SeasoNpowered by AvlingAI

Variable-rate nitrogen from soil, landscape water, this season's weather and your N plan — compared with KdN V3.3.1
BoundaryTerrainSoil testYield

1Field boundary

Shapefile (.zip, or .shp + .dbf + .prj together), GeoJSON or KML.

2Terrain analysis

The buffer pulls in ground outside the boundary so water running onto the field is counted.

3Soil test

Point shapefile (.zip), CSV with latitude/longitude, GeoJSON or KML. CSV template

4Yield

Points only: averages raw yield points over this radius before the Rx is built (0 = off).

5Zone & field

Zone is suggested from the field location when the boundary loads. Suggest again
lb N/ac. Other N = starter, MAP/DAP. Soil nitrate is credited by zone (full in B, C, D; half in A and E).

6N plan & season weather

Timing and product split of the total N. Presets: Spring NH3 · Fall NH3 · Fall NH3 + inhibitor · Split UAN 40/60 · Starter + sidedress
Date (MM-DD)Product% of N
Observed weather since Oct 1, then 30+ past seasons for the rest of the season. Losses are compared with typical ground, a typical year and spring preplant NH3 — the conditions V3.3.1 was tuned under.

7Economics & output

Low-zone guard: N demand never drops below this % of the field-average yield. MRTN pull shifts the whole field toward the university rate and keeps the spatial pattern.
No terrain built: every cell is treated as typical ground (W = 0).
Hover for values · click a cell for its breakdown

Cell breakdown

Soil samples — rate at each sample

Export

Shapefile fields: N_RATE (lb N/ac), PROD_RATE, V331, DIFF, YIELD, CEC, PH, OM, W, LOSSADJ, SATDAYS, OMCREDIT, GROSS, TWI, TPI, ELEV. WGS84.
Season-N · field-testing build. Soil and yield are interpolated by inverse distance; terrain is built from USGS 3DEP lidar or yield-monitor elevation. Weather, the season simulation and the rate model run on AvlingAI servers. Loss rates, zone coefficients and the price adjustment are starting values being calibrated with strip trials.