Water level drawdown makes boreal peatland vegetation more responsive to weather conditions

Climate warming and projected increase in summer droughts puts northern peatlands under pressure by subjecting them to a combination of gradual drying and extreme weather events. The combined effect of those on peatland functions is poorly known. Here, we studied the impact of long‐term water level drawdown (WLD) and contrasting weather conditions on leaf phenology and biomass production of ground level vegetation in boreal peatlands. Data were collected during two contrasting growing seasons from a WLD experiment including a rich and a poor fen and an ombrotrophic bog. Results showed that WLD had a strong effect on both leaf area development and biomass production, and these responses differed between peatland types. In the poor fen and the bog, WLD increased plant growth, while in the rich fen, WLD reduced the growth of ground level vegetation. Plant groups differed in their response, as WLD reduced the growth of graminoids, while shrubs and tree seedlings benefited from it. In addition, the vegetation adjusted to the lower WTs, was more responsive to short‐term climatic variations. The warmer summer resulted in a greater maximum and earlier peaking of leaf area index, and greater biomass production by vascular plants and Sphagnum mosses at WLD sites. In particular, graminoids benefitted from the warmer conditions. The change towards greater production in the WLD sites in general and during the warmer weather in particular, was related to the observed transition in plant functional type composition towards arboreal vegetation.

and decomposition, changes in either of these processes are likely to change the net impact of peatlands on the global climate.
With changing climate, greater air and soil temperatures are likely to lead to increased evapotranspiration and reduced water balances in boreal peatlands (Helbig et al., 2020), while the impacts on precipitation are predicted to be more variable and uncertain (Monier et al., 2013). As climate change is projected to have the most profound effect at high latitudes (Dai, 2013;Sand et al., 2015), future scenarios for the C balance of boreal peatlands are rather uncertain (Charman et al., 2013;Couwenberg, 2011;Dorrepaal et al., 2004).
On the input side of the C balance, vegetation composition and phenology are expected to change, leading to altered biomass production and C uptake (Antala et al., 2022). Concurrently, changed hydrology will lead to drastic shifts in peatland structure and functionality (Tahvanainen, 2011), as peatland properties, such as vegetation composition  and biomass production (Korrensalo et al., 2018;Moore et al., 2002) are primarily defined by their hydrological regimes.
Changes in both vegetation structure and leaf area development are relevant in altering ecosystem productivity (Bu et al., 2011;Koebsch et al., 2020;Mäkiranta et al., 2018;Peichl et al., 2015;Wilson et al., 2007). There are some indications that climate change could lengthen the phenological growing season of peatland vegetation, partly because of the extended period of conditions suitable for plant growth (Mäkiranta et al., 2018), and partly due to species turnover (Korrensalo et al., 2018). Plant functional types (PFTs) are known to differ in their biomass production capacity and potential growing periods, with evergreen shrubs and mosses supporting a longer growing season, while deciduous shrubs, graminoids and forbs have a shorter photosynthetically active season (Antala et al., 2022). Climate change-induced species turnover will likely change the relative proportions of PFTs and, therefore, affect biomass production through changing production capacity and growing season length (Mäkiranta et al., 2018). Two pathways of change have been suggested: (1) Increased mineralisation at lower water tables (WTs) and related aerobic soil conditions are a threat to previously stable C pools (Dorrepaal et al., 2009) but can also result in increased nutrient availability and cause a shift in peatland plant communities towards forest-associated species (Kokkonen et al., 2019). In some instances, such changes may increase photosynthetic productivity (Lohila et al., 2011;Munir et al., 2014). (2) Alternatively, lower WT may disconnect fen vegetation from the lateral flow of minerogenic water, which reduces nutrient availability and the steady source of cations to buffer organic acids, thereby leading to decreased pH levels (Laine et al., 2004). As a result, sedge-dominated fen vegetation could be replaced by bog vegetation dominated by Sphagnum mosses within decadal timescales (Tahvanainen, 2011;Wu & Roulet, 2014).
As the PFT composition varies between peatland types (fens, bogs) it is likely that also the responses to lower WT vary between peatland types.
Altered PFT composition after long-term water level drawdown (WLD) is also likely to modify the response of peatland vegetation to annually varying weather conditions. The productivity of peatland vegetation is known to be lower in cool wet summers than in warm summers (Lund et al., 2012). However, the PFTs vary in their capacity to respond to such annual variations in weather (Bubier et al., 2003;Radu & Duval, 2018), but the combined effects of changing vegetation and weather are poorly known. Thus far, most studies on peatland vegetation responses to the changing climate have been based on measurements carried out within a single growing season or over several years Antala et al., 2022;Mäkiranta et al., 2018), while few studies have focused on long-term changes (Kokkonen et al., 2019;Laine et al., 2021). Short-term responses may differ from those over the longer term (Antala et al., 2022), specifically as they do not consider long-term species turnover. Regardless, an understanding of both the long-and short-term responses of peatland vegetation to environmental change is key to building accurate climate models that account for these potential feedback mechanisms.
Currently, we lack knowledge on how sensitive boreal peatland vegetation may be to climate change-induced long-term drying and interannual growing season weather conditions. This study aims to assess the effect of long-term (>15 years) experimental WLD on boreal peatland phenology (leaf area development during the growing season) and photosynthetic biomass production, that is, parameters that control ecosystem productivity. Our experiment covers three peatland sites of different nutrient status, representing the inherent variation found in boreal peatlands. Phenological responses and photosynthetic biomass production were quantified over two growing seasons (16 and 20 years since WLD) with contrasting meteorological conditions (a wet and cold summer in 2017 vs. a warm and dry summer in 2021). The combination of the long-term and short-term environmental effects allows us to relate the observed results to gradual climatic change and to the increased likelihood of extreme weather events, such as heat waves, and droughts during the growing season.
We hypothesised that: (i) Long-term WLD increases the overall leaf area index (LAI) and the length of the growing season.
(ii) Long-term WLD promotes biomass production in vascular plants, especially dwarf shrubs and tree seedlings but reduces the productivity of Sphagnum mosses.
(iii) Long-term WLD modifies the short-term response of vegetation to different weather conditions.
(iv) The response of vegetation to WLD and weather conditions increases along with nutrient availability (rich fen → poor fen → bog).

