{"id":"us-drought-indices","slug":"us-drought-indices","name":"United States Drought Index Dataset","description":"Four drought indices — the Palmer Drought Severity Index, the Palmer Hydrological Drought Index and the Standardized Precipitation Index at one and twelve months — monthly since 1895 for each of the 344 climate divisions of the contiguous United States.","methodology":"Collected from NOAA’s nClimDiv drought products, computed by NOAA from area-weighted gridded estimates interpolated from GHCN-Daily station data against a 1931–1990 reference period. The four indices are held as four separate metrics because they disagree with one another by design: PDSI answers whether the weather is dry and PHDI whether the water supply is, and the lag between them is a real state of the world rather than a discrepancy. Hydrionis computes no combined drought score, and none should be added. NOAA stamps a release date into every filename with no stable alias, so Hydrionis reads the directory and takes the newest file per index.","domains":["drought"],"metricIds":["pdsi","phdi","spi-1","spi-12"],"subjectScope":"The 344 climate divisions of the contiguous United States","cadence":"monthly","version":"1.0.0","released":"2026-08-27","changelog":[{"version":"1.0.0","date":"2026-08-27","summary":"First release.","additions":["PDSI, PHDI, SPI-1 and SPI-12 for 344 climate divisions, monthly from 1895."]}],"limitations":["These are calculated indices, not measurements. They inherit whatever the underlying interpolation got wrong, particularly in mountainous and sparsely gauged divisions.","Every value is an anomaly against 1931–1990. A value describes conditions relative to that baseline rather than to a recent normal, and in a shifting climate those diverge.","The Palmer indices were calibrated on the central United States and behave less well in snow-dominated and arid divisions, where the moisture balance they assume fits less closely.","Successive SPI-12 values share eleven months of input, so a run of low values is one dry year reported repeatedly rather than a run of independent dry periods.","Man-made changes — irrigation, new reservoirs, industrial use — are excluded from the Palmer computation, so these describe the natural balance and not the managed one."],"context":{"hydrionisClass":"water-driver","whyHeld":"Drought is the condition under which most of the water questions Hydrionis exists to answer become urgent, and it is not one thing. Holding four indices that disagree is the point: a place can be in meteorological drought and not hydrological drought, and knowing which is which is the difference between a dry spell and an emptying reservoir. These are indices computed from weather, not measurements of any river or aquifer.","waterRelevance":["Reservoir storage and drawdown","Low river flow and its persistence","Groundwater recharge failure","Irrigation and agricultural abstraction demand","Drought declaration and restriction decisions"],"suitableUses":["Placing a drought period in a long historical context for the same division","Contrasting meteorological with hydrological drought, where the divergence is the finding","Showing drought conditions alongside water data on a shared time axis"],"unsuitableUses":["Any statement about a specific reservoir, river or aquifer, which requires that water body’s own record","Combining the indices into a single drought score — the weighting would be an editorial judgement presented as a measurement","Comparing divisions in different climates on the Palmer indices, which are not equally calibrated across them","Reading a value as severe relative to today, rather than relative to 1931–1990"],"joinDimensions":[{"kind":"time-only","on":"Shared monthly periods with any other Hydrionis series","limitation":"A shared time axis supports showing drought beside water data and nothing more. Hydrionis holds no climate-division boundaries, so it cannot establish that a given reservoir lies in a given division, and does not claim to."}],"relationshipConfidence":"contextual","eligibility":{"api":true,"download":true,"overlay":true,"publicPage":false}},"observationCount":2168920,"seriesCount":1376,"subjectCount":344,"from":1895,"to":2026,"sourceIds":["noaa-climdiv"],"units":["index"],"redistributable":true,"commerciallyReusable":true,"citation":"Hydrionis (2026), United States Drought Index Dataset, version 1.0.0. Retrieved from https://hydrionis.com/data/datasets/us-drought-indices/","downloads":[{"file":"/downloads/observations-drought-noaa-climdiv.csv","format":"CSV","label":"drought observations from NOAA NCEI, part 1 of 3 (1,000,000 rows)"},{"file":"/downloads/observations-drought-noaa-climdiv-part2.csv","format":"CSV","label":"drought observations from NOAA NCEI, part 2 of 3 (1,000,000 rows)"},{"file":"/downloads/observations-drought-noaa-climdiv-part3.csv","format":"CSV","label":"drought observations from NOAA NCEI, part 3 of 3 (168,920 rows)"}],"apiPath":"/api/v1/datasets/us-drought-indices.json","lastFetched":"2026-08-27","fields":[{"id":"pdsi","name":"Palmer Drought Severity Index","definition":"Meteorological drought severity from a moisture balance of precipitation against demand, on a scale where negative is dry.","unit":"index","nativeGeography":["climate-division"],"comparabilityNotes":"One index among several that disagree by design. It must not be combined with the others into a single drought score, and a place not in drought by this measure may well be in drought by another. PDSI is also known to behave differently between climates — it was calibrated on the central United States and is less reliable in snow-dominated and arid regions, where the moisture balance it assumes fits less well."},{"id":"phdi","name":"Palmer Hydrological Drought Index","definition":"Hydrological drought severity: the same moisture balance as PDSI, but held until the deficit is genuinely ended.","unit":"index","nativeGeography":["climate-division"],"comparabilityNotes":"One index among several that disagree by design. It must not be combined with the others into a single drought score, and a place not in drought by this measure may well be in drought by another. PHDI lagging PDSI is the expected behaviour, not a discrepancy: a month where the weather has recovered and the water supply has not is a real state of the world and both indices are describing it correctly."},{"id":"spi-1","name":"Standardized Precipitation Index, 1 month","definition":"How unusual one month’s precipitation was for that division and that calendar month, in standard deviations.","unit":"index","nativeGeography":["climate-division"],"comparabilityNotes":"One index among several that disagree by design. It must not be combined with the others into a single drought score, and a place not in drought by this measure may well be in drought by another. A single dry month is not a drought. SPI-1 is noisy by construction and is most useful beside a longer window, where the disagreement between them is the finding."},{"id":"spi-12","name":"Standardised Precipitation Index (12-month)","definition":"A normalised measure of how unusual the last twelve months of precipitation are for a given place, expressed in standard deviations.","unit":"index","nativeGeography":["grid-cell","monitoring-station","climate-division"],"comparabilityNotes":"SPI measures meteorological drought only. It says nothing about whether reservoirs, rivers or aquifers are actually depleted, which can lag rainfall by months to years. It is one index among several that disagree by design and must not be combined with them into a single drought score. Consecutive months share eleven months of input, so a run of low values is one dry year being reported repeatedly, not a run of independent dry periods."}]}