Skip to content

Agriculture · Sector explainer

Soil, water and weather, field by field.

Farm weather stations, irrigation, frost protection, livestock and parametric-insurance networks. Six scenes, each a working calculation: evapotranspiration to FAO-56, the root-zone water balance, frost inversions, canal flow, collar battery life over LoRa and the station density behind an index-insurance contract.

Scenarios
5
Live calculations
6
Community fit
Strong

01 · Agromet network

Scenario

Water demand, from one station.

Compact stations and soil probes feeding irrigation, disease and frost models. The station measures the five inputs of the FAO-56 Penman–Monteith equation, here for a site at 50° N and 300 m in mid-July; the mid-season crop coefficient turns reference evapotranspiration into each crop’s daily water use.

Crop, mid-season

Reference evapotranspiration ET₀
5.3 mm/day
Crop coefficient (Maize (grain))
1.20
Crop water use ETc = Kc · ET₀
6.3 mm/day
Net radiation Rn
14.2 MJ/m²/day
Vapour pressure deficit
1.18 kPa
Water over a 40 ha block
2,524 m³/day

02 · Soil water and the trigger

Scenario

Irrigate before the crop feels it.

Probes at four depths track the water left in the root zone. The trigger fires when depletion passes the readily available water, the share the crop can draw without stress, so water arrives before yield is lost. Total available water is 1000 (θFC − θWP) Zr; the crop draws 55 % of it without stress.

Soil texture

Total available water TAW
117 mm
Readily available water RAW (p = 0.55)
64 mm
Root-zone depletion Dr
36 mm
Root-zone water content
21.0 % vol
Days to the trigger
4.7 days
Irrigation to refill
36 mm · 360 m³/ha
Depletion against TAW; the red mark is RAW.

03 · Frost and inversions

Scenario

On a still night, warm air sits overhead.

Dense inversion-aware temperature sensors and wind-machine triggers. Radiation frost leaves colder air at bud height than ten metres up. A wind machine mixes that warmer layer down; it only helps when the inversion is strong enough, so sensors at both heights decide, against the grower’s two settings for the bud stage.

Temperature at 10 m
1.4 °C
Alarm at 1.5 m
On
Wind machine
Run
Warm air available above bud height
2.6 °C

04 · Irrigation district

Scenario

Every canal, metered.

Canal level and flow, soil moisture and pump control over private LTE or low-power radio. A level sensor in a lined channel gives discharge through Manning’s equation, Q = (1/n) A R^⅔ S^½; the district sees deliveries per gate instead of estimates.

Lining (Manning n)

Discharge Q
0.97 m³/s
Mean velocity
0.79 m/s
Hydraulic radius
0.383 m
Volume per day
84.2 ML
Area irrigated at 6 mm/day
1,403 ha

05 · Livestock and rangeland

Scenario

A collar that lasts the season.

Position collars and water-trough sensors over low-power radio. Battery life is set by airtime: the spreading factor that reaches the gateway, the payload, the GNSS fix and how often each collar reports.

Spreading factor

Time on air per report
185 ms
US915 dwell limit, 400 ms
Within
Average current
377 µA
Battery life
941 days · 2.6 years
Reports per collar per day
48

06 · Index insurance

Scenario

Density decides the basis risk.

Station density and tamper evidence for index insurance. An index pays on the station’s reading, not the farm’s loss. The farther the farm from its station, the larger the gap between the two, so the network is sized by distance to the nearest station: on a square grid the mean is 0.383 times the spacing. Every reading carries the station’s identity and calibration, so tampering shows.

Station spacing (square grid)
16.7 km
Mean distance, farm to nearest station
6.4 km
Farthest point from a station
11.8 km
Area per station
278 km²

07 · Proof

Evidence

What the buyer checks.

Each item a cooperative, a ministry or an insurer asks for is produced by the grid as a record: calibration, uptime, battery state and data terms travel with every value.

Show the record for

Asked for
Agronomic model validation
Record kept
Model run against held-out station data, error per crop and season

—Proof

What buyers in agriculture ask to see.

The grid produces each item as a record with the data it supports, not a report written afterwards.

01Agronomic model validation

02sensor accuracy specification

03battery life

04a local-language app

05data-ownership terms

Bring us the site.

Tell us the site, the sector and what you need to observe. We bring the survey, the design and the crew.

Talk to usSector overview