Response of Two Dominant Boreal Freshwater Wetland Plants to Manipulated Warming and Altered Precipitation

This study characterized the morphological and photosynthetic responses of two wetland plant species when they were subject to 2–6°C fluctuations in growth temperature and ±50% of precipitation, in order to predict the evolution of natural wetlands in Sanjiang Plain of North-eastern China. We investigated the morphological and photosynthetic responses of two dominant and competitive boreal freshwater wetland plants in Northeastern China to manipulation of warming (ambient, +2.0°C, +4.0°C, +6.0°C) and altered precipitation (−50%, ambient, +50%) simultaneously by incubating the plants from seedling to senescence within climate-controlled environmental chambers. Post-harvest, secondary growth of C. angustifolia was observed to explore intergenerational effects. The results indicated that C. angustifolia demonstrated a greater acclimated capacity than G. spiculosa to respond to climate change due to higher resistance to temperature and precipitation manipulations. The accumulated effect on aboveground biomass of post-harvest secondary growth of C. angustifolia was significant. These results explain the expansion of C. angustifolia during last 40 years and indicate the further expansion in natural boreal wetlands under a warmer and wetter future. Stability of the natural surface water table is critical for the conservation and restoration of G. spiculosa populations reacting to encroachment stress from C. angustifolia expansion.


Introduction
Wetlands are among the most important terrestrial carbon pools and play an important role in global carbon cycling [1,2]. Boreal wetlands comprise about half of the total global wetland area, store about one third of the world's organic carbon in the form of living or partially decayed vegetation and soil organic carbon and contribute 34% of global atmospheric methane flux [3,4]. Due to their large and dynamic carbon pool, boreal wetlands may not be only significantly impacted by climate change, but may also provide negative (e.g. increasing organic carbon sink) or positive feedback (e.g. increasing greenhouse gas emissions) mechanisms to anthropogenic climate change [5,6] and mitigate climate warming overall.
Boreal wetlands are considered particularly vulnerable to climate change because they depend on specific climatic conditions with low temperature and high water availability [7,8]. To quantify anticipated climate effects on boreal wetland ecosystem processes, field experiments simulating climate change have been undertaken in North America and northern Europe [2,5,9], and these studies suggested that both the performance of a specific plant and the compositions of plant communities will change in different directions and at different magnitudes as a response to warming and/or changes in the depth of the surface water table that usually connected with temperature and precipitation change. Despite these efforts, species-specific eco-physiological responses of wetland plants to changes in temperature and precipitation, and especially the interactive effects of these factors, remain unclear. Simulations provide a mechanism for exploring the importance of these temperature-precipitation interactions on wetland plants under future climate conditions [10]. Multifactor experimental manipulations of temperature and precipitation in the laboratory can complement field studies by enabling controlled adjustment of covarying climate variables based on the historical data to generate relatively realistic climate change experimental scenarios [11]. The responses then could be scaled up to the level of the ecosystem when the initial structure of the wetland and the environmental driving variables are consequently well coupled.
As typical dominant boreal freshwater wetland plants in Northeast China, Siberia and Far East Russia, Calamagrostis angustifolia and Glyceria spiculosa both belong to the family Poaceae, and are often distributed as large, dense clumps in palustrine, riverine and lacustrine wetlands [12,13]. Generally, C. angustifolia covers saturated microhabitats while G. spiculosa covers the flooded microhabitats with greater surface water depth. Both plants play important ecological functions in the wetlands as primary producers for food webs, shelter for some invertebrates and major organic carbon sources for carbon sequestration. In recent years, with climate warming and surface water deficiency in some wetlands of the Sanjiang Plain, the microhabitats once covered by G. spiculosa have been almost replaced by C. angustifolia due to the drawdown of surface water depth, caused either by climate change or anthropogenic activities such as drainage.
In this paper, we simultaneously considered two major factors (temperature and precipitation) involved in climate change effects on boreal wetland ecosystem functioning. The two plant species, C. angustifolia and G. spiculosa, were incubated from seedling to senescence in environmental chambers under manipulated warming and altered precipitation conditions. The objectives of this study were: (1) to characterize the morphological and photosynthetic responses of the two plants to 2-6uC warming and 650% of precipitation changes; and (2) to analyze the resistance and acclimation of the two plants and predict their potential distribution in the future under the projected changed climate conditions in the Sanjiang Plain wetlands of Northeast China.

