After almost 18 years of research, campaigning and negotiation, statutory protection for the most vuulnerable reefs in Lyme Bay became a reality in 2008. This was deemed necessary as, despite voluntary agreements, it was apparent the damage to the reefs was still occurring. The questions that needed to be addressed then became:
- how will the species on the reefs respond?
- How long will full recovery take?
- Will the species that appear to have been hardest hit, the sponges, the seafans, the soft corals, the larger sea squirts, re-establish themselves in a few years or will other species colonise the reefs first?
Between summer 2008 and summer 2010 we undertook a monitoring programmme looking at the species present and how there numbers were changing. In that time we have just started to get the first tantalising glimpse of the changes occurring.
We have now been able to publish our report on the monitoring work we undertook looking at the changes that occurred on boulder reef communities. The full report (as a PDF) can be downloaded Here: Lyme_Bay_Closed_Area_Monitoring_2008-2010_MBI.
However, as this is a fairly lengthy document it seemed a good idea to summarise what we’ve found so far in a fairly non-technical way, so here goes.
Before describing our findings it’s worth going over a description of Lyme Bay and a bit of background regarding the concerns about bottom fishing towed gear and earlier attempts at achieving a degree of protection. This helps explain some of our reasoning in survey design and interpretation of the data. (If you only want to read the outcomes then skip down to What we have found so far? near the bottom)
Lyme Bay is a large, open, south-facing bay in Southwest England, opening into the English Channel (see study layout map, below). The West of the Bay is predominantly fine sediment seabeds, fine sands or mud. The waters in the west also tend to be more turbid. This is in part due to the tides are being weaker in the western part of the Bay (allowing fine sediments to settle out of the water column) and also because much of the rock here is soft, rapidly eroding sandstone. But mostly it is because of the two major rivers, the Teign and the Exe, which flow in to the western side of Lyme Bay carrying large sediment loads which are then deposited in the Bay.
Tidal streams are markedly stronger and no rivers flow into the eastern part of Lyme Bay. The seabed here is much rockier and sediment tends to be much coarser. The greater amounts of exposed bedrock, in particular the high rocky ledges, and stronger currents, generally results in richer assemblages of filter feeding animals such as larger erect sponges, gorgonians, soft corals.
The reefs of greatest concern in relation to damage from scallop dredges and trawlers were located a little to the east of the centre of the Lyme Bay. They lie in a band between (roughly) 20 and 24 metres below chart datum (approximately 22 – 29 metres actual depth depending on the state of the tide). Back in 1992 the Devon Wildlife Trust began getting reports of damage to reefs caused by rockhopper trawls and in particular scallop dredges. We then conducted a series of diving surveys, documenting the damage occurring here for the first time. After many years of negotiation, Devon Wildlife Trust and local fishermen reached an agreement whereby bottom towed fishing gear would not operate within two vulnerable reef areas, known as Lane’s Ground and Saw-tooth Ledges. This agreement came in to effect in 2001. Two other reef areas, known as Beer Home Ground and the East Tennants Reef, were subsequently added in 2006. This was a considerable achievement by both Devon Wildlife Trust and local fishermen, and the agreement was largely adhered to. The problem was it wasn’t adhered to by everyone, and one or two scallop dredgers passing through such an area will cause damage that will linger for years. So, in July 2008 a larger area of 60 square miles within Lyme Bay was closed to all towed bottom gear fishing by Statutory Instrument. This area enclosed the four existing voluntary areas.
At this point the need to begin monitoring of how the newly protected area responded to this protection was recognised. The question was, how to design the monitoring? Ideally monitoring would have commenced several years before the statutory protection came in to place, allowing a time series of before and after comparisons . However that had not happened and nothing could be done about that now. Comparison of change in reef habitats inside and outside of the protected area is another approach, but the problem here is that (as described above) conditions and habitat vary markedly across the bay to the east and west and it was very unlikely that similar habitats existed to the east or west of the closed area box, or that species would reproduce, settle, grow and interact in similar ways east or west of the Closed Area box as they would inside the box. South (offshore) of the box conditions were very different; the water was deeper and the seabed mostly sedimentary, thus it could not be used for comparison either. There are also the four existing voluntary protected areas to consider. One possibility is to compare existing voluntary protected areas with areas in the new Closed Area. If this approach is taken two important factors must be taken in to account. The first consideration is that each of the voluntary protected areas was very different to the others, in terms of habitat and the species assemblage they support. By way of example, if one considers Lane’s Ground Reef and East Tennant’s Reef (See map above).
