In this post Associate Editor mentee Lander Baeten discusses a paper he handled by Beth Atkinson and colleagues ‘A comparison of clearfelling and gradual thinning of plantations for the restoration of insect herbivores and woodland plants’

Since the pioneering work of George Peterken in the 1970s, numerous studies have shown that many forest plant species are extremely slow to re-establish once lost from the ecosystem. Previous work has principally compared the forest ground flora of native forests on ancient forest sites versus plantations on former agricultural land. Results consistently showed that the abiotic legacies (e.g. soil nutrients) and biotic legacies (e.g. seed bank depletion) of past agriculture create persistent thresholds for plant community recovery. It is even likely that post-agricultural forests will never fully recover. However, plantations can be ancient forest as well – yes, even conifer plantations – if they were planted directly after cutting the native forest without some intermediate agriculture. Ancient forests are sites that have been continuously forested at least since a certain threshold date (e.g. 1600 AD in England and Wales). The concept thus refers to the land-use history of a site, not the age and composition of the tree population. It is not a synonym for old growth or primary or virgin forest.

Although ‘ancient’ is a prerequisite for floristic diversity, it is not necessarily a label of actual habitat quality. Indeed, the plant community in conifer plantations on ancient forest sites is often degraded relative to the previous native forest. Detrimental tree species effects on the environment (e.g. soil acidification under conifers) and a changed forest management regime may profoundly influence the ground flora. Yet, conifer plantations are excellent candidates for restoration of native forest because, unlike post-agricultural forest, planting the conifers has not immediately pushed the forest across the same a/biotic thresholds. Remnant populations of typical forest species may still persist and soil nutrients are not dramatically altered. A crucial question for managers then is: how can we restore native forest on ancient forest sites planted with conifers without eradicating remnant plant populations? In their recent paper, Beth Atkinson and colleagues exactly explored this challenge. In the Forest of Dean (UK), they compared existing conversions of conifer plantations on ancient forest sites with similar plantations not undergoing restoration (‘pre-conversion’ state) and native forest stands (‘reference’ state). Two strategies for conversion were compared: clearcutting and replanting with native broadleaves versus gradual removal (thinning) of the conifer trees. Besides its clever design, the study adds to what we know and what more we need to find out in at least three important ways, which I will explain below.

The ground flora of ancient forests can be remarkably species-rich (Le Haute Pays, Belgium)

1–The associated species diversity in tree plantations has been widely studied, but most emphasis has been on planted forests on post-agricultural sites (see above, here and here for overviews). Results of Atkinson’s study indicate that conifer plantations on ancient forest sites can support an equally diverse ground flora compared with native forest, even if we only look at strict forest species. The two restoration methods (clearfelling, thinning) at least maintained the total richness and strict forest plant species richness. Clearcutting even increased total species numbers. It seems therefore crucial to consider the actual land-use history of a site, rather than simple dichotomies such as conifer versus native, if we want to understand contemporary and future community patterns. The study therefore provides new empirical data supporting our recent framework on land-use legacies and vegetation dynamics.

2–In the study of land-use legacies in forests, previous studies have almost exclusively looked at plants. Other groups have mostly been neglected, especially the taxa of higher trophic levels that directly depend on the plant community. Atkinson and colleagues sampled leaf-mining insects in the same plots as the plant community and showed that insect richness increased with plant species richness in most management regimes, as expected. In the clearcutting conversion scenario, however, the relationship was negative with higher plant diversity leading to lower numbers of insect herbivore species. After clearcutting, vegetation growth was more vigorous and in a more dense vegetation the insects have more difficulties finding their host plants because they are hidden. Considering more than one taxonomic group may thus lead to a different choice of conversion method. Clearcutting seems favourable for plant richness (total richness is even highest in this regime), but reduces invertebrate richness. Other groups such as butterflies are maybe more likely to occur in open clear-cut patches. I couldn’t agree more with the authors who conclude that taxa other than plants need to be considered when picking a conversion method.

3–Finally, the restoration management methods were evaluated in two contrasting forest types: acidic oak forest and mesotrophic ash forest. The efficacy of restoration appeared equal for both methods and both taxonomic groups. The success of the conversion management may thus be generalizable across forest types. The question is, however, whether these findings can be translated to other (temperate) regions as well? This is a question about the broader context dependency of the results, that is, beyond the forest type effect. We know that in many temperate forest landscapes a chronically increased nitrogen deposition has led to accumulated nitrogen in the soil. Below a dense overstorey canopy, the ground vegetation is light-limited and does not respond so much to the nitrogen build-up. The ‘nitrogen time bomb’ may however explode when the canopy is opened up, and in this case, the impact of completely removing the canopy in a clearcutting regime versus gradually opening the canopy with thinning may actually become very different. Similar reasoning is obviously also relevant for plantations on sites that have received nutrients in a different way (e.g. in post-agricultural forests).

In sum, I believe this study provides new information that is directly relevant for managers, but also leads to new questions such as the impact of these restoration alternatives on other taxa or the generality of the results for forests in more intensive landscapes or with a more intensive land-use history.