Study site and target species
The study site is located in Xingkai Lake (Khanka) lacustrine wetlands of Sanjiang Plain, the largest expanse of freshwater wetlands in China. Xingkai Lake is the largest lake in Northeast Asia and a transboundary lake shared by China and Russia. According to calculations based on the 54 years of local meteorological records, the annual mean precipitation is 561 mm and the temperature ranges from 239uC in January to 36uC in July with the annual mean of 3.5uC for 1957-2010. According to a linear fitting of local meteorological records in Xingkai Lake lacustrine wetlands (Annual temperature = 252.55 + 0.028*year, r.0.99, p,0.0001; unpublished), there is a significant warming trend of 0.28uC per decade in Xingkai Lake over the past half century. This is even greater than the average level of 0.22uC per decade for the whole country [14].
The target plant species for the three-month incubation experiments were collected in situ from Calamagrostis angustifolia (45u20959" N, 132u18952" E) and Glyceria spiculosa (45u21949" N, 132u19926" E) communities within Xingkai Lake lacustrine wetlands. Representative and homogeneous individuals of C. angustifolia and G. spiculosa were excavated with intact roots and soils using a planting shovel in May of 2011 and 2012, respectively, when the first new shoots were apparent in springtime. The soil types were silty clay meadow marsh soil and humus marsh soil, respectively. The seedlings and soils with the depth of approximate 20 cm were cut into small blocks with rhizome based, transported to the laboratory in Northeast Institute of Geography and Agroecology within two days, and transplanted into incubation pots with 10-cm diameter and 15-cm depth with 4-5 seedlings in each pot. To minimize loss of vitality during transplantation, these potted seedlings were first incubated in a greenhouse for one month to germinate new ramets before placement into environmental chambers. The temperature in the greenhouse was set as the average air temperature of June in Xingkai Lake lacustrine wetlands. All the field studies got the permission from administrative authority of Xingkai Lake National Nature Reserve and did not involve any endangered or protected species.

Experimental design
The design between the warming and the precipitation was a two-way randomized block experiment. Four environmental chambers were deployed to control the temperature. Within each chamber, three plastic container (35 cm 6 25 cm 6 10 cm) were used to control water level fluctuations caused by precipitation manipulation. Each container contained three replicated pots. Before incubation, any pot growing approximate ten vigorous seedlings with similar performance (about 20 cm and 15 cm heights for C. angustifolia and G. spiculosa, respectively) was selected and randomly separated into 4 groups corresponding to the 4 temperature regimes. The ambient temperature was set as the local monthly average air temperatures in July, August and September, respectively fitted. The ambient temperature was adjusted five times per day based on the local monthly (July, August and September) average air temperatures over the time periods 1:00-4:00, 4:00-10:00, 10:00-16:00, 16:00-19:00 and 19:00-1:00. Table 1 provides the temperature and illuminance values for the plant incubations in the environmental chambers.
The illuminance values were based on the field observation. These pots were exchange their positions in the chambers daily to keep the temperature and light intensity similar for a given pot. The warming scenarios were then set according to the ambient temperature regime with +2uC, +4uC and +6uC added for each interval. The lights were plant growth fluorescent lamps that are recognized as suitable for plants' growth [15].
The ambient water depths were set to mimic the natural conditions in Xingkai Lake wetlands, with no net water for C. angustifolia and 10 cm of water above the soil surface for G. spiculosa. The ambient precipitation regime was set according to the local monthly average precipitation. These seedlings were gently watered over one hour using sprinkler nozzles to simulate natural precipitation every 3 days with one tenth of the amount of monthly precipitation. The altered precipitations were set as -50% and +50% based on the ambient precipitation rates.
After the three month incubation in the environmental chambers and subsequent harvesting of all above-ground biomass, one C. angustifolia pot was randomly selected from the 3 replicates of each precipitation treatment container within each environmental chamber. Twelve pots were consequently selected for a secondary growth test to analyze the potential effect of climate change on the next generation of this plant species. These pots were incubated for one month in the same environmental chamber with the average temperature and precipitation in June of 17.4uC and 74.4 mm, respectively in Xingkai Lake wetlands.

