The second meeting of 2016 took the WSG to Ampfield Wood in Hampshire. Here the group were given the opportunity to revisit the site of an experiment instigated in 1954, entitled the ‘Rehabilitation of Derelict Woodland’. The aim of the experiment was to understand how the implementation of a variety of silvicultural interventions might promote the development of silviculturally sound stands from once neglected woodland. The meeting offered WSG members the opportunity to develop this year’s theme of ‘The Silviculture of Mixed Stands’ in a unique setting.
Lead by Prof. Julian Evans, and supported by the Forestry Commission Solent beat forester Ben Phelan, our morning session began with an introduction to the site and an account of the rationale behind the experiments. Following the Second World War, a survey carried out by the Forestry Commission concluded that a significant proportion of woodland in the UK was derelict. The Ampfield site was one in a suite of 10 nationwide experimental sites, created in the 1950s, with the aim of determining the most efficient way of turning derelict woodland into productive woodland (with an emphasis on conversion to coniferous crop). Five different treatments were applied to woodland plots (below), each treatment being replicated three times:
The experiments were subject to assessment until the late 1960s, and then received very little interest during the 1970s. During the 1980s Julian reviewed the experiments, subsequently reporting the merits of enrichment planting over clearing and re-planting (Evans, 1984). Of particular note was the financial differential between such treatments, with enrichment planting requiring 60-70 man-hours per hectare and clearing and replanting requiring closer to 600 man-hours per hectare at establishment. The cost of maintenance operations was also considerably less where treatments focussed on enrichment planting rather than clearing and replanting.
Once part of the initial experimental matrix, following poor performance of Douglas fir, the site was converted to a trial of Populus trichocarpa in 1964. The once strong market for hardwood pulp now having dissolved, poplar can be a challenge to market. The high moisture content of the timber (and propensity to ‘re-wet’) makes poplar a poor firewood species. There does, however, appear to be some potential for export. The question was raised as to whether poplar species should feature in current planning trials looking at future species choice. Although there appears to be limited market potential at present, perhaps it is sensible not to exclude any particular species from such trials? Trees at this site have shown a poor rate of growth, potentially as a result of their narrow spacing (poplar appears to show similarities to black walnut in terms of spacing/light requirements). Examining the canopy of the stand, the phenomenon known as ‘crown shy’ was evident. This is a trait shared by poplars, walnuts and eucalypts whereby branches tend to die off as they begin to touch, preventing the development of overlapping crowns.
Stop two looked at one of the plots in which promising natural regeneration (mostly oak and ash) had been accepted, and limited enrichment planting of oak had taken place. In the 1980s, final crop trees of suitable form and species had been secured by heavy crown thinning. Although results have been variable between plots, this treatment has generally given rise to an acceptable crop at little expense. During the 1980s this plot (amongst others at Ampfield) had also been the site of a set of trials looking at the inhibition of epicormic growth, the stems of individual trees having been wrapped in black polythene for a period of up to four years. The results of these trials suggested that this treatment yielded limited benefits – although many of the oak trees do now lack significant epicormic growth, lenticel production in test trees was hugely increased (a more porous wrapping material may have prevented this), and the trial was visually unappealing.
At our next stop, neglected woodland had been cleared during the establishment of the original experiment, and the site replanted with Douglas fir. A huge time and labour investment was required as part of this approach to woodland improvement – around 1000 man-hours per hectare over the first 15 years. Future management of this stand will now centre on the promotion of natural regeneration in a move towards continuous cover forestry. Looking to an adjacent plot, in which enrichment planting with oak had been carried out sixty years ago, the stand was now overstocked and productivity low. Linking with the third WSG meeting of 2016, yet to come, Andy Poore and David Pengelly outlined how diversity may be introduced to this one-tier stand. Working with the receptive planting bed maintained by a hazel understorey, underplanting and fencing may now be the most suitable approach (deer pressure is high). This should be balanced with a carefully considered opening of the canopy aimed at fulfilling the light requirements of the desired understorey species only.
