Altitude Migration Tracker

In development. This tracker has no published figures. The method is published first, ahead of any data, so that it can be examined and contested before a single value is released.

What it measures

Indian agricultural statistics do not report elevation. Area, production and yield are recorded by district and by block; the height at which a crop is actually grown is absent from the record.

For most crops this does not matter. For a perennial tree crop with a fixed chilling requirement, planted once and held for two decades, it is the variable that determines whether the asset works.

The Orchard Elevation Register introduces elevation as a reported dimension. For any 250-metre band in any apple-growing district of the Indian Himalayan Region, it is built to answer five questions:

• Is orchard area in this band expanding or contracting?

• Is yield rising or falling?

• Is winter chilling still meeting the requirement of the varieties planted here?

• Is land in this band transacting at rising or falling value?

• Is replanting credit flowing into this band, or out of it?

None of these can be answered from public data today. Making them answerable is the whole of the contribution.

Coverage

Version 1 — Himachal Pradesh, complete. Shimla, Kullu, Mandi, Kinnaur, Chamba, Sirmaur, Solan, Lahaul & Spiti and Kangra. Reported at panchayat level, aggregated to elevation bands.

Subsequent phases extend to Uttarakhand, Jammu & Kashmir and Arunachal Pradesh.

Banding. 250-metre bands from 900 m to 4,000 m. The published record of apple cultivation shows movement of several hundred metres per decade. Bands of 500 m would conceal a decade of that movement inside a single unit; bands of 100 m would claim a precision the underlying area data does not support. 250 m is the interval at which the shift is visible and the data still holds.

Series start. Orchard area and chilling accumulation from 2000, the mid-point of the documented elevation shift. Production from 2010, the limit of departmental data that can be reliably reconciled. Land transactions and procurement from 2015, the practical limit of digitised revenue records.

The four layers

Orchard Register

Area, net change, new establishment, removal, age profile, variety mix, plantation system and yield — each reported by elevation band.

The register's hardest measurement is orchard removal. Abandonment is not reported by any agency, in any state, at any resolution. It has to be detected rather than collected. It is also the clearest single signal that a band has stopped working, which makes it the most valuable output the register can produce and the one it is being built around.

Chilling Adequacy

Accumulated hours below 7°C through December to February, interpolated from station records to elevation band, measured against the requirement of the varieties actually planted in that band.

The requirement figure is contested in the published literature. Stated thresholds for traditional varieties vary widely between sources, and early varieties are given substantially lower figures again. The tracker will not adopt one number and present it as settled. It publishes a reconciled table setting out each source and its basis, and where the sources cannot be reconciled, reports adequacy as a range rather than a point.

Bands falling below 300 accumulated hours are flagged separately. That threshold is used because it is the level at which the state's own horticultural university has recommended variety substitution.

Value Transfer

Agricultural land transaction values and volumes, horticulture credit disbursed, high-density scheme uptake and crop insurance enrolment — all by band.

The layer's principal output sets credit disbursement against chilling adequacy and answers one question: is replanting capital flowing into elevation bands whose viability is deteriorating? A high-density orchard loan is a fifteen-year commitment to a fixed parcel at a fixed altitude. If that capital is concentrating in contracting bands, it is a portfolio exposure that no lender currently has the instrument to see.

Privacy. Land values and transaction counts are published only as band-level aggregates above a minimum cell size. No parcel, khasra number, owner name or village-level value is published, and any output permitting re-identification is suppressed. The privacy protocol is published in full alongside the first data release and applies from that release onward.

Market Structure

Market intervention scheme volumes and rates, procurement centres, corporate announced rates, mandi prices, controlled-atmosphere storage capacity, import volumes and landed cost, applied duty, and carton compliance.

Procurement rates are reported in real terms as well as nominal. A rate held flat through a high-inflation, low-volume year is a reduction, and a series that reports only the nominal figure conceals it.

Before any figure is published

The public record on Himalayan horticulture carries inconsistencies that would otherwise pass straight into the tracker. Four are resolved before the first release, and the resolutions are published as a standing document rather than buried in an annexe:

Tonnes and boxes. The conversion assumption and its basis are stated openly. Box weight is itself unstable under non-standard carton practice; where a single conversion cannot be defended, both series are published separately and never merged.

Calendar and fiscal years. Tonnage figures commonly appear against calendar years and box figures against fiscal years. A period concordance table accompanies every series.

Hectares and acres. All area is normalised to hectares. At least one national report has carried an area figure with the units transposed.

Chilling requirement. As above.

Where a single source carries two different values for the same variable, both are recorded and flagged rather than silently reconciled.

Publication calendar

Once live, the tracker follows the horticultural year rather than a fixed editorial schedule.

March. Chilling accumulation for the completed winter. This is the anchor release. It lands before replanting decisions are taken and before intervention terms are set for the season, which is the point at which it is most useful to the people it concerns.

April. Chilling adequacy assessment and band-level flags.

June and July. Intervention scheme terms, analysed in nominal and real terms.

August to October. Procurement and price series, updated weekly through the season.

November. Full annual refresh.

December. Annual analytical paper.

How the data is presented

Every value carries its source. Bands where data is incomplete are shown as incomplete rather than estimated, and no band is classified where completeness falls below the published floor — a confident label on thin data is worse than no label at all. Missing data, data suppressed for privacy, and data withheld by its source are displayed distinctly and never collapsed into a single blank.

Every band has a permanent URL. Annual snapshots are versioned, so a citation made in 2029 resolves to what the tracker actually said in the year cited. The complete dataset is downloadable as CSV and JSON without restriction or registration.

Band summaries are published in Hindi alongside English. A tracker concerning grower land value published only in English has made a decision about whose it is.

Independence

HEF publishes the consulting, commercial and advisory relationships held by the institution and by the authors of any file, on every publication. Where a live commercial relationship touches a file, the author recuses. An independence claim survives disclosure; it does not survive discovery.

Corrections are published openly and every revision is versioned.