본문 바로가기
읽기 · 학술 · 공개 라이선스 원문

Natural disturbance impacts on ecosystem services and biodiversity in temperate and boreal forests

Thom, Seidl · 2015

공개 학술지 Biological Reviews of the Cambridge Philosophical Society 에 실린 총설 논문입니다(2015년). CC BY 4.0 으로 공개된 글이고, 그림 · 표 · 참고 문헌 표시를 뺀 본문을 실었습니다.

출처 Thom & Seidl (2015), Biological Reviews of the Cambridge Philosophical Society · 라이선스 CC BY 4.0 · 원문

장 3개 올림 이 장 2280낱말
← 앞 장

이 장은 원문만 있습니다. 문장별 해석은 앞 장부터 차례로 붙입니다.

3장 · Natural disturbance impacts on ecosystem services and biodiversity in temperate and boreal forests — Discussion

What we know about disturbance impacts on forest ecosystems

We investigated disturbance effects of fire, wind, and bark beetles in a search for general differences in disturbance impacts on ecosystem services and biodiversity. The large number of studies available for analysis not only indicates the importance of disturbance impacts to forest ecosystems, but also provides a suitable basis for a global synthesis on disturbance effects. The increasing number of publications over time may represent a response of the scientific community to the increase in disturbance frequency observed in recent decades (Westerling et al., 2006; Seidl et al., 2014), and should imply a growing understanding of disturbance processes. However, while disturbance impacts on biodiversity are increasingly well researched, we found more variability in information on different ecosystem services. While the main focus of the reviewed papers was on regulating services (predominately on C storage as an important mechanism of climate regulation), supporting and provisioning services are less well studied. The disturbance impact on cultural services has barely been assessed to date (see online Fig. S1).

In addition, the information available on disturbance impacts also differs with disturbance agent and region. The impact of fire on biodiversity and ecosystem services is the most intensively studied disturbance agent, reflecting the dominant role of wildfire in disturbance regimes around the globe (e.g. Conard et al., 2002; Schelhaas et al., 2003; Littell et al., 2009; Newton et al., 2011; Knox & Clarke, 2012). Regional differences were apparent in our database of published studies on disturbance impacts: Asia, for instance, is underrepresented in our analysis; we found only 10 unique studies on disturbance impacts on biodiversity and 11 on ecosystem services for that continent. However, it has to be noted that not the entire geographic imbalance in disturbance studies is likely to be related to regional differences in scientific understanding of disturbance processes. The main cause of such variation in peer-reviewed information available from different regions is likely to be the language barrier (Powell, 2012). Differences in local research agendas are also likely to play a role (see e.g. Kajala & Watson, 1997). Nonetheless, we advocate research programs that facilitate a broader study of disturbance effects (geographically as well as in terms of the indicators studied), in order to close some of the remaining gaps in our understanding of the role of disturbances in forest ecosystems.

Challenges for synthesizing disturbance impacts

One challenge for a global synthesis lies in a comparison of the different methodological approaches used to study disturbance impacts. Simulation approaches appear to underestimate the effect of disturbances on biodiversity perhaps because current disturbance models are rarely able to assess effects on diversity over a broad variety of guilds. Future improvements in simulation modelling should thus aim to capture the multiple impacts of disturbances better on ecosystems and their diversity (see also Seidl et al., 2011a). Another interesting finding was that expert knowledge differed significantly from the results of empirical studies. Part of this difference could be explained by expert knowledge being reported for different systems and contexts, i.e. systems and indicators that are less well represented by empirical studies. However, the finding that disturbance impacts estimated by experts are more negative on ecosystem services and more positive on biodiversity than those estimated empirically strongly suggests that expert opinions should be omitted from further quantitative analysis (Whittaker, 2010). It should also be noted that our data – like most published literature reviews – are likely to incorporate a degree of publication bias (Møller & Jennions, 2001), i.e. neutral or mixed effects are likely to be underrepresented.

A second challenge relates to the general ability to synthesize the published literature. Although we found a large number of papers dealing with disturbance impacts on biodiversity and carbon storage, only a limited number (18.4 and 22.4%, respectively) could be used in a quantitative meta-analysis. In most instances we had to exclude studies due to inconclusive reporting of disturbance severity, or the absence of a proper control, consequently making it impossible to quantify the disturbance effect. We thus call for better reporting, especially the inclusion of summary statistics in publications, and advocate a BACI (before – after, control – impact) design (Stewart-Oaten, Murdoch & Parker, 1986) to facilitate future syntheses on this topic. The increasing requirement to make the results of studies available upon publication, either as an electronic supplement or in archiving services such as Dryad (http://datadryad.org/) should benefit such syntheses in the future. However, some variation in the choice of an appropriate control to disturbed systems is likely to persist, as, for example, the definition of ‘old-growth’ conditions often differs regionally. Note also that historic land-use and management practices may influence reference conditions as well as disturbance drivers and impact (e.g. Carcaillet et al., 2009), an aspect that cannot be rectified in a global review and meta-analysis such as that presented here.

