the-go Proximal Soil Sensing for Agriculture

Transcription

1 AGRI-SENSING 2011 (Haifa, Israel) On-the the-go Proximal Soil Sensing for Agriculture Viacheslav I. Adamchuk Department of Bioresource Engineering McGill University Problem Statement The assessment of soil variability is one of the most important steps in site-specific management Conventional means to attain soil variability data are incapable of accurately identifying spatial inconsistency within a production field at an economically feasible cost There is a need to develop equipment for mapping soil attributes on-the-go February 21, 2011 Sensor Use Approaches Nature of Sensing Listen Smell Touch Real-Time Application Map-Based Approach Look Taste Integrated Approach (Real-Time with Supplemental Base Maps) On-the the-go Proximal Soil Sensors Electrical end Electromagnetic Sensors Electrical and Electromagnetic Optical and Radiometric Electrical Conductivity/Resistivity Sensors Electromagnetic Induction Method Galvanic Contact Resistivity Method Capacitively-Coupled Resistivity Method Dielectric Sensors Magnetic Sensors Acoustic Mechanical Electrochemical Pneumatic Soil type/texture Salinity Water content Organic matter content Depth variability Soil ph / nitrate content Volumetric water content Soil type/structure Salinity EC 1 Subsurface soil impurities Iron H + H + H + H + EC 2 EC a H + EC EC EC 3 Soil Soil 1

2 Galvanic Contact Resistivity Method Electromagnetic Induction Method A I N M U B Equipotentials Transmitter Secondary field Receiver Current flow Primary field Soil Eddy currents Veris Technologies, Inc. (Salina, Kansas) Geocarta (Paris, France) Crop Tehchnologies, Inc. (Spring, Texas) Veris 3100 and MSP (0.3 and 0. m) Geocarta ARP (0., 1, and 2 m) Soil Doctor System (real-time approach) Geonics Limited (Mississauga, Ontario) Dualem, Inc. (Milton, Ontario) Geonics EM-3 horizontal 0. m vertical 1. m DUALEM 1S co-planar 0. m perpendicular 0. m Capacitor analogue Capacitively-Coupled Coupled Resistivity Method Coaxial Transmitter cable Metal shield as a capacitor plate Inner wire Insulation as dielectric material Pneumatic Angular Scanning System (PASS) Soil as a capacitor plate Soil Geometrics, Inc. (San Jose, California) Geometrix OhmMapper TR Dielectric Sensor Dielectric Sensor Map Retrokool (Berkeley, California) Silty clay loam soil Triple replicates Two tests Sensor Output, V Test 1 Test 2 2. Average Gravimetric Soil Moisture, g/g 2

3 Soil Water Content Optical and Radiometric Sensors Laboratory measurement 30% 2% 20% 1% 10% % 0% Gravimetric soil moisture R 2 = 0. SE = 2.% Volumetric soil moisture 0% 0% % 10% 1% 20% 2% 30% 10% Sensor measurement % Laboratory measurement 3% R 2 = 0. SE = 3.% 30% 2% 20% 1% 0% 0% 10% 20% 30% 0% Sensor measurement Subsurface Soil Reflectance Sensors Visual Near-infrared Mid-infrared Image analysis Polarized light Organic matter (carbon) content Soil texture Cation exchange capacity (CEC) Soil water content Soil ph Mineral nitrogen and phosphorous Microwave Sensors Water content Ground Penetrating Radar Water content Geophysical soil structure Gamma Radiometer Potassium Uranium Thorium Small Scale Topography Subsurface Soil Reflectance Sensors Individual Wavelengths VIS/NIR Spectrophotometer 660 nm LEDs Shank Photodiode Purdue University (West Lafayette, Indiana) DGPS Antenna Notebook Computer Coulter Spectroradiometer Light Source Hyperspectral Response Reflectance, % Nebraska soils Shank Wavelength, nm Sapphire Window Predicted Carbon Measured Carbon Veris Technologies, Inc. (Salina, Kansas) Integrate Soil Mapping System (ISMS) Load Cell Sensor Mechanical Sensors Soil strength Draft and vertical force Soil profile Bulk soil strength Vertically actuated Tip-based Tine-based Optical Sensor Verticallyoperated cone penetrometers Single-tip horizontal Multiple-tip horizontal Vertically oscillating and Holland Scientific Retrokool (Berkeley, California) Dielectric Sensor Soil mechanical resistance Soil compaction Water content Soil types Depth of hard (plow) pan Cantilever beam Direct load 3

4 Soil Mechanical Resistance Mapping Discrete Depth Profiling Tools Tool Bar Load Cell Array Three Cutting Blades Strain Gauges Travel Direction Purdue University (West Lafayette, Indiana) UC-Davis (Davis, California) University of Missouri (Columbia, Missouri) Example Soil Mechanical Resistance Map Integrated Soil Physical Properties Mapping System (ISPPMS) Compacted area Old roads Two wavelengths soil reflectance sensor Soil Mechanical Resistance Map (20-30 cm) Yield Map Soil mechanical resistance profiler with an array of strain gage bridges Capacitor-based sensor Bulk Density Prediction Disc Coulter Sensor GPS Antennae & DAQ Moisture R 2 = 0.0 Soil Mechanical Resistance R 2 = 0.1 Ultrasound Depth Sensor 1. Bulk Density Predic ted Bulk Density, g/ml R 2 = Measured Bulk Density, g/ml Rotational Potentiometer

