Effect of nutrient additions and site hydrology on belowground production and root nutrient contents in two wet grasslands

Effect of nutrient additions and site hydrology on belowground production and root nutrient contents in two wet grasslands

Ecological Engineering 84 (2015) 325–335 Contents lists available at ScienceDirect Ecological Engineering journal homepage: www.elsevier.com/locate/...

NAN Sizes 0 Downloads 9 Views

Ecological Engineering 84 (2015) 325–335

Contents lists available at ScienceDirect

Ecological Engineering journal homepage: www.elsevier.com/locate/ecoleng

Effect of nutrient additions and site hydrology on belowground production and root nutrient contents in two wet grasslands Keith R. Edwards ∗ ˇ Department of Ecosystem Biology, Faculty of Science, University of South Bohemia, Ceské Budˇejovice, Czech Republic

a r t i c l e

i n f o

Article history: Received 13 February 2015 Received in revised form 16 July 2015 Accepted 7 September 2015 Available online 22 September 2015 Keywords: Drought Fertilization Flooding NBPP Root CNP

a b s t r a c t Belowground production can be a significant part of total plant biomass and represents a significant pool of carbon and other nutrients. Changes in root biomass and nutrient contents can influence root decomposition and nutrient turnover rates, thereby impacting ecosystem processes and functions. Ingrowth core bags were used in a long-term nutrient addition study to determine net belowground primary production (NBPP), root nutrient (C, N and P) percentages and stoichiometry under different nutrient treatments in two wet grasslands with either mineral or organic soil. It was hypothesized that fertilization will lead to reduced NBPP, but increased root nutrient contents, and that these differences will be greater over time. The fertilizer was added in two half doses in a growing season over seven years with NBPP and the nutrient measures sampled in years 2, 3, 5 and 7. Between-year differences in the measured parameters were tested by repeated measures ANOVA, while one-way ANOVAs were run to compare between-treatment differences in each year within each site. Linear regressions were run to relate NBPP, root nutrient content and the stoichiometric ratios to changes in site hydrology. NBPP and root nutrient contents changed over time, especially in the organic soil grassland, but were largely unaffected by the nutrient additions. Site hydrology, sometimes interacting with the nutrient treatments, appeared to be the more important factor. Prolonged flooding of the two grasslands in the latter two years of measurements led to significantly reduced NBPP, and N and P% especially in the organic soil site. The response of wet grassland belowground structures to changing hydrologic conditions, in tandem with nutrient addition, may be useful in developing management plans to deal with the expected effect of climate change on water availability at the local and regional scales. © 2015 Elsevier B.V. All rights reserved.

1. Introduction The production of belowground structures can be a significant portion of total plant biomass. These belowground structures can be significant pools of carbon and other nutrients, and the subsequent death and decomposition of these structures can represent large inputs of these nutrients to ecological systems (Persson, 1979; Vogt et al., 1998; Boyd and Svejcar, 2012). Root production and the rate of root turnover can greatly affect various ecosystem processes, including nutrient flows and fluxes such as carbon sequestration (Gill and Jackson, 2000; Bai et al., 2010). Recently, there has been increased interest in belowground production. Various factors, such as increased CO2 (Milchunas et al., 2005; Arndal et al., 2013; Sonderegger et al., 2013), greater or lesser moisture levels (Zhang and Zak, 1998; Fiala et al., 2009; Bai et al., 2010) and nutrient

∗ Correspondence to: Faculty of Science, University of South Bohemia, Braniˇsovská ˇ Budˇejovice, Czech Republic. 1760, 37005 Ceské E-mail address: [email protected] http://dx.doi.org/10.1016/j.ecoleng.2015.09.034 0925-8574/© 2015 Elsevier B.V. All rights reserved.

additions (Dukes et al., 2005; Moar and Wilson, 2006; Blue et al., 2011), have been found to greatly influence root production and nutrient turnover rates. Increased nutrient availability is known to result in changes to plant biomass allocation patterns, favoring aboveground production at the expense of belowground structures (Saggar et al., 1997; Detenbeck et al., 1999). The roots produced are of higher quality (lower C:N ratio) with faster turnover rates resulting in more rapid nutrient cycling within a particular system (Olde Venterink et al., 2002; Kaˇstovská et al., unpublished data). A majority of studies on belowground production have been conducted in forested and upland grassland habitats (see appendix in Gill and Jackson, 2000 and later studies such as Milchunas et al., 2005; Meinen and Leuschner, 2009; Dodd and Mackay, 2011; Ladwig et al., 2012, etc). Such studies in wetlands have occurred mostly in riparian meadows (Martin and Chambers, 2002; Kiley and Schneider, 2005; Blank et al., 2006; Boyd and Svejcar, 2012), with non-riparian, non-forested wetland systems, such as wet grasslands, being under-represented. Overall, most nutrient addition studies have generally been of short duration, usually a year or less, with only a few studies investigating the effect of long-term

326

K.R. Edwards / Ecological Engineering 84 (2015) 325–335

(>1 year) nutrient additions on primary production or plant nutrient contents (e.g., Ladwig et al., 2012). Wet grasslands are highly productive, graminoid-dominated wetlands, which can have either low or high vegetation species diversity, depending on the heterogeneity of the plant community mosaic (Ruˇzíˇcká, 1994; Prach and Straˇskrabová, 1996; Joyce and Wade, 1998). In Europe, wet grasslands are maintained by human activities, therefore any change in management practices can greatly impact the state of these systems (Joyce and Wade, 1998; Tallowin and Jefferson, 1999). The area of European wet grasslands has declined over the last 60 years due to agricultural intensification, conversion to upland arable land, or abandonment (Joyce and Wade, 1998; Tallowin and Jefferson, 1999; Prach, 2008). Only a few studies in wet grasslands have focused on belowground production (Fidelis et al., 2013). These studies found that belowground production was negatively affected by late season mowing (Dumortier et al., 1996), but enhanced under greater fire frequency (Fidelis et al., 2013), while flooding could have either a positive (Barbosa et al., 2012) or negative (Fidelis et al., 2013) impact on root growth. No studies have investigated the effect of nutrient additions on belowground production in wet grasslands. A manipulative field experiment was established to investigate the effect of prolonged nutrient additions (over seven years) on belowground production and the nutrient percentages and stoichiometric ratios of the belowground structures in two wet grasslands growing on either organic or mineral soil. This study was part of a larger project investigating how long-term nutrient inputs and soil type affected plant-soil interactions in wet grasslands. Within this larger project, short-term nutrient treatment effects on primary production (both above and belowground) have already been published (Picek et al., 2008; Edwards et al., 2015). In keeping with the results of past nutrient addition studies, it was hypothesized that plants in fertilized plots will have reduced belowground production, but that these structures will have greater nutrient (N, P) contents than those growing in control, unfertilized plots. In addition, it was expected that these differences would become greater over time due to the repeated application of nutrients (hypothesis two). In addition to the manipulative nutrient addition experiment, there were uncontrolled changes in the hydrologic conditions of the two wet grasslands over the seven years of the study. The first few years of data collection occurred at times of drier site conditions while the wet grasslands were flooded for long periods in the later years of the study. Given that future conditions in Central Europe are predicted to be drier, with changed precipitation patterns, as a result of climate change (IPCC, 2014), and that belowground plant C can be an important input affecting other ecosystem processes and functions, it may be interesting to determine how other factors, such as site hydrology, affect belowground plant production, either singly or when interacting with the experimental nutrient additions. Therefore, this paper describes the response of belowground plant production to a long-term, manipulative nutrient addition experiment overlaid by an observational study of water level effects. Such information could be useful in determining how nutrient additions and site hydrology, either singly or in combination, affect plant nutrient inputs to wet grasslands and what management actions may be required to maintain well-functioning ecosystems.

