Weather forecast in offshore operations, how accurate is it?
There's no doubt about it that weather forecast is a critical element in all offshore operations. It always has been and and always will be.
But how can you make the most of your weather forecast? How can you maximize it?
After measuring the ocean surface for almost 40 years we Miros - Real-time Ocean Insights first handedly know that the actual measured sea state at the exact location of your vessel from time to time could be quite different than what the weather forecast states. In all fairness, this discrepancy isn't entirely unexpected. A forecast remains just that – a forecast, not an exact science. We are all probably familiar with how wrong it could be in our daily life onshore from time to time. If that's difficult to forecast, it becomes obvious that adding waves & current will complicate the hell out of it.
In the realm of marine offshore operations, wouldn't it be both fascinating and advantageous to receive an indication of the accuracy of the weather forecast you rely on?
We certainly think so. Consequently, we made the decision to visually represent the disparity and accuracy of the weather forecast directly in our fully digitalized Miros App, eliminating the need for our clients to manually compare them.
And this is how it looks
Check out the new "Forecast" tab located in the upper right-hand corner. The screenshot is captured from a vessel in the North Sea this morning, and we immediately observe a few intriguing details.
First and foremost, there's an unmistakable difference! The forecast indicates an Hs of 5.6m, whereas the vessel is actually experiencing an Hs of 6.9m. That's a substantial variance.
Additionally, it's evident that the forecast has consistently underestimated Hs over the past few hours. It will be interesting to monitor whether the actual measurements align with the forecast later today.
领英推荐
Let's rewind one day and see if there's any interesting observations there as well.
Even more intriguing! Here, we observe the opposite scenario. Yesterday morning, between 10:00 am and 12:00 hrs, the forecast overestimated Hs. It's not entirely inaccurate; it's just that, in reality, the Hs doesn't abruptly spike by 2m in a matter of minutes. As clearly seen, it happens gradually, taking about 2 hours in this instance.
After that the forecast is back underestimating again, but as one can see that the forecast was on its money at 16:00hrs when there was a small drop in weather. Still underestimating, but at least will a lesser margin than prior to 16:00hrs.
While examining the past and present situations is insightful, what about looking into the future? At present, we have the opportunity to look a few days ahead.
The screenshot below pertains to the next 24 hours.
Weather apparently due to come down quite a lot the next 24hrs and it will be interesting to see which line the it will take.
Get in touch if you want real-time ocean surface insights in order to enhance your offshore operations to the next level.
Head of Metocean, UK. DHI
1 年Chris, interesting article, but I had a few questions and comments. firstly, how are you measuring the wave height from the vessel? is derived from the vessel motion? or do you have a radar on the vessel looking outwards? Also, is the vessel stationary or in transit? I also don't think that you comparing 'apples with apples' here. wave model forecasts are inherently smoother than observations as they represent a time period (average say over an hour) and also a spatial area, determined by the model resolution. depending on where you source you forecast from, this could be anything typically from 12km down to 0.5km. your observations are spot location data and at a much higher temporal resolution and so will absolutely show some differences. your model data looks like it has a relatively coarse time step (3hrs?), and higher resolutions should be available. To get a true comparison the observations should be smoothed so that they are representative of the model time averaging. All that said, site observations are absolutely critical in appraising metocean forecast performance and this intelligence should be fed back to the forecaster to help them improve model performance and/or use live data to adjust forecasts operationally.