How a California orchard's water use is measured, signed and sponsored

The question sponsors ask

What can our company say about the water program it funds, and what record supports that sentence? For a corporate sustainability lead, that question connects the orchard to the company’s reporting. For counsel, it determines which description is supported by the evidence. A comparison with a county figure, a change in pumping against the farm’s own history, and an independently reviewed benefit answer different questions.

AgWaterAI’s groundwater program keeps those questions separate. Each number has a label, a season and a record. A sponsor can read the label before deciding whether the number fits its reporting needs. The benchmark number can describe a completed season already on file. A measured reduction needs the project season and the required history. A validated benefit needs an independent review after measurement.

The distinction matters while the program is being set up. A description of the planned process does not mean a sponsor has signed or a volume has been validated. This article explains the evidence, the boundaries of each number, and the steps still ahead as of October 7, 2026. The sponsor’s own counsel reviews any public claim.

The three numbers and their labels

The program uses these labels verbatim. They travel with the number so that a reader can distinguish a performance comparison from a reduction and from a validated benefit. A record for one block and season does not establish results for other blocks or seasons.

1. Benchmark efficiency: Efficiency against a published benchmark, for the farm's own fields. A performance claim, not a reduction.
2. Measured reduction: Measured reduction against the farm's own weather-adjusted history. Measured and signed by AgWaterAI, not independently validated.
3. Validated benefit: Validated volumetric benefit under VWBA 2.0, baseline capped at the allocation, independent validator named.

Where the numbers come from

The starting point is the farm’s own record of water use. Well meters provide measured pumping. Where a block has pump run-time records and an application rate on file, those inputs support an estimate of applied water. The record must say estimated. A run-time estimate should not be described as a flow-meter reading. A full season is needed for an annual benchmark comparison; a partial season does not stand in for it.

Soil moisture probes, pressure chamber readings and weather add the field context the adviser uses. The probes show changes in soil moisture on the blocks. Pressure chamber readings give a plant-based view of water stress. Weather puts those readings and water use in seasonal context. These inputs serve different purposes: a soil reading is not a well meter, and a pump record is not a reading of the tree’s condition.

For measured reduction, the method starts with three or more prior metered seasons and scales each to the assessment season’s weather using CIMIS reference evapotranspiration. The comparison asks what the farm’s own past seasons would have pumped under the assessment season’s weather. It does not substitute the county benchmark for the farm’s history. Where the required history is missing, the method identifies fallback evidence and the number waits.

That distinction also applies to readiness for 2027. Having enough evidence for a benchmark comparison today does not automatically establish a measured-reduction baseline. The block’s history, measurement basis and attribution of pumping to the enrolled acres need to support the number being requested. A sponsor should read those conditions alongside the season, rather than treating the availability of any record as availability of every label.

What the signed record does for a sponsor

A signed record gives a sponsor a specific item to cite and a public place where a reader can check it. Each record identifies its label, season and method version. The public lookup at agwaterai.com/verify/unit re-checks the signature on every lookup. The signing method is patent-pending. Its purpose is to let a reader check that the record has not changed since it was signed.

Signing does not establish measurement accuracy, regulatory compliance or independent validation. Those questions depend on the measurements, the accounting conditions and the review appropriate to the label. Keeping that boundary clear helps a sponsor explain what it funded without attributing more authority to a signature than it carries. A signed benchmark comparison remains a benchmark comparison.

Public records show region, crop, season and acres rounded to the nearest 10, never the farm. The public description lets a reader understand the scope without identifying the orchard. The grower keeps raw data. The evidence behind a proposed sponsorship and the block identities can be shared under a confidentiality agreement with the grower’s authorization. Public lookup and sponsor diligence therefore answer different levels of detail about the same record.

The benchmark and its source

The county applied-water estimates come from the California Department of Water Resources. The citation in the program’s benchmark table is: California Department of Water Resources, Statewide Agricultural Water Use Data 2016-2020 (Excel application tool), county sheets Regional_AW_Unit_County and Regional_ETc_Unit_County, 2024. The source link appears below. The county sheet averages water years 2016 to 2019 even though the tool’s title says 2016-2020. These are modeled county estimates, not measurements taken from a neighboring farm.

The following values and crop descriptions come directly from the program’s benchmark table. Each figure is applied water in acre-feet per acre. Citrus is part of a broader DWR category that includes olives and avocados; almonds and pistachios share a category. The crop category matters when a sponsor interprets the comparison.

1. Kern: Almonds & Pistachios (DWR crop category), 3.96 acre-feet per acre.
2. Kern: Citrus & Subtropical (DWR crop category), includes olives and avocados, 4.05 acre-feet per acre.
3. Tulare: Almonds & Pistachios (DWR crop category), 3.94 acre-feet per acre.
4. Tulare: Citrus & Subtropical (DWR crop category), includes olives and avocados, 3.71 acre-feet per acre.
5. Fresno: Almonds & Pistachios (DWR crop category), 4.04 acre-feet per acre.
6. Fresno: Citrus & Subtropical (DWR crop category), includes olives and avocados, 3.53 acre-feet per acre.
7. Kings: Almonds & Pistachios (DWR crop category), 3.88 acre-feet per acre.
8. Kings: Citrus & Subtropical (DWR crop category), includes olives and avocados, 3.79 acre-feet per acre.
9. Madera: Almonds & Pistachios (DWR crop category), 4.00 acre-feet per acre.
10. Madera: Citrus & Subtropical (DWR crop category), includes olives and avocados, 3.46 acre-feet per acre.