2. Methods 2.1. Study sites In situ experiments were conducted in two wet grasslands, one ˇ Basin on organic soil and the other having mineral soil, in the Tˇrebon

Biosphere Reserve (TBBR, 49◦ 01 N and 14◦ 46 E), Czech Republic. Complete descriptions of the study wetlands are given in Picek et al. (2008). Briefly, the organic site, Záblatské Louky (hereafter referred to as the organic soil site), is located in the inundation area of a large fishpond. Carbon accumulation is typical of poorly flushed marginal wetlands in the TBBR (Prach, 2002). The site is a species poor (11 vascular plant species; H = 0.48) sedge meadow dominated by Carex acuta (50–75% cover) and Carex vesicaria (10–25%). Other species found in the site include Phalaris arundinacea (5%), Lythrum salicaria (5%) and Galium palustre (5–10%) (based on results of phytosociological relevés, L. Rektoris, personal communication). The moss Calliergonella cuspidate is also quite common (25–50% cover). The organic soil site is mown once a year in mid-July. The second site, Hamr (hereafter referred to as the mineral soil site), is located in the floodplain of a small river (Neˇzárka) and is frequently flushed. It is a normal situation in floodplains that wetlands on mineral soil tend to predominate on frequently flushed sites (Hejny´ and Segal, 1998). The site has a silt-sand alluvial substrate and is dominated by Glyceria maxima and C. acuta (both with 25–50% cover) with only four other species present (L. Rektoris, personal communication). In this site, C. cuspidate is quite rare with less than 1% cover. Water levels are similar to those in local drainage ditches, which are connected with the river. Thus, average water level is lower and more variable in this site than in the organic soil site. The mineral soil site is mown two times during the growing season, in early-June and August. The sites are about 11 km distant from each other and have similar elevations (ca. 400 masl). 2.2. Sampling design A field experiment was established at the two sites in spring 2006. See Picek et al. (2008) for a full description of the experimental design. Briefly, four blocks were established in each site, with three treatment plots (12.25 m2 ) per block. Of these, only two of the nutrient treatments were used in this study, with fertilization treatments of 0 and 300 kg NPK ha−1 yr−1 (corresponding to unfertilized control and high fertilization treatments, respectively). This differs from earlier studies (Picek et al., 2008; Zemanová et al., 2008; Edwards et al., 2015) which included data from a third, middle nutrient treatment (65 kg NPK ha−1 yr−1 , corresponding to the Low treatment in those studies). Nutrient addition was administered to randomly selected plots using a commercial NPK fertilizer (Lovofert 15:15:15 NPK, Lovochemie, a.s.), which is recommended for use in wet grasslands. The fertilizer was added in two half doses during the growing season (mid-May and July) to simulate normal agricultural practices. Adjacent plots were separated by at least 1.5–2 m wide buffer strips. Nutrient addition occurred when there was no standing water in the sites and when fertilizer application was followed by at least two days with no precipitation. Laboratory analyses of the soil from the sites showed that the added nutrients were immobilized into the plants and soil microbes within 48 h. This rapidity of nutrient uptake appears to be a common feature of wet grasslands (see Rejmánková et al., 2008). Therefore, it was felt that these precautions greatly reduced the chance of the nutrients spreading to adjacent plots even though the sites were occasionally flooded. Continuous water level measurements were taken from a well established in the center of each site. Water levels were taken at 15 s intervals using the STELA system, which consists of a field datalogger (M4516) with a GSM/GPRS modem (MG40) and water level sensor (Fiedler-Magr, Electronics for Ecology, Czech Republic). For this study, daily water level means were calculated from the continuous water level readings using MOST (Monitoring Station) version 2.3 (Fiedler-Magr, Electronics for Ecology, Czech Republic). Net belowground primary production (NBPP) was measured in both sites in 2007, 2008, 2010 and 2012 (representing years 2,

K.R. Edwards / Ecological Engineering 84 (2015) 325–335

3 5 and 7 from the start of experimental fertilization) using the in-growth core bag method (Vogt et al., 1998). This is a common method for determining belowground production in both agricultural (e.g., Steingrobe et al., 2000; Steingrobe, 2005) and ecological studies (Finer and Laine, 2000; McKee and Faulkner, 2000; Edwards and Mills, 2005). Replicate mesh bags (at least three bags per plot), with dimensions of 7 cm diameter and 15 cm depth, were filled with soil from the respective sites, buried in the root zone of each plot in the two sites at the beginning of each growing season (late April–mid May), so that the top of each bag was level with the top of the soil, and retrieved near the end of the particular growing season, usually from mid-October to mid-November. Ingrowth bags for a particular site were installed or removed on the same day with these operations occurring less than a week apart between the two respective sites. Possible differences in root growth patterns between the core and the surrounding soil were minimized by using the soil of the respective site in addition to limiting soil manipulation (sieving) prior to placing the soil into the mesh bags, thereby maintaining the natural soil properties (Steingrobe et al., 2000). Keeping the core bags in the soils over most of the growing season allowed for adequate root growth into the bags without significant root mortality. Once removed, the bags with soil and roots were taken to the laboratory where they were carefully cleaned and the roots washed, dried at 65◦ C for 72 h and weighed. The resulting dry weights are equal to NBPP. These data were then extrapolated to a m2 basis (gram dry weight per square meter per year; g DW m−2 yr−1 ) for statistical analyses (see below). C, N, and P percentages were determined from the dried belowground samples in each year with the exception of P in 2008. C and N were analyzed using a CN analyzer (ThermoQuest, Italy) while P was determined using a flow injection analyzer (FIA, Lachat QC8500, Lachat Instruments, USA). Stoichiometric ratios (C:N, C:P, N:P) were determined from these data (Sterner and Elser, 2002). 2.3. Data analysis The experiment was set-up using a block design to deal with possible site heterogeneity. Thus, complete randomized block (CRB) ANOVAs were first run on the NBPP and nutrient data (four replicates per treatment per site), using SYSTAT v 11, to determine whether there were any significant block effects. Since no block effects were found, further analyses were conducted. The effect of nutrient additions over a seven year period on NBPP and the root nutrient data in each site over time was determined by repeated measures ANOVA using Statistica v. 10 with time, the nutrient treatments and their interaction as the test factors. Scheffé’s comparison of means test was used in the case of significant differences between years (Sokal and Rohlf, 1995). The effect of the two nutrient treatments (unfertilized control; high fertilization), within any year in which NBPP, the nutrient percentages and stoichiometric ratios were measured, was analyzed by one-way ANOVA using SYSTAT v. 11. When necessary, the data were transformed by natural log or square root transformations to attain normality and homogeneous variances. For each site, several water level parameters were determined from the daily mean water level measurements including the mean, minimum and maximum water levels during the particular period of the growing season in which the ingrowth bags were installed in any particular year as well as the range of water level fluctuation within any particular period (=maximum − minimum). Two other relative water level parameters were calculated from the basic water level data, the first being the number of days in a particular growing season period in which the water level was below the root zone (detected visually as the top 20 cm of the soil layer). Because the number of days in which ingrowth bags were installed in the sites differed between the years, the proportion of days in

327

Fig. 1. Annual net belowground primary production (NBPP = gDW m−2 yr−1 ; means ± 1 SE) in an organic and mineral wet grassland subjected to different nutrient addition treatments using a commercial NPK fertilizer, Nutrient treatments: unfert: unfertilized control; high fert: high fertilization (300 kg NPK ha−1 yr−1 ).

