostatistics;138
8.4.1;5.4.1 Characterization of the Trend;139
8.4.2;5.4.2 Characterization of the Stochastic Residual;139
8.5;5.5 Application of Space--Time Geostatistics to the Lauwersmeer Farm Data;141
8.5.1;5.5.1 Characterization of the Trend;141
8.5.2;5.5.2 Characterization of the Stochastic Residual;143
8.5.3;5.5.3 Space--Time Kriging;144
8.6;5.6 Discussion and Conclusions;147
8.7;References;149
9;Chapter 6: Delineating Site-Specific Management Units with Proximal Sensors;151
9.1;6.1 Introduction;152
9.1.1;6.1.1 The Need for Site-Specific Management;152
9.1.2;6.1.2 Definition of Site-Specific Management Unit (SSMU);153
9.1.3;6.1.3 Proximal Sensors;153
9.1.4;6.1.4 Objective;156
9.2;6.2 Directed Sampling with a Proximal Sensor;157
9.2.1;6.2.1 Complexity of Proximal Sensor Measurements and the Role of Geostatistics;157
9.2.2;6.2.2 Practical Consideration of Differences in Support;158
9.3;6.3 Delineation of SSMUs with a Proximal Sensor;158
9.3.1;6.3.1 Geostatistical Mixed Linear Model;158
9.3.2;6.3.2 Soil Sampling Strategies Based on Geo-Referenced Proximal Sensor Data;160
9.3.3;6.3.3 Applications of Geostatistical Mixed Linear Models to Proximal Sensor Directed Surveys;162
9.4;6.4 Case Study Using Apparent Soil Electrical Conductivity (ECa) -- San Joaquin Valley, CA;163
9.4.1;6.4.1 Materials and Methods;163
9.4.1.1;6.4.1.1 Study Site;163
9.4.1.2;6.4.1.2 ECa-Directed Soil Sampling Protocols for Site-Specific Management;163
9.4.1.3;6.4.1.3 Yield Monitoring and ECa Survey;165
9.4.1.4;6.4.1.4 Sample Site Selection, Soil Sampling and Soil Analyses;166
9.4.1.5;6.4.1.5 Statistical and Spatial Analyses;167
9.4.2;6.4.2 Results and Discussion;167
9.4.2.1;6.4.2.1 Correlation Between Crop Yield and ECa;167
9.4.2.2;6.4.2.2 Exploratory Statistical Analysis;167
9.4.2.3;6.4.2.3 Crop Yield Response Model Development;169
9.4.2.4;6.4.2.4 Site-Specific Management Units;171
9.5;6.5 Conclusion;173
9.6;References;173
10;Chapter 7: Using Ancillary Data to Improve Prediction of Soil and Crop Attributes in Precision Agriculture;178
10.1;7.1 Introduction;178
10.2;7.2 Theory;180
10.2.1;7.2.1 Variogram and Cross-Variogram;180
10.2.2;7.2.2 Cokriging;181
10.2.3;7.2.3 Simple Kriging with Local Means;182
10.2.4;7.2.4 Kriging with an External Drift;183
10.3;7.3 Case Study 1: The Yattendon Site;183
10.3.1;7.3.1 Site Description and Available Data;183
10.3.2;7.3.2 Data Preparation;185
10.3.3;7.3.3 Variograms;187
10.3.4;7.3.4 Leave-One-Out Cross-Validation;189
10.3.5;7.3.5 Patterns of Variation;192
10.3.6;7.3.6 How Small Can the Sample Size of Primary Data be when Secondary Data are Available?;195
10.4;7.4 Case Study 2: The Wallingford Site;199
10.4.1;7.4.1 Site Description and Available Data;199
10.4.2;7.4.2 Leave-One-Out Cross-Validation Using Grid Sampled Data;200
10.4.3;7.4.3 Patterns of Variation;201
10.5;7.5 Conclusions;203
10.6;References;204
11;Chapter 8: Spatial Variation and Site-Specific Management Zones;206
11.1;8.1 Introduction;207
11.2;8.2 Quantifying Spatial Variation in Soil and Crop Properties;208
11.3;8.3 Site-Specific Management Zones;210
11.3.1;8.3.1 Soil Properties, Crops and Geographic Distribution of Management Zones;211
11.3.2;8.3.2 Techniques of Delineating Management Zones;213