How to read the comparison

A comparison against the published county benchmark is a performance comparison, not a reduction. It describes how the farm’s own water figure for a complete season relates to that published figure. It does not establish that sponsorship changed pumping, that the farm pumped less than in its own past seasons, or that a volumetric benefit has passed independent review.

The basis must match as well as the crop and county. Applied water and crop evapotranspiration are different quantities in the benchmark table. An applied-water comparison uses the applied-water row. A record from the GSA’s satellite ETAW, crop ET less the rain the crop consumed, uses the ETAW row and is never compared with applied water. Each record states its source. The source, year and basis belong beside the number so the reader can see what was compared. The method page preserves those citations and qualifications, including the county-sheet period and the regional qualification for Kern.

For the 2026 season, the benchmark route is available where a complete season is on file. Pump run-time with the block application rate produces a labelled estimate; flow meters elected with the GSA support a metered figure. In Madera County the 2026 records use the GSA’s own satellite ETAW, on the calendar year, final in January. Availability now therefore means a route supported by completed evidence, with the measurement basis disclosed, rather than a claim that every enrolled block has the same readiness.

What independent validation adds

Independent validation adds a review by a validator who is named on the record when the review is complete. VWBA 2.0 is the accounting guidance for the volumetric benefit. Citing guidance is not the same as completing validation. Until the independent review is complete, the measured reduction retains its label and the words not independently validated.

For the validated benefit, we propose capping the baseline at the lower of the farm’s history and its GSA allocation, subject to technical and assurance review. A pumping cut the GSA already requires does not count. The method also addresses deductions and water carried over and pumped later. These controls determine the scope of the candidate benefit; a signature does not replace them.

The sponsor should distinguish measurement during the 2027 season from validation of that vintage after the season. The vintage identifies the season the record concerns, not the date someone happens to read it. A completed review is required before the benefit takes the validated label. As of October 2026, no independent validator has been appointed and no volume has been validated.

The grower side

The grower keeps allocation and water rights. The sponsor funds the program’s monitoring and evidence; the program transfers no water to the sponsor. The grower receives a share of sponsor payments, net of validation costs. Nothing is promised until a sponsor signs. The two grower promises for the benefit period are:

1. No sale or transfer of the water left in the ground, for the benefit period.
2. No added irrigated acres on the enrolled farm unit, for the benefit period.

One claim and an honest zero

The program permits one volumetric claim per block and season, by the sponsor that funded it. The grower does not sell or transfer the water behind a recorded reduction. Public programs are screened: acres already paid for through fallowing or land repurposing are excluded for the years paid. Proposed project controls screen GSA transfers, carryover, later pumping, regulatory obligations and overlapping funded activities; final treatment follows the accepted project method and the applicable GSA account rules.

A zero or negative season stays in the record as a zero, and no volumetric claim is issued for it. Measured volume is paid when measured. A benchmark comparison can still describe the season’s performance on its own evidence; it cannot turn a zero reduction into a volumetric benefit. Sponsor pricing is set once sponsors give their indications.

How this differs from sensor and data platforms

Arable, for example, sells an in-field sensor (Mark 3) with apps that measure “in-field rainfall, crop water use, and irrigation data” and markets “Arable Water” to growers and enterprises (source: arable.com, accessed October 7, 2026).

AgWaterAI adds the crop adviser’s call, the comparison against the published county benchmark, a signed record per block and season with a public lookup, and the three-label program. These describe the work and reporting structure of the groundwater program. A sponsor considering a project can use those distinctions to identify the evidence and the type of statement it needs.

How this differs from a promise-based program

A promise-based approach can start with a forward promise of a percentage saving against a modeled regional baseline, checked by the issuer itself. The AgWaterAI benchmark route starts with a measured past season from the farm’s own records, with run-time estimates labelled as estimates. Its county comparison is explicitly labelled as performance, not reduction.

The measured-reduction route uses the farm’s own weather-adjusted history and the project season. Independent validation is the additional step for the validated volumetric claim. These distinctions concern the source of the number, its timing, its label and who reviews it. They do not establish a future result. The 2027 measurements and the subsequent review still have to happen.

The five steps from field records to sponsorship

The proposed process connects the evidence, the adviser’s work and the sponsor’s reporting. Each stage keeps the measurement basis and label attached to the record.

1. The farm’s wells: Gather a full season of the farm’s own pump records or well meters. Label run-time estimates as estimated.
2. Sensors and readings: Use soil moisture probes, pressure chamber readings and weather. The crop adviser makes every irrigation call.
3. Signed record: Prepare a labelled, signed record per block and season, with a public lookup. The signing method is patent-pending.
4. Independent check: A named independent validator reviews the candidate benefit using VWBA 2.0 accounting guidance before it is described as validated.
5. Sponsor payment: Sponsor pricing is set once sponsors give their indications. Measured volume is paid when measured. Nothing is promised until a sponsor signs.

Program status and next steps

Method v1.1, October 7, 2026 is the dated method for this description. A sponsor can discuss the 2026 benchmark records now and plan for the 2027 season where the block’s evidence supports that path. The discussion should identify the desired label, basin, acres and available history before describing any prospective benefit. The method page and sponsor questions provide the next level of detail.