| Experimental design
We studied the effects of long-term WLD and interannual weather conditions on the leaf area phenology of vascular plants and the seasonal photosynthetic biomass production of ground layer vascular plants and Sphagnum mosses in a boreal peatland as part of a long-term drainage experiment. The study site was located at Lakkasuo in southern Finland (61°47′ N, 24°18′ E) ( Figure 1).
Lakkasuo is an eccentric peatland complex (~1.2 km 2 ), located in the southern boreal vegetation zone with a mean annual temperature of 3.5°C and mean annual precipitation of 711 mm (ICOS Ecosystem Station Labelling Report Station: FI-Hyy (Hyytiälä), 2019). Lakkasuo is characterised by a highly diverse ecohydrology and vegetation (Andersen et al., 2011;Laine et al., 2004). Based on its ecohydrology, the peatland complex can be divided into a mesotrophic fen, an oligotrophic fen and an ombrotrophic bog, hereafter called rich fen, poor fen and bog ( Figure 1), respectively. These three are the major peatland types at this latitude (Zeng et al., 2021). The characteristic plant species for each site are given in Table (Table 2).
After more than 15 years of drainage, the vegetation composition had changed in relation to the WLD treatment (Kokkonen et al., 2019;Laine et al., 2021). In the two fen sites that were originally open peatlands, WLD promoted the formation of a tree stand.
The total tree biomass in the WLD areas exceeded that of the control areas but was clearly higher in rich fen than in poor fen ( Table 2)

| Interannual variation in growing conditions
To disentangle the potential effects of short-term changes in growing conditions from the long-term influence of peatland drying/drainage, we compared observations collected during    Table 3). As precipitation alone does not reflect the actual amount of moisture available for the vegetation (Skov & Swenning, 2004), TA B L E 2 Mean (standard deviation shown in parentheses) water table (WT) during the growing seasons (May-September) of 2017 and 2021, and total tree biomass in 2021. Total tree biomass was quantified by determining the biomass of silver birch, Scots pine and Norway spruce according to Repola (2008Repola ( , 2009
we applied a commonly used simple model to calculate the potential water balance as the difference between precipitation and monthly potential evapotranspiration, which is based on the monthly mean temperature (Rana & Tolvanen, 2021;Skov & Swenning, 2004;Stephenson, 1998). At our site, this showed a water deficit during the early growing season (May) in 2017, and during the peak growing season (June and July) in 2021 (Table 3).

