Open Data
Idaho agriculture and water policy, as data
Four datasets built from Idaho public records that their own sources publish one item at a time: what the Legislature did about agriculture and water across the 2019 through 2026 sessions, what the Department of Water Resources is doing about it docket by docket, how much water the system actually carried into each year, and what kills Idaho’s sheep and lambs set beside what one federal program recorded doing about it. The records themselves are public and always were, and the assembly of them is released into the public domain, so none of it needs permission to use.
The articles in the insights library are written on top of these files. Publishing the underlying tables means the work can be checked, and reused, without going through the prose.
Idaho Agriculture, Water, and Natural Resource Legislation, 2019 to 2026
Every bill, resolution, memorial, and proclamation addressing agriculture, water, or natural resources in the Idaho Legislature’s 2019 through 2026 sessions: 367 measures, 168 of them enacted, each carrying the Legislature’s own subject label, its final status, and the session law chapter number where one exists. A second table carries 3,498 recorded actions, an average of 9.5 per measure, which is what answers how long a measure took and which stage it stopped at.
- idaho-agriculture-legislation.json everything, plus coverage, method, caveats, and both field dictionaries
- idaho-agriculture-legislation.csv 367 measures, one row each
- idaho-agriculture-legislation-actions.csv 3,498 actions, joining on year and number
Data last changed .
How a measure gets included
A measure is in when the Legislature’s own subject label for it falls in a reviewed agricultural, water, or natural resource vocabulary. It is not keyword matching on titles. The label list is re-reviewed against every new session, because the Legislature relabels the same policy from year to year: the 2026 list run unreviewed against 2025 returns 22 measures where the reviewed list returns 58. Enacted sets are checked against the session law chapter sequence, which runs 1 to N with no gaps in all 8 sessions covered.
Measure fields
| Field | Meaning |
|---|---|
| year | Session year the measure was introduced in. |
| number | The Legislature's measure number, for example H0503, S1307, HCR010. |
| subject | The Legislature's own subject label for the measure, verbatim. Empty on the five rule-review concurrent resolutions, whose pages carry no subject label; they are in the dataset by explicit override and their description falls back to the measure title. A blank here is the absence of a label, never a parse failure. |
| description | The Legislature's short description of the measure, verbatim. |
| status | The Legislature's final status string for the measure. |
| outcome | Normalized final outcome, one of: enacted, adopted, vetoed, failed, died. 'adopted' covers resolutions and memorials, which are not enacted into code. 'failed' means it lost a recorded floor vote; 'died' means it never reached one. |
| chapter | Session law chapter number when the measure was enacted, otherwise null. Chapters run 1 to N within a session with no gaps, which is how the enacted set is checked for completeness. |
| url | The measure's page on the Idaho Legislature's site. |
| issue | Which of the five policy areas on lpagadvocacy.com the measure was sorted into: water-rights-supply, land-public-lands, production-commodities, wildlife-depredation, or legislative-regulatory. Every measure appears in exactly one. |
| sponsor | Who introduced the measure, as the Legislature records it. Overwhelmingly a standing committee rather than a named legislator: across the 2026 session only 2 of 70 agricultural measures named an individual. Treat this as the originating committee, not as a measure of which legislators champion agriculture. |
Action fields
| Field | Meaning |
|---|---|
| year | Session year, joining to the measures table with `number`. |
| number | The measure's number, joining to the measures table with `year`. |
| seq | 1-based position of this action in the measure's own list, preserving order. |
| date | Date of the action, ISO YYYY-MM-DD. The Legislature publishes MM/DD with no year; the session year supplies it, which is safe because Idaho sessions convene in January and adjourn the same spring. Empty if the source date did not parse. |
| action | The action text as the Legislature records it, verbatim. |
| url | The measure's page on the Legislature's site. Repeated from the measures table on purpose: a row lifted out of this file on its own would otherwise have no route back to its source, and the published terms ask you to verify against the original. |
Coverage by policy area
| Policy area | Measures | Enacted |
|---|---|---|
| Water Rights & Supply | 144 | 82 |
| Land & Public Lands | 87 | 24 |
| Production & Commodities | 83 | 45 |
| Wildlife & Depredation | 26 | 10 |
| Legislative & Regulatory Process | 27 | 7 |
Idaho Water Administration: IDWR Delivery Calls, Mitigation Plans, and Legal Matters
Every legal matter the Idaho Department of Water Resources publishes, with its dated filing trail: 242 matters, 37 of them active, among them 19 delivery calls and 48 mitigation plans, and 10,969 individual filings. A bill changes the rules for next season. A delivery call changes what a farm may divert this one, and a mitigation plan is what lets a junior right keep pumping.
