Paul Dorian

Overview

Artificial intelligence (AI) is a collection of technologies that allow computers to perform tasks that typically require human intelligence, and it is increasingly impacting the world of weather forecasting. The European Center for Medium-Range Forecasting (ECMWF) has made strides with its Artificial Intelligence Forecasting System (AIFS) as it has recently become fully operational and is now run side-by-side with its traditional physics-based Integrated Forecasting System (IFS). According to the ECMWF, the AIFS has outperformed the physics-based model for many measures including, for example, tropical cyclone tracks. In addition to the ECMWF AIFS, there are at least four other known “A.I. trained” weather models including NOAA/Google GraphCast, Microsoft’s Aurora, NVIDIA’s FourCast, and Huawei’s Pangu-Weather.

A side-by-side comparison of the conventional European forecast model (left) and the Euro-AI version (right) of “Total Snowfall Amounts” in the continental US for the period ending Friday AM, April 4th. Maps courtesy ECMWF, Pivotal Weather

Discussion

The traditional approach to weather forecasting has been to make use of numerical weather prediction (NWP) which relies on current conditions, physics-based models, and the solving of complex equations on powerful supercomputers to output such parameters as temperature, pressure, winds, and precipitation at future times. Artificial intelligence (AI) models, particularly machine learning, are being increasingly used to improve weather forecasting by learning from large datasets of weather data to identify patterns and trends. AI models can process data faster and identify complex patterns, potentially leading to quicker and more accurate forecasts. The increasingly important role of AI in weather forecasting will be to complement and enhance traditional NWP models.

The European Center for Medium-Range Forecasting (ECMWF) has made its Artificial Intelligence Forecasting System (AIFS) the first such fully operational weather prediction model that uses machine learning and artificial intelligence. Making such a system operational means that it is openly available and has 24/7 support for the meteorological community. This AIFS can produce a wide range of output parameters including winds, temperatures, and details on precipitation types from snow to rain. The AIFS currently has a grid spacing of 28 km and, according to the ECMWF, it can outperform its physics-based counterpart by as much as 20% on certain measures.

A side-by-side comparison of the conventional European forecast model (left) and the Euro-AI version (right) of “500 millibar height anomalies” across the continental US for the validation time of 8AM, Sunday, March 30th. Maps courtesy ECMWF, Pivotal Weather

The AIFS uses the same initial atmospheric conditions for its forecasts as the IFS. These are based on the combination of a previous short-term forecast with around 60 million quality-controlled observations from satellites as well as many other streams, including from planes, boats, sea buoys and many other Earth-based measurement stations. Every six hours, these initial conditions feed into the AIFS. The machine learning model, trained on how the weather has evolved in the past, assesses how the initial conditions will influence the weather for the coming days. By contrast, the IFS uses physics-based capabilities to arrive at a forecast with a grid-spacing of 9 km over the globe, integrating the laws of physics in its computer code.

Meteorologists rarely put total confidence into a single model but instead, look at a suite of model forecasts. Most global models compute 10-day outlooks every six hours and one way forecasters gauge a model’s reliability is by checking its consistency from run to run. The Euro-AI model did well with this test for Hurricane Francine during September 2024 as it predicted landfall in southern Louisiana every run from 96-hours out until its landfall near Morgan City, LA. Maps courtesy foxweather.com

The first operational version is called the AIFS “Single”. It runs a single forecast at a time, known as a deterministic forecast. However, ECMWF is pushing this model to create a collection of 50 different forecasts with slight variations at any given time to provide the full range of possible scenarios. This is known as ensemble modelling, a technique developed and implemented by ECMWF more than thirty years ago. According to ECMWF, the launch of AIFS “Single” as an operational service is the first step in upgrading its artificial intelligence forecasting capabilities. The next step will be to make ensemble forecasts available by using artificial intelligence and extended range (seasonal) forecasts as well and we’ll continue to monitor the progress here at Arcfield Weather.

Meteorologist Paul Dorian
Arcfield
arcfieldweather.com

Follow us on Facebook, Twitter, YouTube

---

Discover more from Watts Up With That?

Subscribe to get the latest posts sent to your email.

via Watts Up With That?

https://ift.tt/ZpWl854

March 27, 2025 at 08:02AM

However one considers the supply of fresh potable water to properties in the UK, it is definitely an industrial process on a very large scale. The per capita average water consumption is approximately 140 litres per day and there are close to 70 million of us in the UK. Water companies have historically had tens of thousands of rain gauges around the country as hydrology is a primary interest. Being equipped to take meteorological readings makes water companies valuable to the Met Office as an information source and inevitably lots of weather stations developed around the existing systems’s reporting infrastructure – but you can take things a bit too far.

51.00889 -1.33531 Met Office CIMO Assessed Class 4 Installed as Rain Gauge 1/1/1892 Manual Temperature readings from 1/4/1979 Automatic readings from 7/10/2010.

When researching weather stations it is worth pointing out that the set up dates the Met Office quotes are from whenever the site first exists whatever its function may have been. Whilst many sites seem to have exceptionally long history they were nearly all originally just rain gauge sites that often long predated the Met Office. Knowing precipitation and water availability was actually more important than recording the temperature in days when survival was more than just a basic instinct. In the case of Otterbourne though installed originally in 1892, its temperature record only started in 1979 and was remarkably “patchy” at that. Here is just a tiny example from the archive notes of the hundreds that demonstrate how I define “patchy”.

“my Boss forgot to go!!!!” sums up two important issues. Firstly taking readings seemed to be somewhat low priority but secondly that some archivist felt it important to spend no doubt a long time transcribing issues like this into a digital archive. Quite astonishing on both counts but at least it created work to do. The very poor observations standard led to the automation of the site from 2010.