5 Instrumented Tillage Implement Variable Depth Tillage Concept Laptop with DAQ Card GPS Antenna Signal Conditioning Unit Load Cells Strain Gauges Soil Surface Soil Strength Strain Gauges Depth Sensor Custom Protective Shin Load Cells Straight Standard Custom Point US Patent No.,02, Depth Acoustic and Pneumatic Sensors Electrochemical Sensors Soil Penetration Noise Sensors Air Permeability Sensor Ion-Selective Electrodes (ISEs) Ion-Selective Field Effect Transistors (ISFETs) Soil Solution Measurement Agitated Soil Measurement Direct Soil Measurement University of Illinois (Urbana-Champaign, Illinois) Soil clay content (type) Soil compaction Depth of hard (plow) pan Soil structure/tilth Water content Soil type University of Kentucky (Lexington, Kentucky) Conventional Laboratory Analysis Activity of selected ions Soil ph (H + ) Potassium content (K + ) Residual nitrogen (NO 3- -N) Sodium content (Na + ) Automated Soil Testing Automated Soil ph Mapping Systems Soil cutters Purdue University (West Lafayette, Indiana) Shank Coring tube Mixing ISFET Electrode Water jet ~0 g soil core Add 20 ml DI H 2 O Cleaning Sample collection for calibration Purdue University (West Lafayette, Indiana) US Patent No. 6,36,30

6 Mobil Sensor Platform (MSP) Direct Soil Measurement Purdue University (West Lafayette, Indiana) Veris Technologies, Inc. (Salina, Kansas) Ion-selective Electrodes Water Nozzle Soil Sampler EC Surveyor 310 Veris Technologies, Inc. (Salina, Kansas) Soil ph Manager Example Soil ph Mapping Mapping Alternatives Directed Soil Sampling On-the-Go Sensing Soil Sampling Soil ph Maps of a Kansas Field Universal Directed Sampling Field Average Interpolated Grid Map Adjusted Shifted Universal = Intercept Universal + Slope Universal On-the-Go Mapping Conventional 1 ha Grid Sampling Adjusted MSP Shifted MSP ph = InterceptField specific + SlopeField ph = Shift Field specific + specific Soil ph Maps Evaluation Antimony Electrode 2. Acre Grid Raw MSP Adjusted MSP IA1 R 2 = 0. R 2 = Grid Sampling ph R 2 = 0.60 IA1 R 2 = Unprocessed R 2 = 0.1 IA1 R 2 = Adjusted 6 Field Average ph 6 Universal 6 Shifted IA1 1 Field Average IA1 1 Universal MSP IA1 1 Shifted MSP Sandy and stony soils Soil ph Antimony Electrodes R 2 = 0. SE = Soil ph Glass Electrodes 6

7 Portable Probe On-the the-spot Measurement of Soil ph Soil ph Portable Probe Plot A Plot B Plot C Average Plot A Average Plot B Average Plot C Reference Lab ~ ~ 1 ~2 Depth, cm Measured soluble potassium, mg/kg Integrated Direct Soil Measurement 1 Nebraska soils with fixed field water content pk R 2 = 0.2 (0.62 means) RMSE (Precision) = 0.13 pk. SE (Accuracy) = 0.1 pk Reference soluble potassium (AAS), mg/kg Measured ph Measured nitrate-nitrogen, mg/kg R 2 = 0.3 (0.6 means) RMSE (Precision) = 0.12 ph. SE (Accuracy) = 0.16 ph ph Reference ph 10 R 2 = 0.3 (0.61 means) RMSE (Precision) = 0.1 pno3. SE (Accuracy) = 0.12 pno3 pno Reference nitrate-nitrogen (CR), mg/kg Measured soil ph and predicted buffer ph VRT Prescription LR = f (buffer ph) Predicted buffer ph Measured soil ph ( ) Buffer ph = f (soil ph, CEC) R 2 = 0.1 R 2 = 0.6 Measured soluble K and predicted exchangeble K, mg kg Measured buffer ph K rate = f (exchangeable K) Exchangeable K = f (soluble K, CEC) 0 Predicted exchangable K Measured soluble K R 2 = 0. R 2 = Measured exchangeble K, mg kg -1 Integrated Multiple Data Layers Applicability of On-the the-go Soil Sensors Soil property H + H + H + H + H + Soil texture (clay, silt and sand) Soil organic matter or total carbon Soil ph & Clay & OM Soil water (moisture) Soil salinity (sodium) Soil compaction (bulk density) = Lime Recommendation Depth variability (hard pan) Soil ph Residual nitrate (total nitrogen) Other nutrients (potassium) Maps produced by Veris Technologies, Inc. (Salina, Kansas) CEC (other buffer indicators)

8 Targeted Soil Sampling Directed sampling should be used to calibrate and/or validate sensor data Directed samples should be collected from relatively homogeneous field areas away from the boundary and other transitional areas Directed samples should cover the entire range of sensor-based measurements, especially toward low and high ends Directed samples should be physically spread across the entire field It should be possible to process multiple sensor-based data layers OF = S opt Objective Function D S-optimality D-optimality (soil ph) D-optimality (soil EC) H-criteria (soil ph) H-criteria (soil EC) opt ph D opt EC H cr ph H cr EC Summary On-the-go soil can provide high density information about soil properties Many sensor approaches are past initial commercialization stage Sensor fusion provides the ability to separate various agronomic effects Site-specific sensor calibration and validation are essential steps of the mapping process Laboratory soil analysis remains a required supplementary practice E:mail: viacheslav.adamchuk@mcgill.ca