A rainfall guarantee and a statistical estimate need a common definition before their numbers can be compared. The useful questions are what rainfall quantity is promised, how its untreated baseline is constructed, where it is measured and over which period.
Two statements to read together
In its 27 April 2026 announcement, Rain Enhancement Technologies said it would guarantee at least 10% precipitation enhancement for WETA clients, conditional on acceptable historical data and future measurements. The announcement is a company statement; it does not supply a complete contract or independent evaluation of every deployment.
The July 2026 preprint by Tho, Chambers and Welsh reanalyzes historical Oman ionization-trial data. Its revised estimator gives 6.68%, with a bootstrap 95% interval of 2.21–12.88%. Within that analysis, the earlier estimator gives 12.52%. These are estimates under specified statistical definitions, not observed percentage gains promised for every site.
Keep the data and method attached
The preprint revisits the 2013–2018 data; it is not a new independent field trial. Its aggregate uses positive-rainfall downwind gauge-day observations and a modeled natural-rainfall baseline. A seasonal catchment total or a contractual performance metric may be a different quantity. The paper is labeled as a versioned preprint in our library record.
Questions a useful evaluation should settle
- Endpoint: a gauge observation, eligible-event total, seasonal precipitation or usable water delivered?
- Baseline: what would have happened without the intervention, and how is its uncertainty estimated?
- Eligibility: which weather, dates, sites and instrument records enter the comparison?
- Allocation and dependence: how are treatment decisions, neighboring gauges and correlated weather handled?
- Commitment: which conditions, evaluation period and remedy appear in the actual agreement?
These questions protect both the buyer and the science. A business commitment can be assessed against its terms; a scientific claim must be assessed against the design and evidence that support it.
Why this does not decide FogLab
FogLab asks whether powering a collector changes optical attenuation beyond the identical passive collector. Its measurement, mechanism and experimental unit are different from a rainfall-attribution study. The useful connection is methodological: define the outcome and comparison before interpreting a favorable number.