| Seasonal leaf area development
The seasonal development of LAI (m 2 m −2 ) was estimated for each vascular plant species (Table S1), using the method adapted from Wilson et al. (2007). Sampling was done throughout both growing seasons from early May to late August-September at regular intervals, six times in 2017, and five times in 2021. Tree seedlings and saplings (up to 0.5 m in height) were considered part of the ground vegetation cover and were included in the LAI estimations.
In each sample plot, the number of leaves per vascular plant species were counted from five permanent representative subplots (0.08 × 0.08 m). However, large plant species with infrequent or uneven distribution were counted at the plot level. For the rich fen WLD area, which had smaller circular plots, leaf number was counted at the plot level. Simultaneously, the mean area per leaf was measured for each plant species (Table S1). Representative leaves (varying in age and size) were collected adjacent to the plots and scanned with a portable leaf area meter (LI-3000, Li-Cor). Leaf area values were multiplied by the leaf counts to determine the LAI value per species per plot. Finally, the species-wise LAI estimates were summed to provide an estimate of the total plot-level LAI estimate of all leaves present in the plot, which was used in subsequent analysis.
To assess the impact of WLD and the two contrasting growing seasons on the phenological development of LAI, we used a nonlinear mixed effects model with a log normal (ln) unimodal function with parameters that described the leaf area phenology in each plot (Equation 1) as described by Mäkiranta et al. (2018): where LAI pt is the observed total LAI for plot p at measurement time t, DOY denotes day of the year, S is the start of the growing season, C is the evergreen leaf area present at the onset of growing season and e pt denotes the residual error that were evenly distributed around a mean of zero with constant variance. The parameters to be estimated were the seasonal LAI maximum (LAIMAX), the timing of LAIMAX (DMAX) and a unitless parameter that described the length of the growing season (Shape; Figure S1). To facilitate the estimation, the parameters S and C were set constant at day 110, 19 April (C) and 0.08 m 2 m −2 (S) similarly to Mäkiranta et al. (2018). In the full model, (1) Total monthly precipitation (Prec.), average temperature (Temp.) and water balance (WB) during the growing season in 2017 and 2021, recorded at the Hyytiälä SMEAR station. Negative WB values indicate a water deficit. Long-term (1986Long-term ( -2017 mean monthly precipitation and monthly average temperature (standard deviation shown in parentheses) recorded at the Hyytiälä FMI weather station is also shown.

| Seasonal biomass production
Growth over the growing season was defined for each Sphagnum species (Table S1) using the cranked wire method (Clymo, 1970). Cranked wires were installed into intact homogenous patches of each species within each area to serve as a fixed reference point. Sphagnum growth was measured as the increment of individual Sphagnum stems along the wire from spring to autumn. In addition, shoot density and the weight (g) of 1 cm of stem were defined in Sphagnum patches adjacent to the measurement plots in July in both years. To estimate Sphagnum biomass production, the increment (cm) was multiplied by relative coverage and dry weight per cm stem to calculate the annual biomass production of each Sphagnum moss species (g per growing season) and were subsequently summed per plot.
Leaf biomass production of vascular plants was calculated similarly to Korrensalo et al. (2018) by converting LAI to biomass using specific leaf area (dry mass per leaf area, g m −2 ). The specific leaf area was defined for five replicates per species per site per WLD/control area in July of both years. Species-specific biomass production was quantified as the difference between minimal biomass at the start of the growing season and maximal biomass at peak growing season and was subsequently summed to calculate total and PFT-specific leaf biomass production. Specifically, the PFTs were graminoids (sedges and grasses), forbs, shrubs and tree seedlings (up to 50 cm height) ( Table S1).
The effects of WLD and year on biomass growth were analysed using linear mixed effects (LME) models, in which we included sample plot as a random effect to account for repeated measurements.
For biomass production we fitted three LME models with either the biomass production of the total ground level vegetation, the The composition of vegetation in terms of vascular PFTs was highly variable between sites; this was verified by testing biomass production of each vascular PFT with an LME model with site as a fixed effect ( Table S4). Because of the strong influence of the site, we continued to analyse the biomass production of each vascular PFT with site-specific LME models, with year, WLD treatment and their interaction as fixed effects. Normality and homogeneity of variances were verified with Kolmogorov-Smirnov and Levene's tests, respectively, and with diagnostic plots. Where needed, variance structures were applied (VarIdent) or the data were log normal (ln) transformed.