Another difficulty for synthesis and generalization arises from the inherent complexity of disturbance regimes in temperate and boreal forests (see also White & Jentsch, 2001). While we studied three of the most influential disturbance agents globally, other agents of high regional significance were not considered. For example, ash dieback, a disease affecting common ash (Fraxinus excelsior L.) trees of all age-classes, is currently strongly impacting forest ecosystems in many European countries (Halmschlager & Kirisits, 2008; Ogris, Hauptman & Jurc, 2009), but was not included in this analysis. Our first analysis step revealed significant differences in impact among disturbance agents, documenting that the unique ecology of every agent is important for understanding its effects (e.g. which trees are affected and how). In the second step of our analysis we included severity and time since disturbance as covariates in order to generalize across agents in our meta-analysis. Tests of this generalization assumption show that differences among agents could be explained satisfactorily with these two covariates (data not shown), enabling a statistical analysis across agents and scales. This underlines the potential for a process-based analysis of disturbance regimes in synthesizing knowledge from individual observations to reach general patterns and principles (Turner et al., 1993; White & Jentsch, 2001; Miller et al., 2011; Seidl et al., 2011a).

However, this ability to generalize might to some degree be attributed to the inclusion of only temperate and boreal forest ecosystems in our data set. Whether the general patterns deduced for these biomes also hold for tropical forests remains to be tested. Martin, Newton & Bullock (2013), conducted a review on the effects of anthropogenic disturbance on carbon stocks and plant diversity for more than 600 secondary forest sites in the tropics. They show that both biodiversity and carbon storage were negatively affected by clearing (a high-severity disturbance), and took several decades to recover. Assuming that salvage logging after natural disturbance results in an impact comparable to anthropogenic clearing we here find contrasting results for biodiversity effects in temperate and boreal forests: our data suggest a weak positive effect of disturbance on biodiversity (not significantly affected by salvage logging, P = 0.205). This indicates that further studies are needed to establish whether the disturbance paradox described here also applies to tropical forests.

The existence and strength of simultaneous positive and negative impacts of disturbances on objectives of ecosystem management, described here as the disturbance paradox, might not only vary geographically but is likely also strongly dependent on the indicators selected for analysis, and hence the local relevance of specific ecosystem services and aspects of biodiversity. Generalist species might, for instance, benefit strongly from disturbance events while specialists and late-seral species – which are often a priority for conservation – could be negatively affected (Devictor & Robert, 2009). Moreover, disturbances might benefit invasive alien species (see e.g. Crawford et al., 2001), widely regarded as negative for biodiversity. Owing to the broad scope of this study such aspects were not explicitly considered in our analysis. They might, however, be of high relevance in local assessments and management decisions, and could thus strongly modify the disturbance paradox, described here based on a global synthesis for boreal and temperate forests. A context-specific assessment of biodiversity effects at the level of guilds, red-listed species, and alien/native/endemic species in future studies is thus suggested in order to scrutinize further the generality of the disturbance paradox presented here.

The disturbance paradox and how to address it in ecosystem management

We found strong evidence for the existence of the disturbance paradox in our global analysis of disturbance impact. Disturbance effects on ecosystem services and biodiversity clearly differ in the published literature, with ecosystem services being overall negatively affected while biodiversity is predominately positively influenced by natural disturbances. Our meta-analysis of the disturbance effect on species richness and total ecosystem carbon storage aptly illustrates this paradox: while species richness increases by 35.6% on average for a high-severity disturbance event, a simultaneous loss of 38.5% of total ecosystem carbon storage is to be expected. When management goals are to increase carbon storage while at the same time fostering biological diversity, managers are faced with ambiguity with regard to assessing the impact of a disturbance event, and gauging the implications of future disturbance regimes. Are disturbances to be prevented (as far as possible) to reduce negative impacts on ecosystem services, or are they to be welcomed and incorporated into management due to their positive effects on biodiversity?