Response measurements
The morphological indicators (height, stem diameter and leaf area) were observed once per month using a meter stick, vernier caliper and portable area meter (LI-3000, Li-Cor Inc., USA) from seedling to senescence, with each indicator measured by 4 replicates for C. angustifolia and 3 replicates for G. spiculosa. The photosynthetic indicators, including leaf net photosynthetic rate (Pn), leaf stomatal conductance (Gs) and leaf transpiration rate (Ts) were measured by a portable photosynthesis system (LI-6400, Li-Cor, Inc., USA) during the peak growth phase (around August), with each indicator measured by 4 replicates for C. angustifolia and 3 replicates for G. spiculosa. The chlorophyll content was determined by homogenizing and extracting the leaf tissue in 80% acetone until the green colour turn white completely, then measuring the optical absorbance at 663 and 645 nm for chlorophyll a and b, respectively using a spectrophotometer (UV-2500, Shimadzu, Japan), and calculating concentrations using the specific absorption coefficients [16,17]. The reported leaf chlorophyll content was the sum of chlorophyll a and b, with 3 replicates both for C. angustifolia and G. spiculosa.
Above-ground biomass was measured after plant senescence by drying to constant weight at 80uC. The total nitrogen content of harvested leaves was determined using a continuous flow analyzer (SAN ++ , SKALAR, the Netherlands) after digesting 0.3000 g of homogeneous, milled material with H 2 SO 4 /H 2 O 2 , with 3 replicates both for C. angustifolia and G. spiculosa. The belowground biomass was not measured as it is a mixture of newly developed rhizomes/roots within that growing season and the rhizomes/roots from previous growing seasons. Older material usually accounts for most of the total below-ground biomass [18].
The response measurements for the secondary growth test of C. angustifolia were conducted as described above. The measured indicators were height, stem diameter above the soil, leaf area of each leaf on the plant based on measuring a suitable number of leaves, Pn, Gs, Ts and above-ground biomass, with each indicator measured by 3 replicates.

Statistical analyses
Statistical analyses were performed using SPSS Statistics 21.0 (SPSS Inc., USA). The main and the interaction effects of temperature and precipitation on the two plants' morphological indicators were compared by repeated measures analysis of variance (ANOVA). For repeated measures ANOVA, Mauchly's Test of Sphericity was first examined and the Greenhouse-Geisser adjustment was adopted when the null hypothesis was rejected. The main and the interaction effects of temperature and precipitation on the two plants' photosynthetic indicators, leaf chlorophyll contents during the phase of peak growth, the aboveground biomasses and leaf nitrogen contents after senescence were compared by two-way ANOVA. One-way ANOVA was performed to compare the differences of above-ground biomasses of C. angustifolia and G. spiculosa. Differences in morphology, photosynthesis and above-ground biomass between different secondary grown C. angustifolia after the temperature and precipitation manipulation for one month were also compared using one-way ANOVA. The least significant difference (LSD) was used respectively to perform post hoc multiple comparisons when

Morphological indicators
Repeated-measures ANOVA of morphological indicators showed that the interaction effects of temperature and precipitation on the height of C. angustifolia and G. spiculosa were both significant (F = 2.025, p = 0.044; F = 2.520, p = 0.033, respectively). For both species, main effects of temperature or precipitation, and any interaction effects, on basal stem diameter were all insignificant. The main effects of temperature on leaf area of C. angustifolia and the effect of precipitation on that of G. spiculosa were both significant (F = 2.449, p = 0.047; F = 3.271, p = 0.031, respectively) ( Table 2).
C. angustifolia's height increased from July to September. The estimated marginal means (as the average of the results for each month) of the height suggested that the greatest heights were observed in the ambient precipitation for each warming manipulation (Fig. 1A). Under ambient precipitation conditions (Fig. 1B), the greatest height occurred at ambient temperature (except for the early growth of July), while the lowest height was found in the +2uC manipulation (except in September). When the precipitation decreased by 50% from ambient (Fig. 1C), the greatest height occurred in the +4uC manipulation (except for the early growth of July), while the lowest height was in the +6uC manipulation. When the precipitation increased by 50% from ambient, the greatest height occurred in the +4uC manipulation, while the other three treatments produced plants of comparable height (Fig. 1D).
G. spiculosa's height also increased throughout the incubation (Fig. 2B-D). The estimated marginal means of the height suggested that ambient precipitation and 2-4uC of warming could support greater plant height ( Fig. 2A). In the ambient precipitation conditions, there were no significant differences in height between the different temperature treatments (Fig. 2B). For both the 250% and +50% precipitation treatments, the greatest height occurred in the +4uC manipulation (Fig. 2C, D).
The leaf area of the two species increased first and then decreased throughout the incubation. For C. angustifolia, the leaf area with the +6uC manipulation was smaller than other manipulations in July. There was no significant difference in leaf area between the ambient and the +2uC manipulation in July. In    Table 3. Two-way ANOVA of photosynthetic indicators and leaf chlorophyll content for C. angustifolia and G. spiculosa during the phase of peak growth, and above-ground biomass and leaf total nitrogen content after harvest. September, the leaf area with the +4uC manipulation was not significantly different from either the ambient or the +6uC manipulation. There was no significant difference in leaf area with different temperature manipulations in August (Fig. 3A). For G. spiculosa, there was no significant difference between ambient and the 250% precipitation treatments in any of the three months, while the leaf area with the 50% precipitation treatment was smaller than other treatments in August and September (Fig. 3B).

