We conduct extensive verification of our simulation models, comparing them to actual observation data. Thereby, we ensure that our services are delivering top-quality (and continuously improving) forecasts. We are the first commercial weather service that regularly publishes verification data on their website since 2010.
Why do we publish our validations?
- We are transparent: weather is no "chaos" and our customers should know what they receive.
- We deliver quality: our accuracy is so high that it is worth showing.
- We are realistic: you should know what to expect from a forecast - and what not to expect.
- We are competitive: if someone believes we are not good enough - show us how to do better.
What does meteoblue quality mean? Here are some examples.
meteoblue predicts 80% of all temperatures with less than 2°C difference from measured temperature for the next 12 hours; 3 days in advance it is more than 70% of all temperatures. In all meteoblue forecasts, temperature simulations are performed with MOS (model output statistics).
An agreement of 70-80% agreement for wind speed is reached with the meteoblue wind forecast for the period of 12-48 hours ahead. In most of the meteoblue forecasts, wind simulations are performed with MOS, whereby the mean absolute error is less than 2m/s for 96% of the stations.
meteoblue calculates radiation for the land and sea surface and for atmospheric layers, both as incoming direct and indirect sunlight, as well as reflected radiation from clouds or surface. meteoblue radiation simulation is consistent over continents and reaches a monthly mean absolute error of 1-15% in 95% of all places.
meteoblue can forecast more than 75% or 85% of all precipitation events with more than 2 or 5 millimetres (mm) correctly 3 days in advance. This includes forecasting the days on which such events will not occur. Some precipitation variables, such as total precipitation quantity in tropical climates, snowfall and depth, thunderstorm rainfall distribution and hail, can only be simulated with (very) limited accuracy. Contact us for further references.
meteoblue forecasts are high-quality approaches to the reality. Are weather stations still needed?
Absolutely yes, if…
… simulations (historic or forecast) must be validated: only reliable measurements ensure that simulation accuracy is known;
… specific micro-climate situations must be as accurately described (e.g. for wind or solar power generation, fruit production, etc.).
… the reality is characterised by small-scale differences (e.g. valleys, mountain summits);
… short term changes within a range of minutes are important (e.g. airports);
… long-term developments have to be observed (e.g. disease development in a vineyard);
… small differences have a great economic importance (e.g. frost on the road, irrigation management, etc).
Then, weather stations are very important for decision making on weather-dependent measures.
meteoblue simulations can complement such measurements, where decisions have to be taken, either as forecast 6 to 144 hours before the arrival of the event, or as historic dataset for assessment of risks and probabilities.
Model validation: resolution, nowcasting and MOS
A comprehensive verification study of temperature and wind simulations conducted over more than 1000 weather stations and one entire year shows the improvements obtained by model resolution, nowcasting and MOS (Model-Output-Statistics) modelling methods developed by meteoblue. You can download the document below:
For specific variables and uses, meteoblue offers specific validation tools and reports.