- idaho-water-administration.json everything, plus coverage, method, caveats, and both field dictionaries
- idaho-water-administration.csv 242 matters, one row each
- idaho-water-administration-filings.csv 10,969 filings, joining on matter
IDWR read on . Active and archived describe which index each matter sat on that day, not its posture today.
These are real disputes involving named parties, several of them live. The data carries docket status, filing titles, and filing dates exactly as IDWR publishes them, verbatim.
Matter fields
| Field | Meaning |
|---|---|
| matter | Stable key for the matter, taken from the last segment of its IDWR URL. Unique across all 242. Use this to join, not `docket`: 45 matters carry no docket number and one docket number appears on two matters. |
| type | IDWR's own category: Delivery Call, Mitigation Plan, Administrative, District Court, Supreme Court, or Settlements. |
| docket | Docket number as IDWR lists it. Empty string when IDWR lists none. |
| docketLine | The docket line from the matter's page, verbatim. Sometimes names two dockets, for example 'Docket Nos. CM-DC-2010-002 & CM-DC-2010-003'. |
| title | The matter's title as IDWR lists it. |
| status | Which IDWR index the matter appeared on **as of the fetchedAt date**, not a claim about its posture today: 'active' from the legal-actions index, 'archived' from the archived-matters index. This is the one field here that is derived rather than scraped, so it is deliberately time-scoped. IDWR decides which index a matter sits on, and a matter can stay on the active list for many years, which reflects how long conjunctive administration takes rather than a backlog. |
| url | The matter's page on IDWR's site. |
| filingCount | Number of filings parsed from the matter's page. |
| firstFiling | ISO date of the earliest parsed filing, ignoring dates that postdate publication. Empty when no filing trail parsed. Genuine filings here run back to 1940: the WD63 contested case carries Army Corps letters from 1946 as exhibits. |
| lastFiling | ISO date of the most recent parsed filing, ignoring dates that postdate publication. Two of IDWR's published dates are typos (December 3 2104 and October 27 2105); they appear verbatim in the filings table but are excluded here so they cannot put a 2015 case at the top of a sort by recent activity. |
Filing fields
| Field | Meaning |
|---|---|
| matter | Joins to the matters table on `matter`. |
| seq | 1-based position in IDWR's own listing for that matter, which runs newest first. |
| date | ISO date of the filing, as IDWR published it. Two records carry a typo in the source and are reproduced rather than corrected: an Administrative Bulletin 'Volume 14-12' dated 2104-12-03, and a procedural order on the 2015 case CV-2015-1687 dated 2105-10-27. Both are excluded from the matters table's firstFiling and lastFiling. Filter on year if you are computing a range. |
| title | Filing title as IDWR publishes it, verbatim. Titles name parties and describe the document; they are a record of what was filed and when. |
| url | Direct link to the filed document on IDWR's site, present for 9,819 of the 10,969 rows. Empty where IDWR lists a filing without linking one. This is the row's own primary source: the published terms ask you to verify against the original, and this is that original rather than the matter page it was listed on. |
Matters by type
| Type | Active | Archived | Filings |
|---|---|---|---|
| District Court | 4 | 89 | 2,776 |
| Administrative | 10 | 44 | 2,648 |
| Mitigation Plan | 15 | 33 | 1,499 |
| Supreme Court | 0 | 22 | 762 |
| Delivery Call | 4 | 15 | 2,851 |
| Settlements | 4 | 2 | 433 |
Idaho Irrigation Carryover Storage and the Water Administration Record, 1990 to 2025
Carryover storage on 30 September, the practical close of Idaho’s irrigation season, for the ten reservoirs the Upper Snake and Boise systems run on: 360 station-years from 1990 to 2025. A second table sets each year’s storage beside IDWR delivery call, mitigation plan, and filing activity. The driest carryover in the record is 1992 at 29 percent of the series median.
