Precision Agriculture: Variable Rate Irrigation Explained

Precision agriculture variable rate irrigation delivers water to different crop zones based on their actual need rather than uniform scheduling — reducing water consumption by 30-50% while improving yield uniformity. As India faces escalating groundwater depletion (water table falling at 10-25 cm/year in Punjab and Haryana), variable rate irrigation (VRI) using IoT sensors and automated control systems is becoming essential for sustainable farming. This guide explains VRI technology, implementation approaches, and practical builds for Indian farms.

Table of Contents

• What is Variable Rate Irrigation
• VRI vs Uniform Irrigation: Comparison
• IoT Sensor Network Design
• Components Required
• VRI Control Algorithm
• Python VRI Controller Code
• Indian Farm Applications
• Frequently Asked Questions

What is Variable Rate Irrigation

Variable rate irrigation applies different amounts of water to different management zones within the same field. A 10-acre wheat field may have three distinct zones:

• Zone 1 (Sandy soil, south slope): Drains fast, needs 20% more water
• Zone 2 (Loam soil, flat area): Standard field capacity, reference zone
• Zone 3 (Clay soil, north edge): Waterlogging prone, needs 30% less water

Uniform irrigation over-waters Zone 3 and under-waters Zone 1. VRI optimises for each zone by varying sprinkler application time, pivot speed, or drip valve duration.

VRI vs Uniform Irrigation: Comparison

IoT Sensor Network Design

A VRI system requires:

1. Zone delineation: Soil sampling or EC mapping to identify management zones
2. Soil moisture sensors: 1-3 per zone at root depth (15-45 cm)
3. Weather station: ET (evapotranspiration) calculation requires temperature, humidity, wind, and radiation data
4. Field controller: ESP32 or Raspberry Pi processes sensor data and controls valve/pump outputs
5. Communication: LoRa mesh or WiFi mesh links sensors to controller

Components Required

Per zone node:

• ESP32 (zone controller node)
• 2x Capacitive soil moisture sensors (15cm and 30cm depth)
• LoRa SX1278 module (for mesh communication)
• Solar power (2W panel + 2000mAh LiPo)

Central controller:

• Raspberry Pi 4 (for Python VRI algorithm)
• LoRa hat or USB LoRa adapter
• 8-channel relay board for solenoid valve control
• BME280 weather station (wind speed sensor optional addition)

VRI Control Algorithm

The VRI decision engine uses a water balance model:

1. Calculate daily ET: Using simplified Penman-Monteith or Hargreaves equation from BME280 data
2. Read soil moisture: Compare current soil moisture to field capacity for each zone
3. Calculate irrigation need: Depletion = Field capacity – Current moisture (in mm water depth)
4. Set zone run time: Run time = Depletion / Application rate (mm/hour for drip/sprinkler)
5. Apply rain adjustment: Subtract yesterday’s rainfall from scheduled irrigation

Python VRI Controller Code

import time
import math
from datetime import datetime

# Zone configuration
zones = [
    {"name": "Sandy Slope", "area_m2": 2000, "kc": 1.1, "fc_pct": 55, "pwp_pct": 20,
     "mad_pct": 50, "drip_rate_mm_hr": 3.0, "valve_gpio": 0},
    {"name": "Loam Flat",   "area_m2": 3000, "kc": 1.0, "fc_pct": 70, "pwp_pct": 30,
     "mad_pct": 50, "drip_rate_mm_hr": 2.5, "valve_gpio": 1},
    {"name": "Clay Edge",   "area_m2": 1500, "kc": 0.9, "fc_pct": 85, "pwp_pct": 40,
     "mad_pct": 40, "drip_rate_mm_hr": 2.0, "valve_gpio": 2}
]

def hargreaves_et0(tmax, tmin, tavg, ra_mj_m2_day):
    """Hargreaves ET0 estimation (simpler than full PM)."""
    return 0.0023 * (tavg + 17.8) * (tmax - tmin) ** 0.5 * ra_mj_m2_day