Photosynthetic indicators and leaf chlorophyll content
Two-way ANOVA of photosynthetic indicators and leaf chlorophyll content showed that the interaction effects of temperature and precipitation on these two species were both significant, except for the net photosynthetic rate (Pn) for G. spiculosa, which was only significantly affected by precipitation (Table 3).
For C. angustifolia, smaller Pn occurred under ambient precipitation except for the +2uC manipulation, and greater Pn was measured in the +6uC manipulation with the exception of the +50% precipitation treatment (Fig. 4A). The leaf stomatal conductance, Gs, decreased with the enhanced precipitation for the +4uC and +6uC manipulations (Fig. 4B). The greatest leaf transpiration rates, Ts, were found in the +4uC and +6uC manipulations under 250% precipitation (Fig. 4C). Ts rates decreased in both these temperature treatments with increasing precipitation. There was no systematic pattern in Ts at ambient temperature but the +2uC manipulation showed Ts increased with increasing precipitation. There were no consistent trends in leaf chlorophyll content with temperature or precipitation treatments (Fig. 4C) but all concentrations fell within the range 1.83-3.49 mg/g. For G. spiculosa, Pn decreased with precipitation (Fig. 5A). With the exception of the +2uC treatment under the 250% precipitation regimen, the highest Gs results were found under ambient temperature (Fig. 5B). There was no consistent response in Gs to increased precipitation over the temperature treatments. There was no consistent trend in Ts across either temperature or precipitation treatments (Fig. 5C). All temperature treatments showed a decrease in leaf chlorophyll content between 250% and ambient precipitation (Fig. 5D). Increasing precipitation to 50% above ambient resulted in no consistent trend in chlorophyll with increasing temperature regime. Above-ground biomass and leaf nitrogen content Above-ground biomass of C. angustifolia was only significantly affected by temperature (F = 4.512, p = 0.012), while the effect of precipitation and the interaction effect were insignificant (Table 3). Biomass was at a maximum in the +4uC manipulation, increasing by 58% above that at ambient temperature (Fig. 6A). Leaf total nitrogen of C. angustifolia was also only significantly affected by temperature (F = 7.125, p = 0.001; Table 3). All three temperature increments resulted in statistically significant increases above ambient temperature, with a maximum value at +6uC, representing a 39% increase above ambient temperature (Fig. 6B).
Above-ground biomass of G. spiculosa was significantly affected both by temperature and precipitation (F = 4.156, p = 0.017; F = 13.637, p,0.0001, respectively), while the interaction effect was insignificant ( Table 3). As noted with C. angustifolia, biomass was highest in the +4uC manipulation, with an increase of 37% above ambient temperature (Fig. 6C). Both increasing and decreasing precipitation had negative effects on above-ground biomass, with reductions of 29% and 32%, respectively compared to ambient precipitation (Fig. 6D). Any effects of temperature or precipitation, as well as their interaction effect, on leaf total nitrogen of G. spiculosa were insignificant (Table 3).

Secondary growth of C. angustifolia
One-way ANOVA results (Table 4) showed that only aboveground biomass displayed a significant difference with the incubation temperature treatment after one month's secondary growth (F = 4.096, p = 0.049); no other morphological or photosynthetic indicator showed significant differences (Table 4). Above-ground biomass of secondary growth was greatest in the +4uC manipulation, 1.11 times higher than the ambient temperature (Fig. 7).