- idaho-water-supply.json everything, plus sources, method, caveats, and both field dictionaries
- idaho-water-supply.csv 360 station-years, one row per reservoir per year
- idaho-water-supply-years.csv 36 years, storage totals beside IDWR activity
Data last changed .
This table sets water supply beside administrative activity, and filing volume tracks litigation as much as hydrology: 2015 sits near the median for storage and carries the highest filing count in the record, because the Rangen and Surface Water Coalition matters were both active.
The storage series runs from 1990 but the filing record thins sharply before 2005. Annual counts run between zero and twelve through 2000, then 33 in 2001 and 51 in 2004, reaching 298 in 2005; from 2005 on they range between 98 and 484. That is a change in what the department publishes, not a change in what happened, so 2005 onward is the joinable window. The driest year in the whole series is 1992 at 29 percent of median, and it shows two filings.
Storage fields
| Field | Meaning |
|---|---|
| year | Calendar year of the 30 September reading. |
| station | Reclamation Hydromet station code. |
| reservoir | Reservoir name exactly as Reclamation's own station list gives it. The puller re-resolves every code against that list on each run and refuses to write if a name has drifted, because a code silently repointing is the one error that would put a wrong reservoir on a published figure. |
| system | 'Upper Snake' or 'Boise'. The two systems Idaho irrigation runs on, and the ones the conjunctive administration record turns on. |
| state | State the reservoir sits in. Jackson Lake and Grassy Lake are in Wyoming and are included because they are Upper Snake storage serving the Eastern Snake Plain. |
| acreFeet | Contents in acre-feet on 30 September, as Reclamation publishes it. There is no per-row source link because the source is a Hydromet query rather than a page: `station` plus the query documented under `method` reproduces any row exactly. The two tables in this dataset are the only ones here without a url column. |
Water year fields
| Field | Meaning |
|---|---|
| year | Calendar year. |
| upperSnakeAcreFeet | Sum of 30 September contents across the seven Upper Snake reservoirs. |
| boiseAcreFeet | Sum of 30 September contents across the three Boise system reservoirs. |
| totalAcreFeet | Both systems combined. |
| percentOfMedian | totalAcreFeet as a percentage of the median totalAcreFeet across every year in this series. A normalisation for reading the series, not an official Reclamation figure. |
| idwrFilings | Count of IDWR filings dated in this year across every matter in the water administration dataset. Tracks litigation activity as much as water supply; a busy year can mean an active docket rather than a short one. **Thin before 2005**, which reflects how far back IDWR publishes filings rather than what happened: the count runs 0 to 12 a year through 2000, then 33 in 2001 and 51 in 2004, and 298 in 2005. |
| deliveryCallsFirstFiled | Delivery call matters whose earliest published filing falls in this year. A proxy for when a call was brought, not an official docket-opening date: it is the earliest filing IDWR shows on the matter's page, which can postdate the docket itself. |
| mitigationPlansFirstFiled | Mitigation plan matters whose earliest published filing falls in this year, on the same proxy basis as deliveryCallsFirstFiled. |
Caveats that ship with the data
- The storage series runs from 1990 but the IDWR filing record thins sharply before 2005. Annual filing counts run between 0 and 12 through 2000, then 33 in 2001, 10 in 2003 and 51 in 2004, reaching 298 in 2005; from 2005 on they range between 98 and 484. That is a change in what the department publishes, not a change in what happened. Do not read the near-zero counts of the 1990s as evidence that dry years produced no administrative response: 1992 is the driest year in the series at 29 percent of median and shows two filings. Treat 2005 onward as the joinable window.
- This table sets water supply beside administrative activity, and filing volume tracks litigation as much as hydrology: 2015 sits near the median for storage and carries the highest filing count in the record, because the Rangen and Surface Water Coalition matters were both active.
- 30 September is a reasonable proxy for the close of the irrigation season, not an official Reclamation accounting date.
- deliveryCallsFirstFiled and mitigationPlansFirstFiled count matters by their earliest published filing, which can postdate the opening of the docket itself.