| Effect of WLD and weather conditions on LAI development
Vascular plant phenology at fens responded to WLD and/or to the contrasting weather conditions during the two growing seasons. The response was most pronounced in the rich fen ( Figure 3; Tables S2   and S3), where a tree stand had already developed since initiation of WLD (Table 2), and the treatment resulted in a decrease in ground vegetation LAIMAX by 60%-80% ( Figure 3). In the poor fen, the response to WLD was the reverse and less pronounced ( Figure 3; Table S3). In all three sites, LAIMAX values in the WLD areas were greater in the warmer and drier year (2021)

| Effect of WLD and weather conditions on annual Sphagnum moss and vascular plant biomass production
Photosynthetic biomass production of the ground vegetation in all three sites was found to respond to WLD and to the contrasting weather during the two growing seasons (Figure 4; Table 4). The impact of WLD treatment on vascular plants depended on the site.
Vascular plants were less productive in the rich fen WLD area than in the control area, and this effect was most pronounced in the cooler 2017 ( Figure 4; Table 4, Table S4). However, WLD did not affect vascular plant biomass production in the poor fen and bog ( Figure 4).

Overall vascular plant biomass production was greater in 2021 than
in 2017, particularly in the WLD area in the poor fen (171.78 g m −2 ) and in the rich fen control area (178.37 g m −2 ). Regardless of WLD treatment or year, the bog plots displayed the least variation in vascular plant biomass production (Table S4).
Sphagnum mosses comprised a large proportion of total biomass production in most of the sites ( Table 4). In the rich fen, WLD treatment increased Sphagnum biomass production marginally compared to the control plots dominated by vascular plants (Figure 4). In the poor fen and bog sites, Sphagnum biomass production was unaffected by WLD, but was responsive to interannual differences in weather conditions, with greater production in the warmer year (2021). In general, the greatest Sphagnum biomass production (352.42 g m −2 ) was measured in the WLD area in the poor fen, while the least Sphagnum biomass production was observed in the control area in the rich fen (85.35 g m −2 ) ( Figure 4; Table S4).
Total photosynthetic biomass production was not affected by WLD treatment (Figure 4; Table 4), but it was greater during the warmer growing season (2021) ( Table S4). This difference was most pronounced in the poor fen ( Figure 4; Table 4, Table S4). On average, the ground vegetation of the poor fen was the most productive (total biomass production in 2021 was ~503 g m −2 ), outgrowing the rich fen (~199 g m −2 in 2017, and ~272 g m −2 in 2021). The bog had the lowest biomass production in both growing seasons (~174 g m −2 in 2017, ~239 g m −2 in 2021).