any particular growing season period in which the water level was below the root zone was also calculated. It was thought that such relative parameters may be more ecologically relevant than the basic water level parameters. Analyses were conducted separately for each nutrient treatment in each site resulting in four groups of analyses with each analysis containing four data pairs (NBPP or nutrient percentages vs specific water level parameter). Trends, whether linear, quadratic or exponential, were determined visually. In addition, polynomial contrasts were determined for the relation of NBPP and the nutrient percentages to the water level parameters using SYSTAT. Both the visual and polynomial contrast analyses produced similar results. The strength of the relationship between NBPP and the root nutrient data to the selected water level parameters was determined by regression analyses using SYSTAT v. 11 with NBPP or the nutrient data as the dependent variable and the selected water level parameter as the independent variable. Between-site statistical comparisons were not conducted in this study for two reasons. First, the two sites not only have different soils, but they also differ in their trophic status, with the organic soil site being oligotrophic while the mineral soil site may be classified as more minerotrophic (Edwards et al., 2015). In addition, the sites are also managed differently, with the organic site being mown only once in any growing season while the mineral soil site is usually cut twice in the same period. It was felt that any, if not all, of these factors (soil type, trophic status, cutting frequency) could affect belowground production and would be difficult, if not impossible, to distinguish statistically between individual factor effects. Therefore, any comparisons will be only descriptive and not statistical. 3. Results 3.1. Nutrient treatment effects The repeated measures analyses showed that NBPP decreased significantly over time in both sites (Fig. 1; Table 1), with the decrease being greater in the organic soil site (p < 0.001) than in the mineral soil site (p = 0.027). In the organic soil site, maximum NBPP occurred in 2008 then decreased to minimum levels in 2012 for both nutrient treatments. The pattern was somewhat similar for the mineral soil site in that NBPP was significantly greater in the first two years of measurements (2007, 2008) compared to the last two years (2010, 2012). However, maximum NBPP in the mineral soil site occurred in 2007 for both nutrient treatments, a year earlier

328

K.R. Edwards / Ecological Engineering 84 (2015) 325–335

Table 1 Means (±1 SE) of net belowground primary production (NBPP) in the organic and mineral wet grasslands measured in 2007, 2008, 2010 and 2012 as well as the results of repeated measures ANOVA (Year; F statistic) in both sites. Letters in each year column for the two sites represent results of one-way ANOVAs comparing between-nutrient treatments within each site in each year; different letters represent significant differences within each site. Nutrient treatments: unfert, unfertilized control; high fert, high fertilization (300 kg NPK ha−1 yr−1 ).

Organic Unfert High fert Mineral Unfert High fert * **

2007

2008

2010

2012

Year

417.68 ± 33.56a 290.00 ± 4.05b

1579.90 ± 384.77a 1023.05 ± 78.95a

118.61 ± 28.66a 151.25 ± 23.10a

103.87 ± 16.81a 146.02 ± 9.09ab

88.49**

287.83 ± 19.18b 261.80 ± 87.39b

307.20 ± 79.06b 153.96 ± 25.14b

159.48 ± 53.94a 235.80 ± 16.20a

222.02 ± 46.11a 156.48 ± 21.97a

3.85*

P-value < 0.05. P-value < 0.001.

than in the organic soil site, while minimum values were attained in different years depending on the nutrient treatment, being 2010 (year 5) for the unfertilized control but 2012 (year 7) for the high fertilization treatment. In both sites, the changes in NBPP over the full seven years of nutrient additions were not influenced by the nutrient treatments (non-significant treatment effect in the repeated measures ANOVAs). However, there were significant between-nutrient treatment effects in the organic soil site for 2007, when NBPP in the unfertilized control treatment was significantly greater than that in the high fertilization treatment (one-way ANOVA, p = 0.005). A weaker nutrient effect was seen in 2012 (p = 0.067), but this time the high fertilization treatment had in greater belowground production compared to the unfertilized control (Table 1). No significant within-year nutrient treatment effects on NBPP were found in the mineral soil site. As with NBPP, root C, N and P percentages changed significantly over the years in both sites (repeated measures ANOVA, p < 0.001), with the exception of P% in the mineral soil site (p = 0.073). Root C% increased from a low of about 35% in 2008 in both wet grasslands to maxima of 42 and 40% in the organic and mineral soil sites respectively in 2012 (Fig. 2a). Conversely, root N and P percentages decreased over time. In the organic soil site, both N and P% were highest in 2007 (1.14 and 0.18, respectively), but then decreased significantly by 2008 for N% (0.8) and 2010 for P% (0.12); both nutrients then were stable for the remainder of the field experiment. Likewise, N% significantly decreased over the seven years of the study in the mineral soil site, but this decrease only started after 2008 (Fig. 2b and c). A significant nutrient treatment effect over time was found only in the organic soil site, with roots from the high fertilization treatment having higher C% than those growing in the unfertilized control treatment (repeated measures ANOVA; p = 0.016; Table 2). Plants growing in the high fertilization treatment in the mineral soil site had greater root N% than those growing in the unfertilized control treatment in the same site in all years, but this resulted in only a weak overall nutrient treatment effect (repeated measures ANOVA; p = 0.085). There were also only a few examples of the nutrient treatments significantly affecting root nutrient percentages within any year. For the organic soil site, root P% was significantly greater in the high fertilization treatment (p = 0.002) and only in 2012; no other significant nutrient treatment effects were found for this site (Fig. 2). Likewise, in the mineral soil site, root N% was greater in the high fertilization treatment in 2007 (one-way ANOVA, p = 0.037) while root C% was greater in the unfertilized control treatment in 2008 (p = 0.03). Changes in the stoichiometric ratios over the years mirrored the changes in the root C, N and P percentages, with the C:N (p < 0.001 both sites) and N:P ratios (p = 0.05 and p < 0.001 for the organic and mineral soil sites respectively) significantly differing over time (Fig. 2; Table 2). There were significant between-year differences

Table 2 F statistics from repeated measure ANOVAs for the percentages of root carbon (C%), nitrogen (N%), phosphorus (P%) and the C:N, C:P and N:P stoichiometric ratios in the organic and mineral wet grasslands measured in 2007, 2008, 2010 and 2012 (Year). Treat, nutrient treatments; unfert, unfertilized control; high fert, high fertilization (300 kg NPK ha−1 yr−1 ).

C% Organic Mineral N% Organic Mineral P% Organic Mineral C:N Organic Mineral C:P Organic Mineral N:P Organic Mineral * ** *** +

Year

Treat

Year × treat

17.2*** 40.2***

11.1* 0.01

0.6 6.4**

44.03*** 17.39***

1.45 4.24+

1.02 0.11

21.79*** 3.27+

3.48 1.57

1.62 0.69

38.34*** 30.98***

2.22 2.24

1.82 0.32

28.22*** 0.43

4.21+ 0.32

3.94* 0.75

3.88* 26.48***

19.85** 0.69

2.57 0.39

P-value < 0.05. P-value < 0.01. P-value < 0.001. P-value < 0.10.

in the C:P ratio, but only in the organic soil site (p < 0.001) while it remained unchanged in the mineral soil site. The pattern of these changes differed between the two wet grasslands. Both the C:N and C:P ratios in the organic soil site significantly increased each year from 2007 to 2010 before stabilizing. This differed in the mineral soil site in which the C:N ratio only began to increase after 2008, from a low value of 27 in 2007 to a maximum of 51 in 2012, while the C:P ratio was unchanged in this site (Fig. 2d and e). The N:P ratio in the mineral soil site significantly decreased every year from 2007 to 2012 in both nutrient treatments due to the decrease over time in N% coupled with the lack of change in P%. In the organic soil site, the N:P ratio also decreased significantly from 2007 to 2010, but then it increased again in 2012, leading to the observed weakly significant change over time (Fig. 2f). As with the nutrient percentages, the nutrient treatments significantly affected the stoichiometric ratios in only a few cases (Table 2). The only significant overall nutrient treatment effect was a higher N:P ratio in the unfertilized control treatment compared to the high fertilization treatment in the organic soil site (repeated measures ANOVA; p = 0.004). There were also only a few cases of significant nutrient effects within any year. The nutrient treatments weakly affected the C:N ratio in both sites in 2007 (one-way ANOVA; p = 0.069 and 0.054 for the organic and mineral soil sites, respectively), with higher ratios in the high fertilization treatment in the organic soil site, but greater ratios in the unfertilized control