- Jackson Lake and Grassy Lake are in Wyoming. They are included as Upper Snake storage serving the Eastern Snake Plain. Lake Owyhee is a separate project and is excluded.
- percentOfMedian is computed against this series' own median and is not a Reclamation or NRCS published normal.
Idaho Sheep and Lamb Death Loss, Its Causes, and Federal Predator Take, 1988 to 2025
Idaho sheep and lamb deaths from all causes for 38 years, from 1988 to 2025, joined to the cause-of-death breakdown NASS surveys at intervals: 4 survey years and 64 cause rows separating predator from non-predator loss in head and in dollars. Predators were 23.5 percent of all Idaho sheep and lamb deaths in 2016 and 40 percent in 2025, while deaths from all causes fell from 28,500 head to 22,000.
The cause breakdown is the reason this file exists. It is not in Quick Stats. It is surveyed at the request of, and with funding from, the Idaho Animal Damage Control Board, and published as standalone reports that have never been assembled into a series.
- idaho-sheep-death-loss.json everything, plus sources, method, caveats, and all five field dictionaries
- idaho-sheep-death-loss.csv 38 years, deaths against inventory and lamb crop
- idaho-sheep-death-loss-causes.csv 64 cause rows across 4 survey years
- idaho-sheep-death-loss-wildlife-services.csv 954 rows, USDA Wildlife Services activity by species and method, FY2014 to FY2024
Data last changed .
The two layers count different things and must not be compared directly. The annual series counts lamb deaths after docking only; the cause surveys also count lambs lost before docking. In 2025 the annual series reports 15,000 deaths and the survey reports 22,000. Both are correct on their own definition, and the difference is not a revision.
Wolves are a separate line in 2016, 2018 and 2023 and are not in 2025, where NASS folds them into Other/Unknown Predators with bobcats. The 2025 absence is a reporting change, not a fall to zero. The predator share is also a floor rather than a measurement, because NASS states that Unknown Causes, a quarter of all 2025 deaths, may include some predator loss.
The third layer: what one federal program recorded doing
USDA APHIS Wildlife Services publishes what it did, by species and by method, in Program Data Report G. This is 954 Idaho rows covering FY2014 to FY2024. It sits beside the loss layers and is deliberately not joined to them: it runs on the federal fiscal year while the loss data is calendar year, and it covers one program’s own actions, so private trapping, county pest-district work under Idaho Code 25-2601, and landowner take are all outside it.
Read the species rows, not the total. European starlings are 97 percent of everything Wildlife Services killed in Idaho in FY2024, under a separate agricultural bird program, and 97.4 percent across the eleven years. Quoting the all-species total as a predator-control figure overstates it by roughly two orders of magnitude. In FY2024 the agency recorded 3,024 coyotes and 21 wolves.
Annual fields
| Field | Meaning |
|---|---|
| year | Calendar year. |
| sheepDeaths | Sheep excluding lambs lost to all causes, as NASS publishes it in Quick Stats. |
| lambDeaths | Lambs lost to all causes. **After docking only.** The cause surveys in this dataset also count lambs lost before docking, which is why their totals run higher. |
| totalDeaths | sheepDeaths plus lambDeaths. Computed here, not a published NASS figure. |
| inventoryJan1 | Sheep and lambs on hand 1 January of the same year, for scale. |
| lambCrop | Lamb crop for the year, for scale. |
| deathsPer1000Inventory | totalDeaths per 1,000 head of 1 January inventory. A normalisation computed here so the series can be read against a flock that has changed size, not an official rate. |
Survey fields
| Field | Meaning |
|---|---|
| year | Calendar year the survey covers. |
| totalHead | All sheep and lamb deaths from all causes, including lambs lost before docking. |
| totalValueDollars | Published value of all losses. |
| predatorHead | Deaths NASS attributes to predators. |
| predatorValueDollars | Published value of predator losses. |
| nonPredatorHead | Deaths NASS attributes to non-predator causes. |
| nonPredatorValueDollars | Published value of non-predator losses. |
| wolvesReported | Whether wolves are a separate line that year. True for 2016, 2018 and 2023; false for 2025, where wolves are inside Other/Unknown Predators with bobcats. **False does not mean zero.** |