Interaction effects of warming and precipitation alteration
Boreal wetlands are becoming increasingly vulnerable in a warming world [8]. Measuring responses of dominant plant species to warming and precipitation alteration is fundamental to assess and project the vulnerability of boreal wetlands under climate change. The interactions among multiple climate-change factors may cause responses in wetlands that may not be detected when focusing on a single factor, because climate change will include simultaneous changes in temperature and precipitation regimes [19,20]. The above-ground biomass of C. angustifolia and G. spiculosa initially showed an increasing trend and then decreased with continued increments in mean monthly temperature. The +4uC temperature treatment produced the highest biomass (Fig. 6A, C), which is consistent with the single factor study by Breeuwer et al. [10] and multi-factor meta-analysis results from Lin et al. [21].
The monthly heights in different warming manipulations show intersecting trend lines (Fig. 1A, 2A), indicating that interaction effects of temperature and precipitation should be considered, especially for the +4uC manipulation. In the decreased precipitation treatment (250%), greater plant heights were at ambient temperature and in the +4uC manipulation for C. angustifolia and G. spiculosa (Fig. 1C, 2C, respectively). With the exclusion of injury, height is one of the best indicators of an herbaceous plant's growth and competitiveness in light [22,23]. Our results show that any study on species-specific or interspecific plant response to climate change should also include consideration of the accompanying precipitation regime.
A limited increase of temperature within the optimal range often enhances photosynthesis, depending on the species and its environmental acclimation [24,25]. However, this expected enhancement of Pn can be moderated by concurrent water or drought stress on plants [26]. Our results show that Pn of C. angustifolia increased with temperature only in the driest (250% precipitation) treatment, while ambient precipitation resulted in smaller Pn values in two of the three elevated temperatures (Fig. 4A), indicating that a change in precipitation would counterintuitively limit the photosynthetic capacity of C. angustifolia. The Pn of G. spiculosa decreases with increasing precipitation (Fig. 5A), suggesting that soil water availability affects photosynthesis more than temperature variation [27]. Photosynthetic capacity is closely associated with, and approximately proportional to, leaf chlorophyll and nitrogen contents, and all three are favored by increased temperature [28,29,30]. Only the leaf nitrogen content of C. angustifolia increased with temperature (Table 3, Fig. 6B), while that of G. spiculosa was not significantly affected either by temperature or precipitation (Table 3). Although the leaf chlorophyll contents of two species were altered both by temperature and precipitation (Table 3), the temperature responses under different precipitation treatments differed between species (4D, 5D).
These differential, and even opposite, responses of two species to our manipulated temperature and precipitation treatments are consistent with the widely supported observations and modeling that show species respond individualistically to changes in environmental conditions [5].