East Tennants Reef is slightly deeper, exposed to slightly stronger currents and is composed of large slabs of limestone, whereas Lane’s Ground is composed of small boulders on patches of sand, gravel and stones. Lane’s Ground supports relatively few seafans, somewhere in the order of 1-10 per 100mm2. No large seafans are found on Lane’s Ground Reef, presumably due to a combination of greater exposure to wave action (as shallower), reduced feeding currents and reduced stability of the smaller boulders.
East Tennant’s Reef, in contrast, supports very high densities of large seafans, averaging several hundred per 100mm2, mature colonies being 4-5 times the size of those on Lane’s Ground. Thus the biomass and reproductive capacity of seafans on East Tennant’s Reef will be many times greater than that on Lane’s Ground and, as conditions are clearly much more favourable to seafans here, it would be unsurprising if settlement, early survivorship and growth rates were higher here also. So we can see that it is not possible to treat all pre-existing voluntary areas as one condition or ‘treatment’ to compare with newly protected areas outside. The second factor to consider is that the differences between the existing voluntary protected areas and the newly protected Closed Area beyond their borders were subtle. Remember the statutory protection was established because the voluntary protected areas were not being fully complied with. Thus they will not be uniformly ‘better’ than the areas outside; rather they are a patchwork of relatively pristine and recovering areas (the degree of recovery depending on how long ago they were established, 2002 or 2006) and of damaged areas due to recent incursions.
The hypothesis we were required to test under the funding from Natural England was that:
Over time, species assemblages within sites in the new statutory closure but outside the pre-existing voluntary closures would change to more closely resemble those in the pre-existing voluntary closures and become less similar to sites where fishing by towed bottom gear was still permitted.
From what we already know about the Bay and about the voluntary closures we can Immediately see problems here: both in comparing what’s happening inside the Statutory Closure to what’s happening outside and in comparing the voluntary closures to the areas of the new statutory closure outside the voluntary closures.
This is the approach we took.
- We elected to work with one Voluntary Closed area only, Lane’s ground reef, as (as described above) the differences between the different voluntary closed areas was far greater than any likely change in species’ abundances due to cessation of fishing in the three years of the study. To mix habitats through treating multiple voluntary closures as single treatments would simply create vast amounts of ‘noise’ and introduce many other factors than may be responsible for differences in response, other than cessation of fishing, and so making it impossible to interpret the data. This also meant we could very tightly define the habitat we were studying, in terms of seabed composition, relief, depth, tidal streams and wave exposure, all factors we were aware would markedly modify the species assemblage and so potentially compromise our interpretation. There were a couple of other factors in our selection process. Being relatively level with only small boulders present, Lane’s Ground Reef was relatively easy for trawlers and scallop dredgers to work; there were no ledges or large rocky outcrops on which to come fast or damage gear. The fact that it was level also made it much easier to extract reliable data; it is hard to get good quatitative data from reefs with ledges, overhangs, steep slopes etc simply because of the logistics of laying and counting withing transects and quadrats. We also had very good historical data on Lane’s Ground Reef. Numerous surveys (many by ourselves) had been conducted there over the previous 18 years, thus we had a very good handle on the habitat, the species we were likely to find and their distribution. This hugely aided our survey design. A final, but very important consideration was that Lane’s Ground Reef had been identified previouusly as especially important for its sponge assemblages. Now we knew that sponges amongst the most vulnerable to mobile fishing gear and there was a lot of anecdotal evidence and comparative video suggesting that sponges had declined markedly on Lane’s Grouund over the past decade. But the boulder reef habitat was still largely intact, so this seemed an ideal testing ground to find out whether they would recover once physical disturbance cesed.
- The aim, within the hypothesis to be tested, was to compare relatively pristine sites (the voluntary closures) with the newly protected sites within the Closed Area and also with the unprotected areas outside the Closed Area. But we knew the voluntary closures were not pristine and contained many areas similarly damaged to those within the new closure. Clearly, simply randomly assigning areas within the voluntary closure would not achieve this, so instead we elected to conduct a pre-survey of Lane’s Ground voluntary closure to identify relatively pristine areas, from within in which our ‘pristine’ comparisons would be haphazardly located (i.e. a random stratified approach). This proved to be more important than we had expected; significant tracts of Lane’s Ground voluntary closure appeared markedly degraded (i.e. showing clear signs of physical damage or markedly reduced numbers of epifaunal species compared to historical surveys of the reef), with relatively pristine areas appearing no more than small ‘islands’ dotted across the reef.