| roundingResidualHead | Published total minus the sum of the published cause rows, in head. NASS rounds each cause to the nearest 100 and does not force any column to add; its own reports say 'Totals may not add due to rounding'. 2023 carries a residual of 200 head. |
| roundingResidualValueDollars | The same residual in dollars, and it can be negative because the rows can exceed the published total: 2023 is $2,000 over and 2025 is $1,000 over. Published rather than absorbed, so that a total shown here can always be checked against the rows shown here. |
| predatorSharePercent | predatorHead as a percentage of totalHead. Computed here. A floor rather than a measurement, because Unknown Causes may include predator loss. |
| predatorPer1000Inventory | predatorHead per 1,000 head of 1 January inventory that year. Computed here, so the series can be read against a shrinking flock. |
Cause fields
| Field | Meaning |
|---|---|
| year | Calendar year the survey covers, not the year it was released. |
| cause | Cause of loss exactly as the report names it, with one normalisation: NASS has used 'Other Predators', 'Unknown Predators' and 'Other/Unknown Predators' across years and all three are recorded as Other/Unknown Predators. 'Mt. Lions' is recorded as Mountain Lions. |
| predator | Whether NASS counts this cause under Total Predators. Note that Unknown Causes is not a predator cause here, and NASS states it 'may include some predator loss'. |
| head | Head lost to this cause. NASS rounds to the nearest 100. |
| percentOfAllDeaths | This cause as a percentage of all deaths that year, as published. |
| valueDollars | Value of the loss in dollars. The reports publish thousands of dollars; this is that figure multiplied by 1,000. |
Wildlife Services fields
| Field | Meaning |
|---|---|
| year | Federal fiscal year, October through September. **Not** the calendar year the loss layers use. |
| species | Species as Wildlife Services names it. 'Coyotes (Burrow/Den)' is a den-destruction row, counted under burrowsRemoved rather than killed. |
| invasive | 'IS' if Wildlife Services classes the species as invasive, 'Non-IS' otherwise. |
| method | The capture, restraint, or targeting method. APHIS states this 'MAY NOT have been the method or tool used to kill, euthanize, or free/relocate' the animal. Do not read it as cause of death. |
| intent | 'Intentional' or 'Unintentional'. Both are counted, so a species can appear because it was targeted or because it was taken by something aimed at another species. |
| killed | Animals killed or euthanized. |
| burrowsRemoved | Burrows or dens removed or destroyed. Counts structures, not animals. |
| freed | Animals freed, released, or relocated. |
| dispersed | Animals dispersed rather than taken. |
| acresTreated | Acres treated, where the action is measured by area. |
Wildlife Services, by year
| Field | Meaning |
|---|---|
| year | Federal fiscal year. |
| allSpeciesKilled | Everything Wildlife Services killed or euthanized in Idaho that year, every species. |
| starlingsKilled | European starlings killed, a separate agricultural bird programme. |
| starlingSharePercent | Starlings as a percentage of allSpeciesKilled. It runs between 94 and 98 percent every year, which is why allSpeciesKilled must never be quoted as a predator-control figure. |
| livestockPredatorsKilled | Killed across the species in LIVESTOCK_PREDATORS. Computed here. |
| coyotesKilled | Coyotes killed or euthanized, all methods. |
| coyotesKilledAerial | Coyotes whose recorded capture or targeting method was fixed wing or helicopter. |
| aerialSharePercent | coyotesKilledAerial as a percentage of coyotesKilled. Computed here. |
| wolvesKilled | Gray wolves killed or euthanized by Wildlife Services. Not all wolf take in Idaho. |
Caveats that ship with the data
- The two layers count different things and must not be compared directly. The annual series counts lamb deaths after docking only; the cause surveys also count lambs lost before docking. In 2025 the annual series reports 15,000 deaths and the survey reports 22,000. Both are correct on their own definition, and the difference is not a revision. Each survey report publishes the reconciling figure: lambs lost before docking were 10,500 in 2016, 11,000 in 2018, 7,500 in 2023 and 7,000 in 2025, and the annual figure plus that number is the survey total in every one of the four years.