K.R. Edwards / Ecological Engineering 84 (2015) 325–335

329

Fig. 2. Root nutrient percentages (A–C) and their stoichiometric ratios (D–F) in wet grasslands with either organic (org) or mineral (min) soil measured in 2007, 2008, 2010 and 2012, representing 2, 3, 5 and 7 years since the start of nutrient addition treatments, respectively. Nutrient treatments: unfert, unfertilized control; high fert, high fertilization (300 kg NPK ha−1 yr−1 ). Rows on top and bottom represent the results of repeated measures ANOVAs in each parameter; different letters in each row represent significant between-year differences.

treatment in the mineral soil site. Weak relationships were also found for the C:P and N:P ratios (p = 0.058 and 0.051, respectively) in the organic soil site in 2012, with both ratios being greater in the unfertilized control treatment. The N:P ratio in the mineral soil site was also significantly greater in the unfertilized control treatment (p = 0.016), but in 2007. 3.2. Relation of water levels to NBPP and root nutrient percentages The changes in NBPP and the root nutrient percentages and stoichiometric ratios appeared to coincide with changes in site water levels. NBPP, N% and P% were for the most part higher in 2007 and 2008 when conditions in the two wet grasslands were dry (Fig. 3a and b) with mean water levels during the period of ingrowth bag exposure being below the root zone, which was visually estimated to be in the top 20 cm of the soil profile. The subsequent significant decrease in NBPP, N% and P%, and the significant increase in C%, occurred when the sites had standing water for most of the last two growing seasons (2010 and 2012; Fig. 3c and d). Therefore, the plants growing in the two wet grasslands were subjected to quite dry, even drought, conditions in the first two years of the study, but then had to contend with periods of prolonged flooding in 2010 and 2012. It appeared that the site as well as the nutrient treatments interacted to influence how NBPP and the root nutrient percentages were affected by changes in site hydrologic conditions. The two nutrient treatments did not appear to affect the relationships between NBPP and the water level parameters in the organic soil

site, which were best described by either a quadratic (minimum, range, and the days and proportion of days when the water level was below the root zone) or exponential (mean, maximum) function for both nutrient treatments (Fig. 4a). However, there appeared to be a nutrient treatment effect in the mineral soil site, with NBPP changing in a linear fashion relative to changes in all of the water level parameters for the unfertilized control treatment while half of the relationships were best described by a quadratic function in the high fertilization treatment (Table 3; Fig. 4b). The sites also differed in terms of which parameters significantly affected NBPP and whether there was an apparent hydrology × nutrient treatment interaction. In the organic soil site, mean and maximum water levels negatively affected NBPP while the proportion of days when the water level was below the root zone was positively related to NBPP in both nutrient treatments. Nutrient treatment appeared to be a greater influence on these relationships in the mineral soil site. All of the water level parameters, except range, significantly affected NBPP in the unfertilized control treatment in this site, while the opposite was the case (range was the only significant parameter) in the high fertilization treatment (Table 3). It may be that nutrient addition overrides any hydrologic effect in this minerotrophic wetland. Both site differences and the nutrient treatments again appeared to affect how root nutrient percentages and stoichiometric ratios responded to changing hydrologic conditions. The N:P ratio was the lone exception with no water level parameter significantly affecting this ratio in either wet grassland. Root nutrients were more influenced by hydrologic changes in plants subjected to the high fertilization treatment than those which did not receive additional

330

K.R. Edwards / Ecological Engineering 84 (2015) 325–335

Fig. 3. Water levels in the two wet grasslands (organic, mineral) for (A) 2007; (B) 2008; (C) 2010 and (D) 2012. The solid line at zero represents the soil surface while the dashed line at 20 cm below the soil surface represents the lower boundary of the root zone.

Fig. 4. The relation of mean net belowground primary production (NBPP = gDW m−2 yr−1 ) in wet grasslands with either organic (A–C) or mineral (D–F) soil to selected site water level parameters (mean water level, A, B; water level range, C, D; days in a growing period that the water level was below the root zone, E, F).

K.R. Edwards / Ecological Engineering 84 (2015) 325–335

331

Table 3 Regression equations relating net belowground primary production (NBPP) in two nutrient treatments to selected water level parameters in the (A) organic and (B) mineral wet grasslands. Water level parameters: mean, mean water level over the growing season; min, minimum water level; max, maximum water level; range = the amount of water level fluctuation over the growing season (=maximum − minimum); days, the number of days in a growing season that the water level was below the root zone; prop, the proportion of days in a growing season that the water level was below the root zone. Nutrient treatments: unfert, unfertilized control; high fert, high fertilization (300 kg NPK ha−1 yr−1 ). Significant relationships are in bold. Unfert

High fert 2

r2

Equation

F

r

Equation

(A) Organic Mean Min Max Range Days Prop

NBPP = 7.21 × exp−0.74 × Mean NBPP = −406.01 − 61.79 × Min − 0.58 × Min2 NBPP = 9.09 × exp−1.40 × Max NBPP = −4557.38 + 191.87 × Range − 1.49 × Range2 NBPP = 58.15 + 108.41 × Days − 0.99 × Days2 NBPP = 80.48 + 10455.19 × Prop − 13669.39 × Prop2

1226.54*** 0.89 68.60* 87.88+ 96.94+ 245.99*

0.98 0.00 0.60 0.98 0.98 0.99

NBPP = 6.69 × exp−0.51 × Mean NBPP = −168.78 − 37.98 × Min − 0.35 × Min2 NBPP = 7.85 × exp−0.91 × Max NBPP = −2698.95 + 117.30 × Range − 0.92 × Range2 NBPP = 115.64 + 67.35 × Days − 0.62 × Days2 NBPP = 129.70 + 6432.92 × Prop − 8498.95 × Prop2

601.82*** 0.86 103.00** 94.68+ 100.58+ 277.07*

0.91 0.00 0.50 0.98 0.98 0.99

(B) Mineral Mean Min Max Range Days Prop

NBPP = 191.68 − 2.20 × Mean NBPP = 150.06 − 1.99 × Min NBPP = 273.28 − 2.38 × Max NBPP = 67.93 + 2.85 × Range NBPP = 170.23 + 0.92 × Days NBPP = 170.61 + 137.08 × Prop

107.30** 56.83* 121.33** 0.52 128.52** 102.10

0.97 0.95 0.98 0.00 0.98 0.97

NBPP = 205.89 − 0.84 × Mean NBPP = 273.74 + 9.14 × Min + 0.12 × Min2 NBPP = 325.29 + 5.39 × Max − 0.20 × Max2 NBPP = 2772.66 − 91.36 × Range + 0.79 + Range2 NBPP = 203.82 + 0.26 × Days NBPP = 200.25 + 47.10 × Prop

0.21 0.58 0.58 483.77* 0.11 0.17

0.00 0.00 0.00 1.00 0.00 0.00

* ** *** +

F

P-value < 0.05. P-value < 0.01. P-value < 0.001. P-value < 0.10.

