Britain’s offshore wind resource (1952-1982)

Today, I discovered the website of Helen Rozwadowski (@oceanhistories), it is called fathomingtheocean.com/, it deals with Environmental History and the connected History of humans and the ocean. In that context, I realised that I should have enough material about the offshore wind resource to write a short post. Therefore, I quickly scanned a few articles and gathered a few examples, sparse and non-exhaustive, of historical attempts to assess the wind resource over the North Sea; as an example.

For this post, I have used five sources: a 1952 report jointly edited (among others) by Günter Dietrich from the German Hydrological Institute and A.L Lawford from the Admiralty Hydrographic Department, as well as four conference/symposium proceedings between 1979 and 1982; see below.

• (Dietrich, 1952) “Wind conditions over the Seas around Britain during the Period 1900-1949” by the G. Dietrich and K. Wyrkti (German Hydrographic Institute) and J.N Carruthers, A.L. Lawford and H.C. Parmenter (Hydrographic Department, Admiralty)|Link|
• (Moore, 1979) “Offshore Wind Data” by D.J. Moore (CERL), in “Proceedings of the First BWEA Wind Energy Workshop” |Link|.
• (Lindley, 1980) “Assessment of the Offshore Siting of Wind Turbine Generators” by D. Lindley and P.B. Simpson (Taywood Engineering Ltd), U. Hassan (ERA Technology Ltd) and D. Milborrow (CERL), in “Third Internationl Symposium on Wind Energy Systems”|Link|.
• (Rockingham, 1981) “Offshore Wind and Wave Power – a Preliminary Estimate of the Resource” by A.P Rockingham, R.H. Taylor, J.F. Walker (CEGB), in “Proceedings of the Third BWEA Wind Energy Workshop” |Link|.
• (Milborrow, 1982) “The UK Offshore Windpower Resource” by D.J Milborrow, D.J. Moore, N.B.K Richardson and S.C. Roberts (CEGB), in “Fourth International Symposium on Wind Energy Systems” |Link|

The first North Sea wind hindcast dataset ?

The 1952 report is relatively short (38 pages), with only few pages of text and a lot of tables and maps.

It intends to provide, for five areas around Great Britain (four in the North Sea, one in the Irish Sea and one in the Channel), monthly statistics of mean wind speed and direction between 1900 and 1949. Therefore, it is some sort of “hindcast” of the wind conditions, its primary purpose was to support studies in Fishery Hydrography, see the quote below. How ? I don’t know. The question of the surface currents (related to the surface wind) is mentioned in the introduction but no example is provided.

To those whose interest lies in the study of Fishery Hydrography, therefore, it is hoped that the tables and charts presented herein will prove a useful tool.

There is one part of the report which I like in particular, these are the graphical representations of the data in the form of time series of wind vector anomalies, see an example below. The mean wind vector for the entire period is provided, for every month (rows), in the first column; then for each year (columns) only the anomaly vector is plotted.

I have made a single check using the NOAA-CIRES 20th Century Reanalysis version 2, but that one worked: see how the U component of the wind was indeed negative in February 1929.

I was very suprised to find out that hindcast datasets had been computed prior to the 1970’s. I haven’t really investigated if there are other studies like this one, but here is one more thing to do in 2019 ^^.

The UK offshore wind resource

Now, on to wind energy stuff. You may have seen it, the Crown Estate will soon unveil new offshore wind lease areas. The first map of offshore wind lease areas I know of, is from (Lindley, 1980), see below. The selection criteria were the classical combo {water depth < 20 m, soil, distance to coast > 5 km, shipping, etc} (see details in the article).

Some of the names of the clusters (Shell Flats, Docking Shoals) will sound familiar to the offshore wind practitioners. The report mentions three 600MW clusters, with each 196 3MW turbines (80m rotor diameter). A classical parametric study of the LCoE is presented. Note that:

• wake effects have been calculated using the wind tunnel experiments from Mr. Milborrow,
• only extreme wave conditions have been considered (not mention of fatigue loads),
• mean wind speed estimates are provided (see below).

The mean wind speeds estimated at 70 mMSL at Shell Flats, Camarthen Bay and Docking shoals are respectively 8.9, 9.2 and 8.6 m/s. The value for Camarthen Bay seems wrong (too large), but the others are not far off.

Prior to this article, there a very interesting paper dealing with some general aspects of the offshore wind resource, in the first BWEA workshop proceedings, see (Moore, 1979) where the author is not so optimistic about the accuracy of production estimates..

Follows (Rockingham, 1981) where the plans are more ambitious: 165 GW of installed capacity ! Areas with water depth smaller than 30 m, and further than 5 km from the coast are targeted. Shipping has been considered too, and 2.5 MW 80 m diameter rotors have been used (with an array efficiency of 60% only).

See the conclusion of the paper below: with all constraints accounted for, the target installed capacity shrinks to 70 GW (the current target is 30 GW in 2030).

The last report I will mention is (Milborrow, 1982), which refines the methods and findings from the two previous planning reports. There are three main interesting aspects in this document:

To start with, it contains refined turbine performance estimates based on the experience gathered here and there from prototpes onshore, see the table below.

Second, it includes a more detailed description of the spatial planning constraints, see here an example with the Wash area:

Finally, it includes a more refined wind resource assessment, combined with the spatial planning exercise and site specific wake calculations, see below. This is very close to what is being done today for early phase projects in prospection phase.

There is even a wind speed map (the first of its kind, as I know). Apologies for the bad quality of the scan, I will repost a better image later. the map shows values of the first significant decimal digit of the mean wind speed for 10 sqkm cells, and one needs to use the values provided at the margins (mean wind speeds of the cells right outside the map) to derive the ones inside. Not easy, but feasible I guess. This is where the values in the above table come from. I did not understand how they were computed.

Summary

We saw that long before the need for knowing accurately and precisely the value of mean wind speeds over the North Sea for estimating the production of offshore wind farms, a 50-years long hindcast dataset had already been computed by a joint project between Germany and the UK in 1952. The documents from the early 1980’s, in the absence of durable large turbine prototypes, provide examples of state-of-the-art wind resource and production estimates, and demonstrate how offshore wind energy spatial planning became formalised at a high-level.

And, wind speeds estimates were not that off !

Comments, suggestions are welcome.

Rémi