def solar_radiation_estimate(lat_deg, doy):
    """Extraterrestrial radiation (MJ/m2/day) for given latitude and day of year."""
    lat = math.radians(lat_deg)
    dr = 1 + 0.033 * math.cos(2 * math.pi * doy / 365)
    delta = 0.409 * math.sin(2 * math.pi * doy / 365 - 1.39)
    ws = math.acos(-math.tan(lat) * math.tan(delta))
    ra = (24 * 60 / math.pi) * 0.0820 * dr * (
        ws * math.sin(lat) * math.sin(delta) +
        math.cos(lat) * math.cos(delta) * math.sin(ws)
    )
    return ra

def calculate_vri_schedule(weather, soil_readings, rainfall_mm, lat=20.0):
    """Calculate irrigation duration for each zone."""
    doy = datetime.now().timetuple().tm_yday
    ra = solar_radiation_estimate(lat, doy)
    
    et0 = hargreaves_et0(
        weather['tmax'], weather['tmin'],
        (weather['tmax'] + weather['tmin']) / 2, ra
    )
    
    schedule = []
    for i, zone in enumerate(zones):
        etcrop = et0 * zone['kc']  # Crop ET adjusted by crop coefficient
        soil_pct = soil_readings[i]
        
        # Water deficit (mm) = (FC - current moisture) * root zone depth (mm)
        root_depth_mm = 300  # 30cm root zone
        fc_vol = zone['fc_pct'] / 100 * root_depth_mm
        current_vol = soil_pct / 100 * root_depth_mm
        deficit_mm = max(0, fc_vol - current_vol)
        
        # Also account for ET since last irrigation
        effective_rain = rainfall_mm * 0.8  # 80% efficiency
        net_deficit = max(0, deficit_mm + etcrop - effective_rain)
        
        # Apply MAD (Management Allowed Depletion) threshold
        mad_threshold_mm = (zone['fc_pct'] - zone['pwp_pct']) / 100 * root_depth_mm * zone['mad_pct'] / 100
        
        if net_deficit > mad_threshold_mm * 0.5:
            duration_hr = net_deficit / zone['drip_rate_mm_hr']
            duration_min = duration_hr * 60
        else:
            duration_min = 0  # No irrigation needed
        
        schedule.append({
            'zone': zone['name'],
            'et0': round(et0, 2),
            'etcrop': round(etcrop, 2),
            'deficit_mm': round(net_deficit, 1),
            'duration_min': round(duration_min, 0)
        })
        
    return schedule

# Example usage
weather = {'tmax': 32, 'tmin': 20}
soil_readings = [42, 68, 78]  # % soil moisture per zone
rainfall_mm = 0

schedule = calculate_vri_schedule(weather, soil_readings, rainfall_mm, lat=20.5)
for s in schedule:
    print(f"Zone: {s['zone']:15s} | ET0: {s['et0']}mm | Deficit: {s['deficit_mm']}mm | Run: {int(s['duration_min'])} min")

Indian Farm Applications

VRI provides the biggest benefit in fields with high spatial variability:

• Laterite soil farms (Maharashtra, Goa, Karnataka): Shallow, variable depth patches need zone-specific scheduling
• Sugarcane (Maharashtra, UP): VRI during grand growth phase reduces waterlogging-induced red rot
• Wheat (Punjab, Haryana): With groundwater declining at 25 cm/year, VRI can extend aquifer life by decades
• Banana (Jalgaon, Anand): High water requirement (2000mm/year) with critical timing at bunch emergence

Frequently Asked Questions

How do I identify management zones in my field?

Three approaches from simplest to most accurate: 1) Visual observation of crop stress patterns over 2-3 seasons, 2) EM38 soil EC mapping (contact local Krishi Vigyan Kendra), 3) Drone-based NDVI mapping in stressed conditions. Divide field into 2-5 zones based on the variation pattern.

What is the ROI of a DIY VRI system on an Indian farm?

For a 5-acre farm in Punjab with borewell at 150 feet depth: ESP32 + sensors + relay system costs Rs 25,000-35,000. Water savings of 40% = 2-3 lakh litres/season = Rs 3,000-8,000 in electricity savings. Yield improvement (10-15%) on a Rs 2,00,000/year operation adds Rs 20,000-30,000. Payback in one season is achievable.

Can VRI work with traditional flood irrigation?

VRI is most effective with drip or sprinkler systems where application rate can be controlled per zone. For flood irrigation, a simpler approach is to use soil moisture sensors to stop irrigation when each zone reaches field capacity — this reduces over-application by 20-30% without zone-specific control.

Shop Precision Farming Sensors at Zbotic