nutrient inputs (Table 4). This nutrient treatment effect differed between the sites depending upon the particular nutrient or stoichiometric ratio. For example, different water level parameters were significantly related to C and N% between the two nutrient treatments in the organic soil site, while this nutrient treatment effect was greater in the mineral soil site when describing the relationship between P% and the C:N ratio and the various water level parameters. The relationships between the C:P ratio and the water level parameters were similar in both wet grasslands with significant responses being found for the high fertilization treatment. Overall, the proportion of days in a growing season in which the water level was below the root zone was a good descriptor of the changes in root nutrient percentages and stoichiometry in almost

all cases, while mean water level was often a significant parameter in the high fertilization treatment and only rarely in the unfertilized control treatment (Table 4). The two wet grassland sites also differed in how closely correlated were the various water level parameters. These parameters were all highly correlated to each other (Pearson correlation, r > 0.80) in the organic soil site. However, in the mineral soil site, range was only weakly related to the other parameters, which were all highly correlated to each other (Table 5). This weak correlation of water level range to the other parameters is likely indicative of the greater flushing that occurs in this floodplain site, with its greater connection to the river levels, compared to the more stable hydrologic conditions of the organic soil site.

Table 4 Regression equations relating root nutrient percentages (C%, N%, P%) and stoichiometric ratios (C:N, C:P, N:P) in two nutrient treatments to selected water level parameters in the organic and mineral wet grasslands. Only significant relationships are shown with F statistics in parentheses. Water level parameters: mean, mean water level over the growing season; min, minimum water level; max, maximum water level; range = the amount of water level fluctuation over the growing season (=maximum − minimum); days, the number of days in a growing season that the water level was below the root zone; prop, the proportion of days in a growing season that the water level was below the root zone. Nutrient treatments: unfert, unfertilized control; high fert, high fertilization (300 kg NPK ha−1 yr−1 ). Organic unfert C%

N%

0.76 + 0.48 × lnprop (100.00** )

P%

0.09 + 0.001 × range (174.92* ) 0.12 + 0.001 × days (226.24* ) 0.12 + 0.08 × prop (233.74* ) 56.66 + 0.25 × min (20.75* ) 64.76 − 0.33 × range (116.19** ) 54.14 − 0.17 × days (23.55* ) 54.19 − 25.48 × prop (41.56* )

C:N

C:P

N:P * ** ***

P-value < 0.05. P-value < 0.01. P-value < 0.005.

Organic high fert

Mineral unfert

Mineral high fert

41.75 + 0.12 × mean (18.70* ) 34.78 + 2.37 × ln max (1503.30*** ) 57.26 − 3.23 × ln range (1715.84*** ) 42.43 − 0.76 × ln days (3272.27*** ) 42.12 − 6.27 × ln prop (2303.34*** )

36.94 + 0.06 × max (18.13* ) 39.52 − 0.02 × days (19.85* ) 39.50 − 3.42 × prop (17.82* )

1.28 − 0.18 × ln max (72.35* ) 0.74 + 0.43 × ln prop (68.81* ) 0.14 − 0.001 × mean (31717.44 *** ) 0.12 + 0.001 × range (156.97* )

0.92 + 0.48 × ln prop (118.35** )

33.52 + 1.63 × ln max (12285.18*** ) 41.11 − 1.05 × ln days (528.50*** ) 40.34 − 6.98 × ln prop (572.71*** ) 1.01 + 0.07 × ln days (75.95* ) 1.02 + 0.55 × ln prop (175.53** ) 0.29 + 0.001 × mean (154.59* ) 0.29 − 0.001 × days (463.86* ) 0.29 − 0.05 × prop (733.32* ) 25.49 + 3.55 × ln mean (113.32** ) 40.30 − 20.76 × ln prop (104.85** ) 139.24 + 17.34 × mean (305.25* ) 139.16 + 0.12 × days (224.38* ) 139.24 + 17.34 × prop (305.25* )

ln max (170.43** )

360.31 − 1.29 × range (3448.98* ) 318.71 − 0.73 × days (15842.70*** ) 318.66 − 106.32 × prop (21165.34*** ) 6.94 − 0.06 × max (177.02* )

332

K.R. Edwards / Ecological Engineering 84 (2015) 325–335

Table 5 Pearson correlation matrices of the relationships between the water level parameters in the (A) organic and (B) mineral wet grasslands. Parameters: mean, mean water level over the growing season; min, minimum water level; max, maximum water level; range = the amount of water level fluctuation over the growing season (=maximum − minimum); days, the number of days in a growing season that the water level was below the root zone; prop, the proportion of days in a growing season that the water level was below the root zone. Mean

Min

Max

Range

Days

Prop

(A) Mean Min Max Range Days Prop

1 0.944 0.857 −0.926 −0.986 −0.972

1 0.890 −0.985 −0.971 −0.981

1 −0.799 −0.825 −0.819

1 0.974 0.988

1 0.997

1

(B) Mean Min Max Range Days Prop

1 0.984 0.982 −0.484 −0.995 −0.998

1 0.953 −0.609 −0.967 −0.973

1 −0.340 −0.995 −0.989

1 0.396 0.429

1 0.999

1

4. Discussion Nutrient inputs and water levels are two main factors that can influence the biomass, production and nutrient contents of belowground plant structures (Lieffers and Shay, 1981; Hefting et al., 2004; Dukes et al., 2005). Changes to these parameters can impact root decomposition rates and therefore affect nutrient turnover, which can alter ecosystem processes. Many nutrient input studies have shown strong nutrient effects on NBPP (such as Tilman, 1987; Saggar et al., 1997; Detenbeck et al., 1999) with higher nutrient contents in plant structures (for wetland plants see Nadelhoffer et al., 2002; Güsewell et al., 2003; Ságová-Mareˇcková et al., 2009). In this study, nutrient addition significantly affected NBPP, but only in the organic soil site and only in 2007 and weakly in 2012. Likewise, there was a significant nutrient effect on the N:P ratio in both wet grasslands, but C% was only affected in the organic soil site. The greater effect of the nutrient treatments on plants growing in the organic soil site is in agreement with past results from these two wet grasslands (Picek et al., 2008; Zemanová et al., 2008; Edwards et al., 2015). Still, the initial expectation of reduced belowground production coupled with greater nutrient contents in plants subjected to the high fertilization treatment can only partially be supported. Likewise, the second hypothesis of increased between-nutrient treatment differences over time in NBPP and root nutrient contents was not supported by the results. As in this study, others have found a similar lack of a nutrient effect on NBPP, especially in upland and semi-arid grasslands (Dodd and Mackay, 2011; Ladwig et al., 2012). Chen et al. (2014) also found decreased NBPP with greater nutrient availability in a global survey of forest ecosystems, as did Ket et al. (2011) in tidal marshes, while Kearney and Zhu (2012) found no effect of nitrate addition on root growth or N and P contents. 4.1. Hydrologic effects Although the decline in NBPP with continual nutrient additions over time is what would be expected, it can also be explained by the changing site hydrology. Also, the changes in root nutrient percentages and stoichiometric ratios cannot be so readily explained by any nutrient effect, but are more closely tied to the hydrologic changes seen in the sites. In this case, site hydrology can overshadow nutrient effects on production and nutrient contents under particular conditions (Güsewell et al., 2003). The weather pattern in 2007