| Effect of WLD and weather conditions on annual PFT biomass production
Sites differed in photosynthetic biomass production of graminoids (F 2,50 = 20.13, p < .001), dwarf shrubs (F 2,50 = 6.22, p = .004) and tree seedlings (F 2,50 = 3.28, p = .046) (data were ln transformed before analysis) (Figures 5 and 6; Tables S4 and S5). Due to their limited occurrence in the poor fen, the differences in leaf biomass production of forbs between sites could not be tested.
In the site-specific LME models, the four PFTs showed varied responses to WLD depending on the site ( Figure 5; Table S4). In the rich and poor fen where graminoids accounted for the majority of vascular plants, WLD decreased their photosynthetic biomass production (Table S5), although the decrease was significant only at the rich fen (F 1,14 = 137.34, p < .001). In contrast, WLD marginally increased graminoid biomass production at the bog site ( Figure 5; Table S4).
Forbs only made a minor contribution to total vascular plant photosynthetic biomass production (between 0.1 and 9.0 g m −2 ) and in rich fen exhibited similar responses to the graminoids: WLD decreased forb biomass production (F 1,14 = 5.54, p = .034), particularly in 2017 ( Figure 5; Table S5), while in 2021, there was slightly F I G U R E 3 Model parameter estimates with 95% confidence interval for seasonal leaf area index maximum (LAIMAX), the timing of LAIMAX (DMAX) and the length of the growing season per site (Shape), water level drawdown (WLD) treatment and year. Different letters by the estimates refer to significant differences (p < .05) between the sites and years based on pairwise contrast testing and Tukey correction for multiple testing.
greater forb biomass production (F 1,14 = 3.74; p = .074). At the bog, forb biomass production (largely composed of half-shrub Rubus chamaemorus) did not respond to WLD treatment or to year ( Figure 5; Table S5). At the poor fen, the number of forb observations was insufficient to allow for statistical testing.
At the rich fen, where WLD had resulted in a closed tree stand, the photosynthetic biomass production of shrubs decreased (F 1,14 = 33.96, p < .001), in additions, there was a marginal decrease in the production of tree seedlings (F 1,14 = 0.01, p = .928).
Concurrently, WLD decreased the relative contribution of shrubs and tree seedlings to total biomass production (Table S6). In contrast, at the poor fen (where a closed tree stand had not yet been established by the WLD) photosynthetic biomass production and the contribution of shrubs and tree seedlings to total production  Note: Models included variance structures to account for heterogeneity as indicated by (a) combined VarIdent for year and WLD treatment, or (b) log transformed (ln) transformed data. An interaction term was not included in the model if it was not a significant addition.
p < .05 shown in bold.
increased due to WLD (Figures 5 and 6; Tables S4-S6). At the bog site, shrub and tree seedling biomass production did not significantly respond to WLD ( Figure 5). In all sites and treatments (except in the bog control), shrubs and tree seedlings benefitted from the warmer and drier year (2021), which was also reflected in their greater relative contribution to total vascular plant biomass production ( Figure 6; Table S6).

| DISCUSS ION
To our knowledge, this is the first in situ study to address the impacts of long-term WLD and contrasting growing season weather conditions on leaf area phenology and biomass production. Our focus was on the parts of the ground vegetation that photosynthesise (so does not include biomass in the roots, and stems and branches of shrub and tree seedlings), and which have a key role in peatland C assimilation. The experiment included the three dominant boreal peatland types, namely a rich and a poor fen, and a bog. Across all three study sites, we observed that WLD made leaf area development and biomass production more responsive to the interannual variation in weather conditions. Yet, the response of biomass production to WLD was complex and depended on the nutrient status of the site and the functional composition of the vegetation.