Addressing the site itself indicates all the problems of the numerous water (and also sewage) treatment plant sites that the Met Office operates. These are industrial plants and in no way represent the real world. The Met Office tape measure standard dictates this is a Class 4 site which, with a +/- 2 degree uncertainty added due to siting, renders the site unreliable for any historic temperature reconstruction. Clearly the screen is effectively in a walled garden of wholly unnatural construction surrounded by buildings, treatment tanks, roadways and virtually no natural vegetation other than probably the square metre the screen stands above. If this were fully certified and calibrated equipment but under private ownership any readings would be ruled out as unacceptable by the Met Office themselves. That it comes under their jurisdiction so they can authorise them is absurd.

That does not stop the Met Office from using its readings (of which only the last 14 years are frequent enough to have any semblance of meaning) to be used in compiling Location Specific Long Term Averages for inevitably non existent locations. With almost boring predictability the Met Office presents this list of the “undead”.

Martyr Worthy closed 2002, Southampton, Mayflower Park closed in 2000 and Solent closed in 2015. At least Middle Wallop (Class 4 Heliport and semi circular car parking Stevenson Screen island) and Boscombe Down (Class 5 aviation site with doubled up data) are still alive/open but one does wonder why anyone would refer to such distant sites.

Despite this poor siting in a completely unnatural location the Met Office are perfectly happy proclaiming the day’s “Extremes” from here. No doubt if someone covertly wafted a hot air blower over the screen on a hot day and produced a new national record it would duly be verified as acceptable. That is just plain wrong.

via Tallbloke’s Talkshop

https://ift.tt/oi4jwgY

March 27, 2025 at 06:36AM

By Paul Homewood

The OBR have now published their detailed fiscal outlook following Rachel from Accounts’ Emergency Budget Spring Statement.

Of interest to us is the table for Environmental Levies, AKA Renewable Subsidies:

https://obr.uk/download/economic-and-fiscal-outlook-march-2025/?tmstv=1743072930

As has been the case for the last few years, they don’t include subsidies embedded in Feed in Tariffs, which they used to count as Levies. So I have included them in the modified table below, along with the Climate Change Levy imposed on electricity and gas bills for businesses, which are separately shown in Table 3.9:

In total therefore, the various subsidies and levies will cost £17.1 billion this year, rising to £19.8 billion in 2029/30. This will be £5.0 billion more than last year.

Given that all of this, except for RHI, is loaded on to energy bills, it is not credible that bills will fall by £300 a year, as Labour promised.

These projected costs are all either already in existence or in the pipeline. They will go even higher if Miliband starts spraying more money around on his pet projects, such as carbon capture.

Neither do these costs include the massive extra costs of balancing and upgrading the grid required to  meet Labour’s Clean Power 2030 objective, as already highlighted by NESO.

One of the OBR’s tables is worth a closer look at:

What this is saying is that wholesale electricity prices will quickly drop back to below £80/MWh, after the current spike.

The strike price for new offshore wind projects agreed last year is around £84/MWh at 2025 prices, and will likely rise to over £90/MWh by they time they begin operating in four or five years time.

This demolishes claims that they are cheaper than gas generation.

via NOT A LOT OF PEOPLE KNOW THAT

https://ift.tt/eS2zBfc

March 27, 2025 at 06:31AM

As if we didn’t know…’A large increase in energy storage will be critical to ensuring a clean power system by 2030′ – ELN summary. Two problems there: it’s unaffordable and impractical. Using words like ‘must’ and ‘critical’ is just foot-stamping, leading nowhere. Where would all the storage come from in that timescale, or any timescale for that matter, and how would the vast cost be covered? Not going to happen, as storage with grid-scale capacity doesn’t exist anyway (batteries discharge only once before recharge!), which leaves the original intermittency problem out in the open with no viable solutions. Some more transmission lines will be strung up over the countryside on massive pylons, and some local battery ‘farms’ will appear, but that’s about it. Gas will have to fill most of the gaps, as now.
– – –
The UK must significantly expand energy storage to achieve its clean power system target by 2030, as a tenth of wind-generated electricity is currently wasted, says Energy Live News.

A report by Drax and Imperial College London highlights that last year, wind power became the UK’s largest electricity source (31%), surpassing fossil fuels for the first time in 140 years.

However, 8.3 TWh of wind energy—enough to power two million homes—was lost due to grid congestion, costing consumers nearly £400 million.

Investing in long-duration energy storage (LDES) and battery energy storage systems (BESS), alongside grid improvements, could store surplus wind power and release it when needed, reducing reliance on gas during low-wind periods.

Without urgent investment and policy support, the country risks wasting substantial clean energy.
. . .
”The rapid and continued growth of wind power gives us grounds for optimism on the journey towards clean energy, but that growth presents its own very real and different challenges” says Dr Iain Staffell of Imperial College London, the lead author of the report.

“Last year we saw a 17% reduction in emissions compared to 2023 but with more wind comes more intermittency. This was evident throughout the winter of 2024 and into 2025, with three separate periods of cold, calm weather – known as ‘dunkelflaute’ – exposing Britain’s reliance on costly imported energy and drawing down the nation’s gas storage to ‘concerningly low’ levels.
. . .
Wasted wind

The curtailment of wind energy – where turbines are shut down when there is no capacity to transport their power [Talkshop comment – or supply exceeds demand at any time] – is rising at an unsustainable rate.

In just one year, curtailment doubled from 5.5% to over 10%, largely due to Scotland’s wind farms producing more energy than the grid can carry south to demand centres in England.

Full article here.
– – –
Image credit: impactlab.net

via Tallbloke’s Talkshop

https://ift.tt/HUbu1fa

March 27, 2025 at 04:46AM