resulted in normal site hydrologic dynamics for the conditions of the Czech Republic, with a wet spring followed by lower water levels in summer due to drier conditions and then increased water ˇ and Jeník, 2002). levels in fall following the autumnal rains (Pˇribán However, the plants in the wet grasslands experienced stress in the other years due to the drought-like conditions in 2008, especially in the mineral soil site, and the effects of prolonged flooding in 2010 and 2012. The drought conditions in 2008 resulted in decreased photosynthetic rates due to water stress (unpublished data; H. ˇ zková personal communication) and likely to reduced nutrient Ciˇ uptake rates, but greater biomass allocation to belowground structures. This is similar to the results of Dreesen et al. (2012), who found larger root mass under drier conditions in upland herbaceous species, but is counter to the results of Bai et al. (2010) again for upland plant species. NBPP was also significantly lower when there were the prolonged floods that occurred in the 2010 and 2012 growing seasons, a finding which is similar to results from other wetland studies (Lieffers and Shay, 1981; Jones et al., 2000; Baker et al., 2001). The emergence of anaerobic conditions due to prolonged flooding leads to decreased photosynthesis rates, which can result from increased stomatal closure, reduced Rubisco activity or lower photosynthate transport (Pezeshki, 2001; Chen et al., 2005, 2013). Reduced photosynthesis rates can lead to changed biomass allocation patterns as root growth can be affected earlier and to a greater extent than shoot growth (Kludze and DeLaune, 1994; Pezeshki, 2001). In addition, there is increased growth of new roots when oxygen is present, since that element is the most effective for oxidative phosphorylation (Mitsch and Gosselink, 2000; Pezeshki, 2001). Such effects have been clearly shown to occur in Lepidium latifolium (Chen et al., 2005), Typha domingensis (Chen et al., 2013) and Spartina patens (Pezeshki, 2001), but Rodgers et al. (2003) found that prolonged flooding had no effect on NBPP of Atlantic white cedar (Chaemaecyperis thyoides) while resulting in greater root abundance and longevity. The changes in root nutrient percentages and stoichiometry also appear to be linked more to site hydrology than nutrient additions, although these two factors may be interacting in the two sites (Güsewell et al., 2003). Drought can result in reduced plant nutrient uptake due to a lack of available soil water for transporting nutrients to the roots. At the same time, the slower growth rates of plants experiencing drought stress reduces their need for nutrients (Lambers et al., 1998). Prolonged flooding can also reduce plant nutrient uptake rates by affecting biogeochemical processes due to the onset of anaerobic conditions (Pezeshki, 2001). The increase in root C% to levels >40% with prolonged flooding likely means that there was a greater content of more structural material, such as lignin, in the roots over time (Barbosa et al., 2012). The increase in lignin content would produce roots of lower quality and decreased decomposability, a conclusion further supported by the significant increase in the C:N ratio in both wet grasslands stemming from the increase in C% coupled with decreased N% (van Vuuren et al., 1993; Fig. 2). Increased input of more recalcitrant compounds should not only retard litter and SOM decomposition, but also reduce C and other nutrient turnover rates even under conditions of increased nutrient inputs (Waldrop and Zak, 2006). The increase in more recalcitrant compounds should also likely affect soil microbial community composition thereby influencing the activities of particular enzymatic groups (Fog, 1988). Such changes have been observed in the soils from these two wet grasslands (Picek et al., 2008; Kaˇstovská et al., 2012; Edwards et al., 2015). Changes in soil biogeochemical processes resulting from prolonged flooding can also affect N and P availability. For N, ammonification tends to be the dominant process when water levels are within −10 cm from the soil surface (Hefting et al.,

K.R. Edwards / Ecological Engineering 84 (2015) 325–335

2004; Edwards et al., 2006). The dominant plant species in the two wet grassland sites (C. acuta, G. maxima) preferentially take ˇ ˚ cková, up nitrates compared to ammonium (Kaˇstovská and Santr uˇ 2011), therefore flooding could have limited the amount of available N to their roots. The effect of flooding on P uptake is more complicated, being influenced to a great extent by soil type (Chen et al., 2005). In this study, root P content significantly decreased with higher water levels in the organic soil site, but was unaffected in the mineral soil grassland. The lack of a flooding effect on root P content in the mineral soil site may be due to the greater amount of iron (Fe) in this grassland compared to the organic soil site (Picek et al., 2008). Under anaerobic conditions, Fe3+ is changed to Fe2+ , thereby releasing P sorbed to soil particles and increasing P solubility (Pezeshki, 2001; Chen et al., 2013; Prem et al., 2015). Thus, this greater P availability, and presumably increased plant uptake, may cancel out any reduction in P uptake due to the lack of oxygen in the rhizosphere. In this study, there are significant individual effects of nutrients or hydrology on NBPP and root nutrient percentages and stoichiometry, but the results also show that these two factors interact. The decrease in NBPP can be due to either increased nutrient availability or the effect of prolonged flooding. However, how NBPP was affected by changes in site hydrology differed depending on the nutrient treatment, especially in the mineral soil site (see Fig. 4). In addition, changes in site hydrology more significantly affected root nutrient percentages and stoichiometry in the high fertilization treatment of both sites compared to the unfertilized control treatment. Similar water level × nutrient interactions were noted by Neill (1990) and Güsewell et al. (2003). Further manipulative experiments are needed to make clearer how site hydrology interacts with nutrient availability in wet grasslands. The importance of soil type is another factor that requires further study. Unfortunately, the role of soil type could not be elaborated in this study as the differences in soil type between the two studied wet grasslands corresponded with differences in trophic status and management practices. Experiments under more controlled conditions are needed.

4.2. Management implications The climate change scenarios developed by the IPCC show an overall warming globally, but that the effect of climate change on water availability will differ between regions (IPCC, 2014). For example, Southern Europe should become much drier in the future with increased chance of droughts, while there is expected to be greater amounts of precipitation in Northern Europe (Kovats et al., 2014). The results of this study show possible responses of wet grassland plants belowground to both of these possible future conditions. The drier conditions expected in Southern Europe and other regions of the world would be expected to lead to greater water stress, resulting in plants having lower photosynthesis and growth rates with biomass allocated more to producing belowground structures at the expense of aboveground growth (this study; Picek et al., 2008; Edwards et al., 2015). On the contrary, wet grasslands in regions like Northern Europe, which are expected to have increased precipitation levels, mostly in the form of rain, would conceivably experience periods of prolonged flooding. In such a case, there would be reduced belowground production with structures of lower quality (higher C:N ratio), which would likely reduce rhizodeposition and root decomposition rates, thereby affecting nutrient turnover and sequestration (Olff et al., 1994). In either case, wet grassland plants would be subjected to increased stress leading to shifts in plant allocation patterns, changed plant structure quality and likely leading to changes in plant species composition and diversity (Sjöberg and Danell, 1983).