| Impact of WLD on plant phenology and biomass production
Sixteen (2017) and 20 years (2021) since the start of the WLD experiment, the water level in the WLD areas was, on average, 7 cm lower than in the control areas, and the vegetation had started to adjust to new conditions (Kokkonen et al., 2022). As we expected, WLD F I G U R E 5 Seasonal biomass production per site, year and water level drawdown (WLD) treatment for four vascular plant functional types (PFTs): graminoids, forbs, shrubs and tree seedlings. Mean values and standard deviations are given in Table S4. Due to their limited occurrence in the poor fen, the biomass production response of forbs could not be tested. Different letters indicate significant differences per PFT per site based on Tukey post hoc test.
affected the plant phenology, but in contrast to our expectations the relatively small WLD had only a small effect on the photosynthetic biomass production of the ground vegetation. The responses varied between the peatland types.
The rich fen site had the greatest species turnover since the start of the WLD experiment (Kokkonen et al., 2022). The shading induced by the developing tree canopy (Kokkonen et al., 2019) caused a shift in the understory vegetation from vascular plants to mosses. Even though the decrease in the WT was small compared to typical forestry drainage (Hökkä et al., 2021), it was evident that the changes were comparable to those observed in forestry drained peatlands (Laine et al., 1995). ing tree stand (Kokkonen et al., 2022;Laine et al., 2021;Mäkiranta et al., 2018). Still, the observed vigorous growth of the tree seedlings at the poor fen site indicates the ongoing development of the tree stand towards canopy closure, similarly, but at a slower pace, to the rich fen. Therefore, further changes in the productivity may be expected also at these more nutrient poor peatland types as time passes.
We hypothesised that WLD will have contrasting impacts on the different PFTs, as the specific adaptation to varying moisture conditions and microtopography (Breeuwer et al., 2009;Koebsch et al., 2020;Riutta et al., 2007) provides a competitive advantage or disadvantage under drier conditions. In our study, the clearest impact of lower WT was the reduced growth of graminoids in the fens, which is in line with the findings of Breeuwer et al. (2009) who suggested that graminoids are rather sensitive to WT fluctuations.
In the poor fen, increased growth of shrubs and tree seedlings counterbalanced the decreased graminoid biomass, but in the rich fen the biomass had already moved to the canopy of established tree stand. Our results support the views that climate change-induced drying likely leads to shrubification, followed by tree stand formation (Nowakowska et al., 2021;Tahvanainen, 2011) and relocation of the plant biomass from the field layer to the tree canopy, as seen in the rich fen WLD area.
In contrast to our hypothesis and earlier findings on the strong effect of WLD on photosynthetic capacity, cover and intraspecific trait variation of Sphagnum mosses (Laine et al., 2021;Riutta et al., 2007), we observed a rather moderate response of Sphagnum mosses to WLD. It seems that moderate WLD, as seen in our sites, does not compromise the minimal requirements of water supply required by Sphagna (Loisel et al., 2012), also seen in a short-term WLD experiment by Mäkiranta et al. (2018).

Against our expectations, we actually observed increased
Sphagnum growth in the rich fen. In their natural state, rich fens are not noted for the extensive presence of Sphagnum species, so it is likely that WLD changed the environmental conditions to be more favourable for their establishment. It is known that the disconnection from the lateral flow of minerogenic water reduces nutrient availability and decreases pH levels (Laine et al., 2004) and therein lowers the toxic effect of calcium, and this in turn F I G U R E 6 Mean proportional photosynthetic biomass production of the vascular plant functional types (graminoids, forbs, shrubs, tree seedlings) in response to the water level drawdown (WLD) treatment during 2017 and 2021 at the three peatland sites. Proportional values for biomass production are provided in Table S6.
increases the growth of Sphagnum mosses (Clymo, 1973). Similarly, to our results, such an effect of lowered WT on the growth of Sphagnum has been reported in Sweden (Granath et al., 2010), Netherlands (Diggelen et al., 1996;Veeken & Wassen, 2020), the United States (Borkenhagen & Cooper, 2018) and Central Europe (Singh et al., 2022;Vicherová et al., 2017). In the poor fen and bog the apparent lack of moss growth response to WLD was more likely achieved by species-specific growth responses that counteracted the impact of each other. For example, lower WT is known to reduce the growth of hollow species, such as S. cuspidatum Ehrh. ex Hoffm. but increase the growth of lawn-hummock species, such as S. magellanicum Brid. (Breeuwer et al., 2009). Therefore, lower WT can induce a shift in the dominant Sphagnum species, and, in turn, change the microtopography of the peatland, (Breeuwer et al., 2009;Gong et al., 2020;Loisel et al., 2012) while the overall Sphagnum productivity remains stable.