While our global study cannot resolve this paradox of ecosystem management – which needs to be addressed in the local context of stakeholder preferences, habitat quality, and other constraints – several interesting insights for disturbance management can be deduced from our analysis. Since negative disturbance impacts on carbon storage are strongly reduced with time since disturbance, but positive effects on biodiversity do not vary significantly over time, our global meta-analysis suggests that managing for a low- to medium-frequency disturbance regime would result in limited impacts on provisioning services while still benefiting biodiversity. In other words, our data indicate that the disturbance event itself matters for biodiversity, while having enough time between these events ensures recovery of ecosystem services. Albeit not significant in our analysis, the same is true with regard to severity, i.e. moderate- or mixed-severity disturbances (see e.g. Perry et al., 2011) are likely to be the best balance between negative effects on ecosystem services and positive effects on biodiversity. Traditional disturbance management approaches such as salvage harvesting and prescribed burning, for instance, are not able to moderate between negative ecosystem service impacts and positive diversity effects according to our analysis. We even found a higher proportion of papers reporting negative effects from prescribed burning on ecosystem services provisioning compared to wildfire. However, due to sample-size limitations we were not able to analyse these data for differences in effect size, although differences in severity (i.e. mean severity over all studies for prescribed burning = 26.2%, wildfire = 88.1%) suggest a positive effect of prescribed burning (Hurteau & North, 2009; Meigs et al., 2009).

Ongoing climatic changes will likely increase disturbance frequency and severity in many parts of the world (Li et al., 2013; Temperli et al., 2013; Seidl et al., 2014) which – according to our findings – may have negative implications for ecosystem service provisioning. Hence, adaptation of forest ecosystems to such changes in disturbance regime is of great importance in current forest ecosystem management, in order to sustain future ecosystem services provisioning to society. However, as many important drivers of the disturbance regime such as species composition respond to management changes only on time scales of decades to centuries (e.g. Hicke & Jenkins, 2008; Thom et al., 2013), such management considerations need to take long lead-times into account. On the other hand, our analysis indicates that intensifying disturbance regimes may also represent an opportunity to foster biodiversity in forest ecosystems, and might thus to some degree alleviate the ongoing biodiversity crisis (Stuart et al., 2004; Thomas et al., 2004). In this context it is interesting to note that more diverse ecosystems are often more resistant and resilient to disturbance impacts (Bengtsson et al., 2000), so that in the long term disturbance effects on ecosystem services might be buffered by increasing structural and compositional diversity.

CONCLUSIONS

(1) Over the last decades, the number of peer-reviewed publications on forest disturbances and their effects has increased, mirroring the increasing relevance of disturbance regimes and the changes therein. However, the available literature is heterogeneously distributed over agents and regions, with most studies addressing forests in North America and Europe, and mainly focusing on fire impacts.

(2) Disturbances in forest ecosystems can have both positive and negative impacts on objectives relevant to ecosystem management. We here find that ecosystem services of all four categories defined by the MEA (2005) (provisioning, supporting, regulating, and cultural) are predominately negatively impacted by natural disturbances. Biological diversity, as represented by species richness, habitat quality, and diversity indices is, on the other hand, predominately positively affected by natural disturbances.

(3) In a meta-analysis we determined that on average a disturbance event decreases total ecosystem carbon by 38.5% (standardized coefficient for a stand-replacing disturbance event in the year of the disturbance), while species richness increases by on average 35.6%.

(4) For ecosystem management, which aims to provide ecosystem services sustainably to society while preserving and fostering biodiversity, these divergent disturbance impacts present a paradox – they are at the same time risk factors and facilitators of management objectives. Our analysis suggests that measures of disturbance management such as salvage logging and prescribed burning do not significantly moderate these diverging impacts. However, a meta-analysis of carbon storage (an important regulating service in the context of climate change mitigation) and biodiversity suggests that managing for a disturbance regime of low to medium frequency and severity could limit impacts on ecosystem services while still being beneficial for biodiversity.

(5) Our review suggests that intensifying disturbance regimes under climate change will largely benefit biological diversity of forest ecosystems. Ecosystem services provisioning on the other hand will mostly be negatively impacted by such changes in the disturbance regime. This might require a timely adaptation to changing disturbance regimes in order to provide important ecosystem services sustainably in the future.

이 글은 위 출처의 원문을 그 라이선스(CC BY 4.0)에 따라 옮겼습니다. 참고문헌 목록과 인용 번호, 그림 · 표 · 각주 · 수식 · 부록은 빼고 절별로 나누어 실었으며, 그 밖의 문장은 원문 그대로입니다. 원 저작자와 발행 기관은 이 화면을 만들거나 검토하지 않았고, 이곳의 해석과 낱말 풀이는 공식 번역이 아닙니다.