First
Excess water
Radar sees water even through clouds. The detector got 74 of the 79 cases in the flood golden set right.
Parametric insurance · Res. SSN 315/2026
The rule is here.What's missing is an index you can defend.
Since July 23, 2026, Argentina has rules for parametric insurance: historical index frequency, an alternative source, an independent actuary and payment within 15 days of the event. Ucotron builds the index file and calculates each event with frozen data.
Index · hours with THI ≥ 79
Simulation
0h
Measuring the summer
Cumulative payout
0 %
of the sum insured
Payout clock
No event
From the resolution to the payout
There's a rule and no product.Not yet.
According to the SSN's agro survey, no insurer sold a parametric product in 2024 or 2025. The resolution laid out the path: this is what each step requires and who provides it.
-
21/7/2026
The resolution is signed
Res. SSN 315/2026 adds item 23.8 to the insurance industry regulation.
The rule
-
23/7/2026
Published and in force
It appears in the Official Gazette and takes effect that day. Parametric moves to case-by-case authorization.
The rule
-
File
Feasibility study
Index, thresholds, historical frequency across several baselines, basis risk and alternative source.
Ucotron
-
Validation
Independent actuary
Registered with the SSN, validates the study and the index. The signature is theirs.
The actuary
-
Authorization
The SSN approves the product
With the State as policyholder or over 1,000,000 UVA, it goes out without prior authorization and is reported within 30 days.
The insurer
-
Event
The index crosses its threshold
Ucotron calculates the index, freezes the data version and publishes it.
Ucotron
-
≤ 15 days
Payout
Fifteen calendar days from the event, not from when the data is published. A person decides.
The insurer
0
insurers sold parametric products in 2024 and 2025, per the SSN agro survey
15
calendar days to pay, counted from the event
≈ USD 1.42 M
1,000,000 UVA: from there, or with the State as policyholder, the product goes out without prior authorization
Res. SSN 315/2026 (RESOL-2026-315-APN-SSN#MEC), Official Gazette of July 23, 2026.
What we measured · heat on Rafaela dairy farms
Same index, same threshold.Frequency ranges from 12% to 46%.
The rule asks for historical frequency “over a statistically representative period” and doesn't say which. With 41 summers of ERA5 in the Rafaela cell, the choice of series changes the trigger frequency and nearly triples the premium. Choosing the baseline and justifying it is part of the file.
Hours with THI ≥ 79 per summerIllustrative shape
Summer that would have paidSummer with no payoutOutside the series
What we measured: hours with THI ≥ 79 between 12/1 and 2/28, hourly ERA5 via Open-Meteo, 41 summers (1985/86 to 2025/26), own calculation. The bars show the shape of the series; the frequencies and premiums are the calculated ones. Linear payout between 330 h and 480 h on a hypothetical sum of USD 10,000 per dairy farm, with hypothetical loadings.
When the event happens
The index crosses its trigger.The 15-day clock starts.
The deadline runs from the event, and public data doesn't always arrive on time. Every step is recorded, with the data version that was used.
Day
0
of 15
- Day 0 The window closes or the event happensThe 15-calendar-day clock starts here, not when the data is published.
- Day 5 The data arrives and the index is calculatedUsing the available version of the primary source, at the time and place the policy sets.
- Day 5 The data version is frozenPublication date, version and calculation are sealed. If the data changes later, a revision is issued with its reason.
- Day 6 It's cross-checked against the alternative sourceIf the two sources disagree, the case goes to a person.
- Day 7 The signed report is publishedThe insured sees the calculation. The report is verified with SHA-256, offline.
- Day 8 A person decides the payoutUcotron neither pays nor rejects: the insurer decides with the calculation in view.
- Day 9 PayoutSix days before the deadline.
Illustrative timing: each source has its own delay, and the policy sets which one is used and when.
The right instrument
We start with what we measure well.And we say what we don't.
1The field: open water in two or more passes
2Event water, separated from the usual
3The same place, in the basin
4New water by radar, even through clouds
1Zones by class: open water and likely waterlogged
2Days under open water, pixel by pixel
3No open water in the whole period
- 1The field: open water in two or more passes
- 2Event water, separated from the usual
- 3The same place, in the basin
- 4New water by radar, even through clouds
- 1Zones by class: open water and likely waterlogged
- 2Days under open water, pixel by pixel
- 3No open water in the whole period
74of 79
cases in the flood golden set correctly identified by the excess-water detector.
Each Sentinel-1 radar and Sentinel-2 optical pass is compared against a dry baseline of the same place. That separates event water from the usual lagoon and counts the days under water.
In this field: 66.8 ha, 68%, under open water for at least 37 days.
Second
Dairy heat stress
Hourly ERA5 since 1940, cell by cell. Final data arrives two to three months later: payouts use the frozen version.
For verification only
Hail
The lightning record starts in 2018. It's enough to confirm an event, not to price an index.
Not yet
Drought
Our instrument is still weak for this. We don't offer it as an index.
What we provide and what we don't
The file is ours.The signature and the payout aren't.
Ucotron provides
- The index, thresholds and historical frequency on several baselines, with the rationale for each.
- Basis risk measured against real losses, not against the index itself.
- The primary source and the fallback, with the rule for switching from one to the other.
- Each event's calculation using frozen data, and a signed report the policyholder can see.
What we don't provide
- The actuarial signature: it belongs to the independent actuary.
- The decision to pay: a person at the insurer makes it.
- The rate: with our study, the company sets it.
- A hail index: today it works for verifying, not for pricing.
Once per product
The file
Index, thresholds, frequency on several baselines, basis risk and fallback source, ready for the actuary.
Per season
The monitored index
Place by place across the whole policy term, with its data version.
Per event
The calculation and the report
The index value, frozen and signed, in time to pay within 15 days.
For insurers launching a parametric product, for the State as policyholder, and for the reinsurance behind it.
Parametric recipes
The same thing, switched on as a recipe.Via API or with the signed report.
Excess water
Parametric trigger
When the index's measurement window closes.
Computes the index, freezes the data version and publishes it so a person can decide the payout.
APISigned PDF
Excess water
Excess water by department
When the rainfall index exceeds its threshold or a river port goes on alert.
Alerts with the index value, its threshold and the alternate source.
EmailAPI
Heat
Heat index for dairies
When the temperature-humidity index stays above its threshold for several hours in a row.
Counts the hours above the threshold and leaves the calculation ready to audit.
APISigned PDF