11.4;8.4 Statistical Evaluation of Management Zone Delineation Techniques: A Case Study;220
11.5;8.5 Conclusions;226
11.6;References;227
12;Chapter 9: Weeds, Worms and Geostatistics;231
12.1;9.1 Introduction;231
12.2;9.2 Weeds;232
12.3;9.3 Nematodes;238
12.3.1;9.3.1 Lives of Nematodes;238
12.3.2;9.3.2 Geostatistical Applications;239
12.3.3;9.3.3 Case Study;241
12.3.4;9.3.4 Economics;244
12.4;9.4 The Future for Geostatistics in Precise Pest Control;249
12.5;References;250
13;Chapter 10: The Analysis of Spatial Experiments;252
13.1;10.1 Introduction;253
13.2;10.2 Background;254
13.3;10.3 Management-Class Experiments;256
13.3.1;10.3.1 Case Study I: REML-Based Analysis of a Management-Class Experiment;259
13.4;10.4 Local-Response Experiments;262
13.4.1;10.4.1 Case Study II: Analysis of a Local-ResponseExperiment;266
13.5;10.5 Alternative Approaches to Experimentation;270
13.6;10.6 Issues for the Future;272
13.7;10.7 Conclusions;273
13.8;References;274
14;Chapter 11: Application of Geostatistical Simulation in Precision Agriculture;277
14.1;11.1 Introduction;278
14.1.1;11.1.1 Basics of Geostatistical Simulation;279
14.1.2;11.1.2 Theory;282
14.1.2.1;11.1.2.1 Spatial Random Variable and Spatial Random Function;282
14.1.2.2;11.1.2.2 Stochastic Simulation;282
14.1.2.3;11.1.2.3 Overview of Methods for Geostatistical Simulation;282
14.1.3;11.1.3 Sequential Gaussian Simulation;283
14.1.4;11.1.4 Transformation of Probability Distributions;285
14.2;11.2 Case Study I: Uncertainty of a pH Map;286
14.2.1;11.2.1 Introduction;286
14.2.2;11.2.2 Materials and Methods;286
14.2.3;11.2.3 Results and Discussion;288
14.2.3.1;11.2.3.1 Simulation;289
14.2.3.2;11.2.3.2 Comparison of Kriging and Simulation;290
14.2.3.3;11.2.3.3 Prediction Error of the Interpolated Map;292
14.2.3.4;11.2.3.4 Probability that a pH Value is Outside the Optimal Range;293
14.2.4;11.2.4 Summary and Conclusions;294
14.3;11.3 Case Study II: Uncertainty in the Positionof Geographic Objects;295
14.3.1;11.3.1 Introduction;295
14.3.2;11.3.2 Methods;296
14.3.2.1;11.3.2.1 Definition of Positional Error Model1;296
14.3.2.2;11.3.2.2 Identification of the Model;297
14.3.2.3;11.3.2.3 Application;298
14.3.2.4;11.3.2.4 Data and Scenarios;299
14.3.3;11.3.3 Study Site;299
14.3.3.1;11.3.3.1 Software;300
14.3.3.2;11.3.3.2 Results and Discussion;300
14.3.4;11.3.4 Conclusions;304
14.4;11.4 Case Study III: Uncertainty Propagation in Soil Mapping;304
14.4.1;11.4.1 Introduction;304
14.4.2;11.4.2 Materials and Methods;305
14.4.3;11.4.3 Results and Discussion;306
14.4.4;11.4.4 Conclusions;308
14.5;11.5 Application of Geostatistical Simulation in Precision Agriculture: Summary;308
14.6;References;309
15;Chapter 12: Geostatistics and Precision Agriculture: A Way Forward;312
15.1;12.1 Introduction;312
15.2;12.2 Weather, Time and Space;313
15.3;12.3 Farmers, Advisors and Researchers;315
15.4;12.4 Issues, Ideas and Questions;317
15.5;12.5 Past, Present and Future;319
15.6;References;319
16;Appendix: Software;320
16.1;A.1 Geostatistics in GenStat;320
16.2;A.2 VESPER;322
16.2.1;A.2.1 Background;322
16.2.2;A.2.2 The Software;323
16.2.3;A.2.3 Applications;327
16.3;A.3 SGeMS and Other Software;328
16.3.1;A.3.1 SGeMS;328
16.3.2;A.3.2 Other Software;329
16.4;References;329
17;Index;331