The ability of short-term responses to predict the long-term consequences of conservation management actions: The case of the endangered Paeonia mascula (L.) Mill.
Introduction
Loss of biodiversity is one of the ecological hallmarks of human activity in recent times (Dirzo & Raven, 2003). Conservation strategies aim to maintain biodiversity while helping some species to expand their distributions and colonize new habitats (Butchart et al., 2012). These goals can be achieved by preventing damage to habitats or by improving habitat quality as part of a well-defined management program, taking into consideration different levels of ecological hierarchies starting from the species and up to the ecosystem level (Perevolotsky, 2005). Due to the role of biodiversity in a wide array of ecological processes and its importance to environmental stability, it is not surprising that active management is a common practice in biological conservation (Kovac, Hladnik, & Kutnar, 2018; McCarthy & Possingham, 2007). The rationale is that human activity has negative effects on ecosystems and therefore human intervention is needed to minimize this damage.
Conservation actions executed as part of an adaptive management plan (Holling, 1978) can be positioned along a pulse-press perturbation spectrum (Bender, Case, & Gilpin, 1984). At one extreme, pulse perturbation management practice involves short-term interventions aiming to shift the ecosystem from its current undesired state to a more desired one. At the other extreme, press perturbation practice involves continuous management of the ecological system in order to maintain it in a desired state.
Due to the complex nature of ecological systems, it is often very hard to predict how conservation management actions will play out through time. For this reason it is widely agreed that successful management should be science-based rather than based on intuition and/or expert opinion (Fazey, Fazey, Salisbury, Lindenmayer, & Dovers, 2006). Therefore, conservation efforts are usually based on preliminary experiments that test-run some measurable effects of the evaluated management protocol. However, such experiments are often short-term and limited in their spatial scope; partially to minimize possible damage to an already damaged ecosystem. Since environments are often spatially heterogeneous it is recommended to spread the experiments over a variety of habitats better representing the system’s complexity, such that more spatially comprehensive and general insights can be achieved (Possingham, Franklin, Wilson, & Regan, 2005).
Unlike spatial variability, temporal variability is harder to study because of the long time required to understand long-term processes (Young et al., 2010). This limitation is especially critical in conservational studies. Consequently, although it is widely accepted that ecosystem responses to perturbations should be examined over long timescales (Birkhead, 2014; Hastings, 2010; Hughes et al., 2017; Likens, 1989; Lindenmayer et al., 2012), most often the employed conservation actions are based on short-term studies that were initiated after a problem was recognized, leaving relatively little time to spare and requiring quick answers. These limitations lead to the somewhat unavoidable practice of planning conservation actions using short-term data. When doing so, we implicitly assume that the short-term responses quantified in the short-term study can be extrapolated into the future. This, however, might not always be the case.
Four qualitatively different scenarios can follow a short-term successful management action. Over time, the short-term positive effect (presented from the most desired to the worst) can be: a) enhanced, b) maintained, c) diminished (the system returns to its pre-management state, d) overturned (the system sinks below its pre-management state) (Fig. 1). While the first two scenarios are desired, the last scenario can result in risk to the system’s perpetuity, requiring further management actions (Williams, 2011). The last scenario is by far the worst since it is only in this case that the management action actually causes net-damage to the system. The likelihood of these four different scenarios is related to the stability of the post-management system state (Hobbs, Hallett, Ehrlich, & Mooney, 2011).
The stability of post-management plant communities is affected by their position along the systems successional trajectory (Niering, 1987). Specifically, effects that shift the community into early successional stages are likely to result in relatively unstable states changing within short time scales. In such cases, the positive short-term responses (detected in the short-term perturbation experiment) can diminish over time as the system shifts along its successional trajectory. This problem can be solved by repetitive cycles of the management actions to maintain the system in its desired state (i.e., a press perturbation management action). Without such repetitive management actions, short-term positive results can turn into long-term negative consequences in cases where the positive effects not only diminish, yet rather are overturned in the long-term (Fig. 1b–c). It is in these cases that the net effect of management is negative, and the system would have been better off untouched.
To this end, we suggest that actions that shift the system into a new stable state are less prone to erroneous interpretation of short-term results. However, if processes in the new stable state are more sensitive to rare events (unlike in the pre-managed state), taking place after and not during the short-term study (that led us to think that the management has positive results), then our ability to understand and predict the long-term dynamics will be jeopardized (Fig. 1b). Since some rare events have the potential to reduce the ecosystem performance below that of the pre-management state (Fig. 1b), relying on short-term data can result in net-damage to the ecosystem we are trying to protect.
We present a case study comparing the long-term effects of a pulse perturbation management action involving canopy opening, on the population density and reproduction of the locally endangered southernmost population (Shmida, Pollak, & Fragman-Sapir, 2007) of Paeonia mascula (L.) Mill. (Paeoniaceae), inhabiting a small area in Mount Meron, Israel. Paeonia mascula is endangered in Israel due to its rarity, limited distribution and its attractiveness (Shmida et al., 2007). Although all populations are within the Mt. Meron Nature Reserve, farming and forging roads are a threat to the plant’s populations. In the past 70 years, grazing and tree harvesting have been minimized and vegetation formation in many areas has turned into closed and dense woody vegetation (Hughes et al., 2017).