- The cause surveys are irregular, not annual. Four Idaho years are published: 2016, 2018, 2023 and 2025. Anything between those years is interpolation, not observation, and nothing here should be presented as a continuous cause series.
- Wolves are a separate line in 2016, 2018 and 2023 and are not in 2025, where NASS folds them into Other/Unknown Predators with bobcats. The 2025 absence is a reporting change, not a fall to zero.
- The predator share is a floor, not a measurement. NASS states that Unknown Causes, which is 25.5 percent of all 2025 deaths, 'may include some predator loss'.
- NASS rounds each cause to the nearest 100 and does not force any column to add. In 2023 the head column is 200 short of its own published subtotal, the value column is $2,000 over it, and the percent column is 0.1 point short; 2025 carries a $1,000 value residual. Both residuals are published per year, as roundingResidualHead and roundingResidualValueDollars, rather than absorbed.
- The take layer and the loss layers are on different year bases and must not be joined row for row. Wildlife Services reports on the federal fiscal year, October through September; NASS death loss is calendar year. They are published side by side here so both can be read, not so one can be divided by the other.
- The take layer covers one federal program's own actions and nothing else. Private trapping, county pest-district work under Idaho Code 25-2601 through 25-2604, permit holders, and landowner take are all outside it. It is not a measure of predator control in Idaho, and Idaho's own wolf take is reported separately by Idaho Fish and Game.
- European starlings are 97.4 percent of everything Wildlife Services killed in Idaho across FY2014 to FY2024, under a separate agricultural bird programme. Quoting the all-species total as a predator-control figure overstates it by roughly two orders of magnitude. Use livestockPredatorsKilled, or the species rows, instead.
- The method column in the take layer is the capture, restraint, or targeting method. APHIS states it 'MAY NOT have been the method or tool used to kill, euthanize, or free/relocate' the animal, so it is not a cause of death. Take is also split intentional and unintentional and both are counted.
- Idaho cause data for earlier years exists inside the national Sheep and Goats Death Loss reports and is not pulled here. Those years are listed under unpulledSources so the gap is visible.
Terms, and what these files are not
- What this is
- An independent compilation of public records, assembled and published by LP Associates & Agriculture, LLC. It is not an official product of, endorsed by, or affiliated with any agency named as a source. Where this disagrees with the agency's own record, the agency's record governs.
- Accuracy
- Best effort, no warranty of any kind. These files are built by automated pullers from public web pages, and both the sources and the parsing can be wrong. Verify anything you intend to rely on against the linked original.
- Scope
- A summary of what the named sources published as of the date given, limited to agriculture, water, and natural resource matters. It is not a complete record of the Idaho Legislature, of Idaho water administration, or of any proceeding it references.
- Not legal advice
- Published for general information only, to explain public law and public process. It is not legal advice, does not create an attorney-client relationship, and is not a characterisation of the merits of any matter or the conduct of any person named in it.
- Corrections
- Corrections and takedown requests: info@lpagadvocacy.com. Errors are fixed at the source data and republished, so a correction reaches every distribution at once.
These terms travel inside each JSON distribution as a disclaimer object. The CSVs carry none of it: a comment preamble is not valid RFC 4180 and breaks strict parsers, so they stay clean and the terms live here.
Rights
There are none to clear. The records in all four datasets are Idaho and United States public record and were never restricted. The only right anyone could claim here is a thin one in how they were selected and arranged, and that is released Creative Commons Zero 1.0 Universal (public domain dedication). Use any of it, for anything, commercially or otherwise. No permission is needed and none is being granted, because there is nothing here to grant permission over.
Attribution is not required either. If it is useful to cite the assembly rather than the underlying records:
LP Associates & Agriculture, LLC. Idaho Agriculture, Water, and Natural Resource Legislation, 2019 to 2026, Idaho Water Administration: IDWR Delivery Calls, Mitigation Plans, and Legal Matters, Idaho Irrigation Carryover Storage and the Water Administration Record, 1990 to 2025, and Idaho Sheep and Lamb Death Loss, Its Causes, and Federal Predator Take, 1988 to 2025. August 17, 2026. https://lpagadvocacy.com/data/