Our afternoon session examined two additional sites, distinct from the experimental plots visited during the morning session.
Our first stop of the afternoon visited a site which had been felled, raked and burned in 2002 as part of an early scheme to restore PAWS using a clearfell approach (since identified as lacking merit). Mixed conifer and broadleaved natural regeneration has since become well established at the site. A local FC decision was made to carry out enrichment planting of broadleaves (policy at the time was to fell and allow natural re-colonisation), with the aim of expanding the mixture of species. Regeneration at the site is now prolific and diverse. In terms of the future management of the site, re-spacing will continue to be carried out, which is likely to yield limited mixed firewood. Comparisons were drawn between this, and privately owned woodland where re-spacing in such a stand would be hard to justify financially. The point was raised that creating diversity whilst maintaining production can be difficult to achieve together. Discussions at this stop ended as they so often do, on the topic of grey squirrels. A strategic decision was made not to plant beech at the site due to the inevitable impact of grey squirrels. Around the country, the effective restoration of ancient woodland continues to be severely hampered by this destructive pest.
Led by Ian Willoughby (Forest Research) our final stop was at an experimental site where the direct sowing of seed in a clearfell had been undertaken. The aim of the experiment was to look at whether direct seeding could be used to improve diversity and potential for timber production, rather than relying on either planting or natural regeneration. Clearfelling had taken place in 2005, after which four treatments had been applied to separate blocks in 2006:
Direct sowing has been considered as an alternative to planting in the afforestation of lowland sites. This is because the practice is potentially cheaper than planting (around two thirds of the cost), the site may be captured more easily with less herbicide use required, and the quality of timber can be increased due to the close spacing at planting. However, there are potential problems. The success of direct sowing can be unpredictable, in certain situations it is unsuitable (e.g. where seed predation is high, or where soils are particularly heavy), and certain seeds (e.g. acorns) can be expensive.
Examining the separate treatments ten years after establishment, it was evident that direct sowing at this site had not been particularly effective. Treatment A (rake, burn and fence) had given rise to profuse regeneration of birch, while planted trees (treatment B) appeared to be growing poorly and of poor form. Treatment C (direct seeding) had also produced trees of poor form, with oak having performed particularly badly (both in terms of survival and growth). The most effective treatment appeared to be the untreated control (D), being the cheapest to establish and now comprising a varied mixture of broadleaved and coniferous trees of acceptable form.
The creation of this experimental site represented a significant investment, and unfortunately this investment could not be maintained beyond the establishment phase. Had stands been thinned and manipulated post-establishment, the poor growth and form evident within plots may have been improved, and the potential for further assessment enhanced. Thus, an important point was illustrated - if initial investment in such projects is to be made, it is often important to maintain such investment.
Current research into direct seeding is now underway in North West Scotland, focussing on low productivity upland sites and riparian zones.
Our visit to Ampfield Wood offered the WSG a unique opportunity to assess the outcomes of a number of experiments, some up to sixty years after their creation. Over the course of these experiments, views and policy priorities have changed significantly, impacting the silvicultural decisions being made. Such views and policies will continue to change. What remains unchangingly invaluable is the role of Forest Research in recording and archiving information, allowing knowledge and information to be passed on whatever the prevailing trend of opinion. This was one of the key points brought to light during our day at Ampfield Wood. As well as the clear value of long-term research, it was apparent silviculturally that in some situations limited intervention can be a sensible option where the owner or manager simply aims to make the best of what they have, and to work with what nature already affords.
Evans, J. (1984). Silviculture of broadleaved woodland. Forestry Commission Bulletin 62. HMSO, London.
Willoughby, I., Jinks, R.L., Gosling, P.G., and Kerr, G. (2004). Creating new broadleaved woodlands by direct seeding. Forestry Commission Practice Guide. HMSO, London.