- Comparing with areas outside the statutory Closed Area, as the contract required, proved most problematic. In the end we elected to compare with areas running along the same depth contour as Lane’s Ground Reef, where similar boulder reef habitat occurred immediately outside the Closed Area. This was a difficult choice. Selecting areas immediately outside left our study open to criticisms of ‘edge effects’ such as ‘fishing the line’ (where fishermen tend to work along the edges of a protected area more so than further afield). Yet as we knew, conditions varied markedly as one travelled east, west or south of the Closed Area, and that similar habitat was very difficult to find outside the Closed Area, thus rendering data from further afield unsuitable for comparison.
The layout of the study is shown here. Four monitoring stations were located within Lane’s Ground Reef (what had been the old voluntary closure and was now referred to, conforming with the terminology defined by Plymouth University, as the Closed Control), three within the new Closed Area, but outside of the voluntary closure (termed the New Closure) and three just outside of the new Closed Area, termed the Open Controls. So, testing the hypothesis
What have we found so far?
So, what we found was the hypothesis (comparing station outside the Closed Area to those within) didn’t work – essentially because the underlying assumptions were incorrect, namely that the conditions outside of the Closed Area were similar to those inside. Even when carefully selecting habitat type the results demonstrate that environmental conditions outside of the Closed Area are too dissimilar for meaningful comparisons (in terms of change likely to be due to bottom fishing effects) to be made. Given what we already knew about the Bay this was no great surprise. The data also suggested that the New Closure and the Closed Controls were also different when one looked at all the species studied, but with a fair degree of overlap when individual species were studied. Again this was not terribly surprising. What we, however, did see was a certain amount of change in both New Closed and Closed Controls. Now three years (essentially three data points on our time series) is a very short timescale for the species we are looking at, but what it does suggest is that both new closures and existing voluntary closed areas may be responding to the cessation of trawling and dredging. Again this is not hugely surprising. We know that the existing voluntary closure had suffered from incursions and so was not pristine at the start of the study; areas not impacted by direct disturbance (i.e. trawls or dredges passing directly over) are likely to have suffered indirectly from the effects of increased sediment plumes as gear passed nearby and mobilised seabed sediments. As most of the species of concern (sponges, hydroids, soft corals, gorgonians, sea squirts) are filter feeding organisms it is quite likely this had a deleterious effect on them. Thus even areas that we had pre-selected as relatively pristine may have deteriorated due to the proximity of mobile fishing gear. This may seem like a bit of a failure, but it is useful as there is often considerable pressure to design studies that are statistically elegant but do not take in to account the complexities and variability of the real environment.
Signs of recovery?
Perhaps the most exciting of all is the possible early signs of sponge assemblage recovery.
Sponges, in particular erect branching sponges, are possibly the most vulnerable of the prominent species found in Lyme Bay. They are soft bodied and easily destroyed by physical contact. They are also filter feeders and so likely to suffer from significant increases in sedimentation. Many are believed to be very slow growing, studies at Skomer and Lundy Island Marine Nature reserves indicate that axinellid sponges (a significant group of erect branching sponges) suggest they are very long lived. Sponge assemblages have also previously been identified at one of the most notable features of the reefs in Lyme Bay, with Lane’s Ground Reef highlighted as previously supporting particularly rich sponges assemblages and that these rich sponge assemblages were, probably more than any other feature, what made the reefs of such high conservation importance, with many unusual or rare species and others not yet fully identified. Thus determining whether sponge assemblages recover, and over what timescale, is fundamental to identifying whether the protection afforded to Lyme Bay is a success.
We believe this work is extremely important. The opportunity to conduct such a study as Lyme Bay affords us comes only very rarely. Lyme Bay Closed Area is the first such closed area established for conservation purposes in England’s waters. Furthermore, there are few areas of coastal seabed as well studied as Lyme Bay, thus although we don’t have the ideal pre and post closure monitoring we do possess a wealth of data on what these reefs used to be like almost 20 years ago. Given the uniqueness of this opportunity and the very encouraging signs in the data from the first three years it would seem essential that the monitoring is continued. Currently, although the signs are both encouraging and more or less exactly what we would expect given what we already know about Lyme Bay they are, with only three years data, simply an indication of where change might be heading and no more. Consequently we are now actively seeking funding to restart monitoring in 2013.
[Rockhopper trawls: bottom trawls fitted with extra large rubber discs on the footrope, allowing them to bounce or roll over boulders and small rock outcrops and so work rocky seabeds that other trawls could not]