AgTech - Precision Agriculture: Lecture 2
RFID in Livestock
- Purpose: identification, traceability, data integration with herd management and NLIS (National Livestock Identification System).
- Tag types: Active or Passive (NLIS is passive).
- Timeline: RFID introduced in livestock around 1999; compulsory tagging for cattle; also used for sheep.
- Uses: track movements property-to-property, assist management decisions, treatments, and performance data.
- How it works (passive): tag, reader, antenna; energy from antenna wakes tag; tag transmits data to reader.
Walk-over-Weigh (WOW)
- Concept: weight data captured as animals walk over a weigh platform with load cells; RFID tag read by panel reader during passage.
- Data: liveweight logged and linked to tag number; typically located at water points or attractants.
- Hardware overview (box on page): single-board computer, battery, 4G LTE modem, charger controller, RFID panel reader, RFID antenna, weighing platform, load bars.
- Benefits: daily individual weight monitoring, rapid management interventions, reduced weight loss, identify poor performers, lower labour for mustering.
- Challenges: learning curve (e.g., cows ~10 days), multiple animals on platform, getting all sheep to use it, weight accuracy, data transfer.
OptiWeigh
- Description: portable, movable in-paddock weighing system.
- Benefits: voluntary use, minimal labour, reduced site setup; in-paddock weights.
- Limitations: only part of mob weighed; need training to use a portion of mob; requires attractant.
On-animal Sensors
- Components: GPS/GNSS for location; accelerometer for behaviour; 3G/4G connectivity for data transmission.
- Concept: continuous monitoring of location and activity to infer health, welfare, and production signals.
- Example: Sheep Health Monitor shows real-time alerts (e.g., sick sheep) via mobile messages.
- Advantages: 24/7 monitoring, early issue detection, behavioural insights.
- Challenges: device robustness, power consumption, data processing, and connectivity.
Virtual Fencing (VF)
- Principle: audible warning + electrical stimulus when an animal crosses a virtual boundary.
- Zones: Permitted Zone vs Exclusion Zone (EZ).
- Potential benefits: environmental protection (riparian areas), precise pasture management, labour reductions, monitoring of movements.
- Challenges: animal welfare considerations, learning for all animals, longevity of sensors, form factor, public perception.
Plant Sensing: Remote vs Proximal
- Why: substantial variation in crop yield; sensing helps manage variability for future seasons.
- Terminology:
- Proximal = close to plants
- Remote = from a distance
- Active = uses its own light source
- Passive = relies on external light sources
- Sensing categories:
- Remote sensing: satellites, aircraft; passive and active depending on sensor type.
- Proximal sensing: ground-based vehicles or low-flying platforms; active sensors illuminate the crop.
- Advantages of proximal over remote: usable anytime, not dependent on sun/cloud, higher spatial resolution.
Electromagnetic Spectrum and Leaves
- Visible spectrum and chlorophyll interactions: red and blue absorbed, green reflected; NIR reflects due to leaf structure.
- Internal leaf anatomy influences reflectance in different bands (e.g., palisade cells, spongy mesophyll).
- Importance: spectral signatures enable assessment of plant condition and vigor.
NDVI and Spectral Signatures
- NDVI definition: NDVI = rac{NIR - R}{NIR + R} where NIR is near-infrared, R is red reflectance.
- Typical range: -1 \,\le\, NDVI \le 1; vegetation tends to higher NDVI values.
- Use: correlate with biomass; example shows strong correlation (R^2 ≈ 0.82) between NDVI and biomass measures.
- Other spectral trends: different crops and soils have distinct spectral signatures across bands (Blue, Green, Red, NIR).
Data Applications and Decision Making
- Biomass estimation: calibrate NDVI to available biomass to set stocking rates.
- Zoning and variable rate applications: apply fertilizers or inputs variably within paddocks.
- In-season management: variable rate N application, yield prediction, and targeted weed control.
- Horticulture example: WeedSeeker targets weeds in fallow; precise spraying reduces chemical use and environmental impact.
WeedSeeker and Fallows
- Active sensors detect weeds; spray nozzle activates only where weeds are present.
- Benefits: chemical savings, reduced environmental impact.
- Disadvantages: expensive to set up; best for fallow with intermittent weeds.
- Related systems: WEED-IT, John Deere See & Spray.
What You Should Know (Key Takeaways)
- Understand remote sensing vs proximal sensing concepts and uses.
- Know how plant sensors work and their production applications.
- Be aware of precision livestock sensing options, purposes, and challenges.
- Recognize main challenges associated with various livestock technologies (hardware, data, welfare, power, processing).
Quick Reference Equations
- NDVI: NDVI = rac{NIR - R}{NIR + R}
- NDVI range: -1 \,\le\, NDVI \le 1
Practical Takeaways
- RFID and WOW enable automated, high-frequency individual animal data collection.
- On-animal sensors enable continuous monitoring of location, movement, and health signals.
- Virtual fencing offers a boundary mechanism without physical fencing, with welfare and public-perception considerations.
- Plant sensing (remote and proximal) supports spatial management of crops and rangelands via spectral indices like NDVI.
- Data analytics (biomass estimation, variable-rate inputs, yield prediction) drive efficient, site-specific management.