In a previous study, Ne’eman (2003) have marked ten 25 m² plots with non-flowering P. mascula plants located under closed canopy. Five of the plots were cleared from trees in 1997, and five plots remained closed (control). Tree-cutting increased the radiation to ∼70 % light availability in plots cleared from trees, while also increasing the proportion of flowering P. mascula plants over a 4-year period. However, it is unknown how this pulse perturbation management action affected the P. mascula population over a longer time scale. The aim of the present study was to examine the long-term effect of the pulse (past) perturbation management action, on the population density, soil seed bank and flowering of P. mascula, and on the germinable soil seed bank density and richness of P. mascula neighboring species. We hypothesized that the short-term increase in the percentage of flowering P. mascula plants (Ne’eman, 2003) would be reversed (lower than pre-treatment levels) due to the successional change expected in this system. Specifically, dense foliage and complex regrowth of sprouts occurring in the first few decades of post clearcutting succession could result in reduced resource availability including light (Clarke et al., 2013; Radim et al., 2020). This reversion would be evinced by the reduction of P. mascula population size in plots that were cleared from trees in 1997. Therefore, we hypothesized that P. mascula population size components, such as plant density and percent of flowering plants, should be smaller in the cleared plots.
Since very little data exists regarding the longevity of the soil seed bank of P. mascula (L.) Mill., we were unable to hypothesize whether the density of its seed bank would be higher or lower in the treated plots.
Section snippets
Study site
The study area was located in Mt. Meron nature reserve (35°25'E, 32°58'N), northern Israel (Fig. 2). Since it was declared a nature reserve in 1965, most of the woodland canopy in this area has gradually closed to form a dense woody canopy (i.e., an increase of more than 30 % in tree cover), mainly due to a reduction in grazing and deforestation (Carmel & Flather, 2004; Carmel & Kadmon, 1999; Ne’eman, 2003). The area is characterized by an east Mediterranean climate with hot and dry summers and
Canopy closure
LAI measurements did not differ significantly between cleared and control plots (z = 0.55, P = 0.582; Fig. 3d, Supporting Information).
Paeonia mascula population
In total, 300 and 366 P. mascula plants (including both mature plants and seedlings) were recorded in the study site in 2017 and 2018, respectively. The density of P. mascula plants was significantly higher in control than in cleared plots (z = 2.390, P = 0.017; Fig. 3a). Specifically, the total density of P. mascula in control plots was ∼290 % and ∼420 % higher
Discussion
The data utilized for making conservation management decisions often originates from short-term studies. In cases where short-term responses to perturbations do not predict the long-term consequences, a potential for erroneous decision making arises. Our results illustrate that the short-term positive effect of a pulse canopy opening on population of P. mascula observed ∼20 years ago not only diminished over time, rather it was reversed and ended up having a net negative effect on the
Article impact statement
Conservation actions that look promising in the short-term might result in net-damage to the environment in the long run.
Declaration of Competing Interest
The authors report no declarations of interest.
Acknowledgement
The authors wish to thank Itamar Giladi, David Saltz and Yohay Carmel for their critical review of earlier versions of the manuscript. We thank Gidi Ne’eman for giving us the information regarding the plots. We would also like to thank the professional crew of the Israel Nature and Parks Authority for their help in the field, with special thanks to Dr. Iftach Sinai,the Upper Galilee ecologist. We are also grateful to Ella Dagon, Nitzan Malachy and Ravid Sapir for their help with the soil seed
References (48)
- et al.
Population dynamics of Paeonia officinalis in relation to forest closure: From model predictions to practical conservation management
Biological Conservation
(2017) - et al.
Positive interactions in communities
Trends in Ecology & Evolution
(1994) - et al.
Management type affects composition and facilitative processes in altoandine dry grassland
Acta Oecologica
(2013) - et al.
Biodiversity in (the Natura 2000) forest habitats is not static: Its conservation calls for an active management approach
Journal for Nature Conservation
(2018) To be or not to be—The effect of nature conservation management on flowering of Paeonia mascula (L.) Miller in Israel
Biological Conservation
(2003)- et al.
Short-term successional change does not predict long-term conservation value of managed arable field margins
Biological Conservation
(2010) Adaptive management of natural resources—Framework and issues
Journal of Environmental Management
(2011)- et al.
Response of photosynthetic plasticity of Paeonia suffruticosa to changed light environments
Environmental and Experimental Botany
(2003) - et al.
Effect of successional stages on the establishment of Quercus calliprinos in an East Mediterranean maquis
Israel Journal of Plant Sciences
(1996) - et al.
Forest-cutting rapidly improves the demographic status of Paeonia officinalis, a species threatened by forest closure
Population Ecology
(2013)