| WLD-induced modification of short-term vegetation responses to weather conditions
Our results support our hypothesis that WLD modifies the shortterm responses of vegetation to varying weather conditions. While the effect of interannual variation in weather conditions was very subdued and insignificant in the control areas, under the WLD conditions LAI and biomass production increased during the warmer growing season in all three peatland sites. Deeper WT improves the habitat conditions for most vascular plants and increases their diversity and biomass (Zeng et al., 2021), as well as it also increases their capacity to respond to warmer temperatures (McPartland et al., 2019;Weltzing et al., 2000). In addition, long-term WLD changes the plant community composition (Kokkonen et al., 2019), and this in turn will affect the productivity of the ecosystem as PFTs vary in their photosynthetic capacity (Chen et al., 2022;He et al., 2019), but also in their ability to respond and take advantage of increasing temperatures, with graminoids showing particularly clear responses . Similarly, in our study the growth of graminoids clearly increased at all WLD sites during the warmer growing season in 2021. A similar tendency was observed for the shrubs at the poor fen and bog, while forbs and tree seedlings exhibited rather stable biomass production between sites and growing seasons.
Although the growth response of Sphagnum mosses was rather small, our results indicate that greater temperatures have the capacity to increase their growth, provided the minimal requirements of water supply are met (Bengtsson et al., 2020;Gunnarson, 2005;Loisel et al., 2012). The species composition of Sphagnum moss community changes under WLD conditions (Breeuwer et al., 2009;Loisel et al., 2012) and this change makes the Sphagnum carpet more resilient to the effect of heat waves (Bragazza, 2008;Breeuwer et al., 2009), as experienced in a milder form in our study area during summer 2021. It is probable that species turnover during the last 20 years of WLD has led to a Sphagnum community composition (Kokkonen et al., 2019) that is resilient to the heat waves and droughts currently experienced at these sites. Changes in air temperatures are invariably accompanied by a changed precipitation regime (Crhová & Holtanová, 2018;Wu & Roulet, 2014).
In our study, the vascular plants did not seem to be affected by the negative water balance that occurred at the peak of the growing season in June and July 2021. This finding is in line with Lund et al. (2012) who showed that the timing, severity and duration of drought are important regulators of photosynthesis, and that dry conditions that occur at the peak of the growing season have a smaller effect on peatland biomass production than drought at the beginning of the growing season. Based on this, the slower development and lower maximum of LAI as well as biomass production observed during 2017 may be explained by the lower temperatures and/or the spring drought in 2017. Our results support the view that in boreal and subarctic regions, warming is expected to increase vegetation biomass production by increasing the growing season length (earlier spring warming and snow melt) (Loisel et al., 2012;Menzel et al., 2006;Schulze et al., 2005), although the greater growth rates observed here were also likely additionally boosted by the warmer temperatures during the growing season (McPartland et al., 2020).
As has been shown by previous studies (Kokkonen et al., 2019(Kokkonen et al., , 2022Laine et al., 1995;McPartland et al., 2019;Weltzing et al., 2000), the response of the plant community to both long-and short-term changes in weather seems to be related to the nutrient status of the site. The fen sites are highly responsive, while only modest responses were seen in the bog.

| CON CLUS IONS
Current climate change projections indicate increased air temperatures, accompanied by changed precipitation regimes. Combining the long-term (>15-year experimental WLD) and short-term (two contrasting growing seasons) environmental changes allowed us to explore how the changes operating at different timescales affect peatland functions independently and jointly.
Our results indicate that in the long term, lower WT will lead to increased net primary production of vascular plants on boreal peatlands. This impact is further enhanced by warmer air temperatures and longer growing seasons. However, this production increase comes at the expense of changed ground layer composition and, therefore, results in changed functioning of the current ecosystem.
Our results suggest that peatlands with different relative contributions of PFTs respond to WLD at different speeds and the differences originate from the nutrient status. Rich fens with graminoiddominated vegetation respond to WLD faster and stronger, although the direction of change in all peatland types was towards an arboreal vegetation type. Similarly, the response to short-term weather variability was affected by the nutrient status of the site and enhanced by long-term gradual drying. We can conclude that the drying trend of boreal peatlands will have the strongest effect on the vegetation of rich fens. However, lower WTs induced by changing climate will make all types of boreal peatlands more responsive to weather conditions. Furthermore, our results indicate that global climate change not only increases the production of boreal peatland vegetation but also alters other ecosystem functions that are controlled by vegetation composition.