333

Wetland plants as a rule can adapt to water level changes, although this ability is species-specific. Certain wetland species, such as C. acuta, one of the dominant plant species found in the two wet grasslands studied in this project, can form tussocks when growing in areas which are flooded for longer periods of time. However, this possible adaptational response will likely not be available for European wet grassland species since many are still mown for hay production (Joyce and Wade, 1998; Tallowin and Jefferson, 1999) because mowing would prevent tussock formation. The formation of aerenchymal tissue is another adaptation to flooding, but the benefits of this adaptation are limited under prolonged flooding due to the increased competition for oxygen between the roots and rhizosphere (Pezeshki, 2001). Lastly, shoot elongation can allow the leaves to grow above the water level and re-connect to the atmosphere (Banach et al., 2009). However, there is a range even among wetland plant species in the degree of tolerance to flooding (Sjöberg and Danell, 1983; Güsewell et al., 2003). Since many wet grasslands in Europe are managed for hay production (Tallowin and Jefferson, 1999), one possible management tool is to use already existing water management structures (ditches, weirs) to produce the desired hydrologic conditions suitable for maintaining wet grasslands under either drier or wetter future climates. Good management practices would be required in order to maintain the proper functioning of these water management structures, including regular clearing of the ditches to prevent clogging in areas expected to have increased precipitation levels in the future, such as Northern Europe. Clogging of the drainage ditches may have been a possible extenuating factor resulting in the prolonged flooding of the mineral soil site in 2010 and 2012 (Edwards, personal observation). Since many wetland plants cannot tolerate long-periods of drought, such an approach could also be used to keep more water in site in places like Southern Europe which are expected to suffer more from future droughts (Kovats et al., 2014). Maintaining well-functioning wetlands, or restoring disturbed wetlands, in locations that may experience lower precipitation levels would also help to increase local water availability and assist in maintaining or developing smaller, closed water cycles within a given landscape, which could help to mitigate the negative effects of larger-scale climate change (Kravˇcik et al., 2008). Such an approach could help to connect wet grasslands and other wetlands to the surrounding landscape, resulting in the formation of a better functioning integrated human-ecological system that could be to the benefit of both human society and nature (Mitsch and Jørgensen, 2004). Further studies, political, social as well as ecological, will be required before such a picture can come close to realization. Acknowledgements ˇ Basin Biosphere Reserve This study was supported by the Tˇrebon and Protected Landscape Area. Numerous people helped both in the field and laboratory most notably Lukaˇs Bareˇs, Ville Närhi, ˇ Funding Mirka Káplová, Marketa Applová and Pavla Stanková. for this project came from grants from the Grant Agency of the ˇ Czech Republic (GACR: 526/06/0276; 526/09/1545; 13-17398S) and the Grant Agency of the University of South Bohemia (GAJU: 143/2010/P). The paper was greatly improved by the recommendations of two anonymous reviewers. References Arndal, M.F., Merrild, M.P., Michelsen, A., Schmidt, I.K., Mikkelsen, T.N., Beier, C., 2013. Net root growth and nutrient acquisition in response to predicted climate change in two contrasting heathland species. Plant Soil 369, 615–629. Bai, W., Wan, S., Niu, S., Liu, W., Chen, Q., Wang, Q., Zhang, W., Han, X., Li, L., 2010. Increased temperature and precipitation interact to affect root production,

334

K.R. Edwards / Ecological Engineering 84 (2015) 325–335

mortality, and turnover in a temperate steppe: implications for ecosystem C cycling. Glob. Change Biol. 16, 1306–1316. Baker, T.T., Conner, W.H., Lockaby, B.G., Stanturf, J.A., Burke, M.K., 2001. Fine root productivity and dynamics on a forested floodplain in South Carolina. J. Soil Sci. Soc. Am. 65, 545–556. Banach, K., Banach, A.M., Lamers, L.P.M., De Kroon, H., Bennicelli, R.P., Smits, A.J.M., Visser, E.J.W., 2009. Differences in flooding tolerance between species from two wetland habitats with contrasting hydrology: implications for vegetation development in future floodwater retention areas. Ann. Bot. 103, 341–351. Barbosa, R.I., dos Santos, J.R.S., da Cunha, M.S., Pimentel, T.P., Fearnside, P.M., 2012. Root biomass, root:shoot ratio and belowground carbon stocks in the open savannahs of Roraima, Brazilian Amazonia. Aust. J. Bot. 60, 405–416. Blank, R.R., Svejcar, T., Riegel, G., 2006. Soil attributes in a Sierra Nevada riparian meadow as influenced by grazing. Rangeland Ecol. Manag. 59, 321–329. Blue, J.D., Souza, L., Classen, A.T., Schweitzer, J.A., Sanders, N.J., 2011. The variable effects of soil nitrogen availability and insect herbivory on aboveground and belowground plant biomass in an old-field ecosystem. Oecologia 167, 771–780. Boyd, C.S., Svejcar, T.J., 2012. Biomass production and net ecosystem exchange following defoliation in a wet sedge community. Rangeland Ecol. Manag. 65, 394–400. Chen, G., Yang, Y., Robinson, D., 2014. Allometric constraints on, and trade-offs in, belowground carbon allocation and their control of soil respiration across global forest ecosystems. Glob. Change Biol. 20, 1674–1684. Chen, H., Qualls, R.G., Blank, R.R., 2005. Effect of soil flooding on photosynthesis, carbohydrate partitioning and nutrient uptake in the invasive exotic Lepidium latifolium. Aquat. Bot. 82, 250–268. Chen, H., Zamorano, M.F., Ivanoff, D., 2013. Effect of deep flooding on nutrients and non-structural carbohydrates of mature Typha domingensis and its post-flooding recovery. Ecol. Eng. 53, 267–274. Detenbeck, N.E., Galatowitsch, S.M., Atkinson, J., Ball, H., 1999. Evaluating perturbations and developing restoration strategies for inland wetlands in the Great Lakes basin. Wetlands 19, 789–820. Dodd, M.B., Mackay, A.D., 2011. Effects of contrasting soil fertility on root mass, root growth, root decomposition and soil carbon under a New Zealand perennial ryegrass/while clover pasture. Plant Soil 349, 291–302. Dreesen, F.E., De Boeck, H.J., Janssens, I.A., Nijs, I., 2012. Summer heat and drought extremes trigger unexcpected changes in productivity of a temperate annual/biannual plant community. Environ. Exp. Bot. 79, 21–30. Dukes, J.S., Chiariello, N.R., Cleland, E.E., Moore, L.A., Shaw, M.R., Thayer, S., Tobeck, T., Mooney, H.A., Field, C.R., 2005. Responses of grassland production to single and multiple global environmental changes. PLoS Biol. 3, e319. Dumortier, M., Verlinden, A., Beeckman, H., van der Mijnsbrugge, K., 1996. Effects of harvesting dates and frequencies on above and belowground dynamics in Belgian wet grasslands. Ecoscience 3, 190–198. Edwards, K.R., Mills, K.P., 2005. Aboveground and belowground productivity of Spartina alterniflora (smooth cordgrass) in natural and created Louisiana salt marshes. Estuaries 28, 252–265. ˇ zková, H., Zemanová, K., Santr ˇ ˚ cková, H., 2006. Plant growth and Edwards, K.R., Cíˇ uˇ microbial processes in a constructed wetland planted with Phalaris arundinacea. Ecol. Eng. 27, 153–165. ˇ zková, H., Máchalová-Zemanová, K., Stará, A., 2015. NutriEdwards, K.R., Picek, T., Cíˇ ent addition effects on carbon fluxes in wet grasslands with either organic or mineral soil. Wetlands 35, 55–68. ˚ Fiala, K., Tuma, I., Holub, P., 2009. Effect of manipulated rainfall on root production and plant belowground dry mass of different grassland ecosystems. Ecosystems 12, 906–914. Fidelis, A., di Santi Lyra, M.F., Pivello, V.R., 2013. Above- and below-ground biomass and carbon dynamics in Brazilian Cerrado wet grasslands. J. Veg. Sci. 24, 356–364. Finer, L., Laine, J., 2000. The ingrowth bag method in measuring root production on peatland sites. Scand. J. For. Res. 15, 75–80. Fog, K., 1988. The effect of added nitrogen on the rate of decomposition of organic matter. Biol. Rev. 63, 433–462. Gill, R.A., Jackson, R.B., 2000. Global patterns of root turnover for terrestrial ecosystems. New Phytol. 147, 13–31. Güsewell, S., Bollens, U., Ryser, P., Klötzli, F., 2003. Contrasting effects of nitrogen, phosphorus and water regime on first- and second-year growth of 16 wetland plant species. Funct. Ecol. 17, 754–765. Hefting, M., Clément, J.C., Dowrick, D., Cosandy, A.C., Bernal, S., Cimpian, C., Tatur, A., Burt, T.P., Pinay, G., 2004. Water table elevation controls on soil nitrogen cycling in riparian wetlands along a European climatic gradient. Biogeochemistry 67, 113–134. ´ S., Segal, S., 1998. General ecology of wetlands. In: Westlake, D.F., Kvˇet, J., Hejny, ´ A. (Eds.), The Production Ecology of Wetlands. Cambridge UniverSzczepanski, sity Press, Cambridge, UK, pp. 1–77. Intergovernmental Panel on Climate Change (IPCC), 2014. In: Core Writing Team, Pachauri, R.K., Meyer, L.A. (Eds.), Climate Change 2014: Synthesis Report. Contribution of Working Groups I, II and III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. IPCC, Geneva, Switzerland. Jones, R.H., Henson, K.O., Somers, G.L., 2000. Spatial, seasonal, and annual variation of fine root mass in a forested wetland. J. Torrey Bot. Soc. 127, 107–114. Joyce, C.B., Wade, P.M., 1998. European Wet Grasslands: Biodiversity, Management and Restoration. John Wiley and Sons, Chicester, UK. ˇ ˚ cková, H., 2011. Comparison of uptake of different N forms Kaˇstovská, E., Santr uˇ by soil microorganisms and two wet grassland plants: a pot study. Soil Biol. Biochem. 43, 1285–1291.