Species-specific responses to temperature and precipitation change
The effects of climate change on boreal wetland ecosystems ranges from regional, ecosystem, vegetation type, functional type to species [21,31,32,33]. Our results suggest that within the same functional plant type, species-specific response was a significant determinant in height (Table 3, Fig. 1, 2), leaf area (Fig. 3), photosynthetic indicators, leaf chlorophyll contents (Table 3, Fig. 4, 5), and above-ground biomass ( Table 3, Fig. 6).
The most marked species-specific difference was leaf area and above-ground biomass ( Table 2, 3). Leaf area is only affected by temperature for C. angustifolia (Fig. 3A) and precipitation for G. spiculosa (Fig. 3B). Above-ground biomass of C. angustifolia is only affected by temperature (Fig. 6A), while both temperature and precipitation influence G. spiculosa (Fig. 6C, D). Given the direct relationship with net primary productivity, above-ground biomass is considered as one of the plant characteristics responding strongly to climate change [34]. The +2uC manipulation increased above-ground biomass 23% and 31% compared with ambient conditions for C. angustifolia (Fig. 6A) and G. spiculosa (Fig. 6C), respectively. According to the linear fitting of local meteorological records in Xingkai Lake lacustrine wetlands, a 2uC warming is likely to take place by the 2080s. Consequently, this species-specific difference in predicted biomass would affect carbon and nutrient budgets in boreal wetland ecosystems responding to future temperature change [31].
Previous species-specific comparisons under simulated climate change are mostly based on only one growth season, and the intergenerational response is not readily apparent [35,36]. The significant difference in above-ground biomass of secondary growth (Fig. 7) with the different temperature treatments indicates that effects of climate change would continue to the next generation, because C. angustifolia mainly reproduces clonally by rhizomes [12]. These rhizomes would store a large amount of biomass below-ground, translocated from above-ground when the shoot is harvested. Greater biomass results in better performance during secondary growth [37]. Considering the biomass accumulation of biomass year by year, C. angustifolia is more acclimated to temperature and precipitation change. This acclimated capacity supports the continued presence of C. angustifolia under climate change and perhaps even expansion, because a competitor is not as evolutionarily successful in adapting to change and hence will decline leaving an ecological niche to be filled [31].
Inter-specific interaction to temperature and precipitation change Interspecies interactions play a key role in regulating the distribution and composition of plant communities. The impacts of inter-species interactions can be altered by external drivers such as climatic conditions or anthropogenic activities [38]. For well conserved boreal wetlands, climate change may affect wetland plant communities by directly limiting or fostering the performance of particular species or by altering species competition and abundance [10,39].
As the two dominant species in Sanjiang Plain wetlands of Northeast China, C. angustifolia competes with G. spiculosa both in mesic and hydric habitats. Based on sequentially documented surveys of plant communities in a mesic habitat from 172 sampling plots from 1973 to 2003, the ''importance values'', as the average of relative density, frequency and coverage which gives an overall estimate of the influence of importance of a plant species in the community, of C. angustifolia decreased from 0.55 in the 1970s to 0.50 in the 2000s and that of G. spiculosa from 0.30 in the 1970s to 0.13 in the 2000s, accompanied by an increase in mesophytes.
Although importance values for both hygrophytes declined, the former decreased by 9%, while the latter decreased 57%. Another sequentially documented Carex lasiocarpa community in a hydric habitat showed that the importance value of G. spiculosa decreased from 0.19 in the 1970s to 0.13 in the 2000s, while that of C. angustifolia increased from 0.021 to 0.023 [40]. Consequently, C. angustifolia is more competitive than G. spiculosa and G. spiculosa is more environmentally sensitive than C. angustifolia, which might be partly attributed to the greater acclimated capacity of C. angustifolia responding to precipitation and/or temperature change (excluding anthropogenic activities). Firstly, the greatest above-ground biomass of C. angustifolia presents in the +4uC treatment (Fig. 6A), while that of G. spiculosa is in the +2uC and +4uC manipulations (Fig. 6C), suggesting that C. angustifolia might be more acclimated to greater warming, even occasional extreme high temperatures. Secondly, there is no significant difference in the precipitation manipulations for C. angustifolia's leaf area (Table 3) or above-ground biomass (Table 3), while either -50% or +50% of precipitation significantly decreased the leaf area (Fig. 3B) in August and September and above-ground biomass (Fig. 6D) of G. spiculosa.
Considering the warming trend of 2.5-5.4uC in the Sanjiang Plain within this century [41], it is predicted that the distribution of C. angustifolia will expand while that of G. spiculosa will shrink. However, the composition dynamics of these communities will be determined both by the intrinsic biological characteristics of each species and the external environment including climate change [5,42]. The species-specific responses described in this research coupled with inter-species interactions demonstrate the complexity of multi-factor effects. In addition to climate, other factors, e.g. soil nutrient concentrations, will modify or attenuate temperature and precipitation effects [35]. Consequently, minimization of other external (anthropogenic) factors is required in order to assess Table 4. One-way ANOVA of aboveground biomass, morphological and photosynthetic indicators of C. angustifolia after secondary growing for one month. climate change impacts and this can best be achieved using wellconserved boreal wetlands.

Implications for boreal wetland conservation and restoration
Boreal wetland ecosystems, especially those that are well conserved, have an innate resistance and acclimated self-regulation potential, allowing them to respond to climate change and variation in habitat conditions [7,8]. However, long-term stress caused by high temperature, extreme drought or flooding, or the degradation and eventual loss of dominant species or functional groups will alter fundamental ecosystem properties and processes, because changes in the abundance, production, or distribution of species are often sufficient to alter both structure and function of the ecosystem [43]. Therefore, investigating the thresholds of irreversible climate change effects is both urgent and critical to the conservation and restoration of boreal wetland ecosystems.
Stability and acclimation can by delineated by ecological thresholds. Within these thresholds, the species interactions of freshwater wetland plants tend to shift from competitive to facilitative with increased stress [44]; however, when the environment changes beyond these thresholds, perhaps through a combination of habitat alteration by climate change and direct damage by humans, boreal wetlands may shift to other ecosystems such as grassland or shrubland [45]. This may result in loss of the organic carbon sequestered in these wetlands for hundreds or thousands of years, ultimately reducing the net cooling effect of a carbon sink on global warming [7,45]. To avoid this consequence, conservation policies and restoration practices will need to be revised in the face of potential thresholds for irreversible change to key wetland plant species under climate change. For example, hydrologically damaged wetlands will experience greater vulnerability to climate change effects compared to wetlands with an intact hydrological regime [8]. According to our results (Fig. 6D), the stability of the natural surface water table is critical for the conservation and restoration of the G. spiculosa population, which indicates special actions may be required at both the management and policy level if this species is to be safeguarded in the future.