Kaˇstovská, E., Picek, T., Báarta, J., Mach, J., Cajthaml, T., Edwards, K., 2012. Nutrient addition retards decomposition and C immobilization in two wet grasslands. Hydrobiologia 692, 67–81. ˇ ˚ cková, H., Root physiological traits and distribution Kaˇstovská, E., Edwards, K., Santr uˇ of belowground carbon of two grasses as affected by N loading Appl. Soil Ecol. (unpublished data). Kearney, M.A., Zhu, W., 2012. Growth of three wetland plant species under single and multi-pollutant wastewater conditions. Ecol. Eng. 47, 214–220. Ket, W.A., Schubauer-Berigan, J.P., Craft, C.B., 2011. Effects of five years of nitrogen and phosphorus additions on a Zizaniopsis miliacea tidal freshwater marsh. Aquat. Bot. 95, 17–23. Kiley, D.K., Schneider, R.L., 2005. Riparian roots through time, space and disturbance. Plant Soil 269, 259–272. Kludze, H.K., DeLaune, R.D., 1994. Methane emission and growth of Spartina patens in response to soil redox intensity. Soil Sci. Soc. Am. J. 58, 1838–1845. Kovats, R.S., Valentini, R., Bouwer, L.M., Georgopoulou, E., Jacob, D., Martin, E., Rounsevell, M., Soussana, J.F., 2014. Europe. In: Barros, V.R., Field, C.B., Dokken, D.J., Mastrandrea, M.D., Mach, K.J., Bilir, T.E., Chatterjee, M., Ebi, K.L., Estrada, Y.O., Genova, R.C., Girma, B., Kissel, E.S., Levy, A.N., MacCracken, S., Mastrandrea, P.R., White, L.L. (Eds.), Climate Change 2014: Impacts, Adaptation, and Vulnerability. Part B: Regional Aspects. Contribution of Working Group II to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, United Kingdom/New York, NY, USA, pp. 1267–1326. ´ J., Kohutiar, J., Kováˇc, M., Tóth, E., 2008. Water for the Recovery Kravˇcik, M., Pokorny, of the Climate. Typopress, Koˇsice, Slovak Republic. Ladwig, L.M., Collins, S.L., Swann, A.L., Xia, Y., Allen, M.F., Allen, E.B., 2012. Above- and belowground responses to nitrogen addition in a Chihuahuan Desert grassland. Oecologia 169, 177–185. Lambers, H., Chapin, F.S., Pons, T.L., 1998. Plant Physiological Ecology. SpringerVerlag, New York. Lieffers, V.J., Shay, J.M., 1981. The effects of water level on the growth and reproduction of Scirpus maritimus var. paludosus. Can. J. Bot. 59, 118–121. Martin, D., Chambers, J., 2002. Restoration of riparian meadows degraded by livestock grazing: above- and below-ground responses. Plant Ecol. 163, 77–91. McKee, K.L., Faulkner, P.L., 2000. Restoration of biogeochemical function in mangrove forests. Restor. Ecol. 8, 247–259. Meinen, C.H., Leuschner, C.D., 2009. Root growth and recovery in temperate broad-leaved forest stands differing in tree species diversity. Ecosystems 12, 1103–1116. Milchunas, D.G., Mosier, A.R., Morgan, J.A., LeCain, D.R., King, J.Y., Nelson, J.A., 2005. Root production and tissue quality in a shortgrass steppe exposed to elevated CO2 : using a new ingrowth method. Plant Soil 268, 111–122. Mitsch, W.J., Gosselink, J.G., 2000. Wetlands, third ed. Wiley, New York. Mitsch, W.J., Jørgensen, S.E., 2004. Ecological Engineering and Ecosystem Restoration. John Wiley and Sons, Hoboken, NJ, USA. Moar, S.E.L., Wilson, S.D., 2006. Root responses to nutrient patches in grassland and forest. Plant Ecol. 184, 157–162. Nadelhoffer, K.J., Johnson, L., Laundre, J., Giblin, A.E., Shaver, G.R., 2002. Fine root production and nutrient content in wet and moist arctic tundras as influenced by chronic fertilization. Plant Soil 242, 107–113. Neill, C., 1990. Effects of nutrients and water level on emergent macrophyte biomass in a prairie marsh. Can. J. Bot. 68, 1007–1014. Olde Venterink, H., Davidsson, T.E., Kiehl, K., Leonardson, L., 2002. Impact of drying and re-wetting on N, P and K dynamics in a wetland soil. Plant Soil 243, 119–130. Olff, H., Berendse, F., De Visser, W., 1994. Changes in nitrogen mineralization, tissue nutrient concentrations and biomass compartmentation after cessation of fertilizer application to mown grassland. J.Ecol. 82, 611–620. Persson, H., 1979. Fine-rrot production, mortality and decomposition in forest ecosystems. Vegetatio 41, 101–109. Pezeshki, S.R., 2001. Wetland plant responses to soil flooding. Environ. Exp. Bot. 46, 299–312. Picek, T., Kaˇstovská, E., Edwards, K., Zemanová, K., Duˇsek, J., 2008. Short term effects of experimental eutrophication on carbon and nitrogen cycling in two types of wet grassland. Community Ecol. 9, 81–90. Prach, K., 2002. Human impact on vegetation around Roˇzmberk pond. In: Kvˇet, J., Jeník, J., Soukupová, L. (Eds.), Freshwater Wetlands and Their Sustainable Future. Parthenon Publishing, Boca Raton, FL, USA, pp. 187–194. Prach, K., 2008. Vegetation changes in a wet meadow complex during the past halfcentury. Folia Geobot. Phytotax. 43, 119–130. Prach, K., Straˇskrabová, J., 1996. Meadows in the Luˇznice River floodplain in ˇ Biosphere Reserve – potential for restoration. Pˇríroda, Praha., pp. the Tˇrebon 163–168. Prem, M., Hansen, H.C.B., Wenzel, W., Heiberg, L., Sørensen, H., Borggaard, O.K., 2015. High spatial and fast changes of iron redox state and phosphorus solubility in a seasonally flooded temperate wetland soil. Wetlands 35, 237–246. ˇ K., Jeník, J., 2002. Climatic and hydrologic setting of the wet meadows. In: Pˇribán, Kvˇet, J., Jeník, J., Soukupová, L. (Eds.), Freshwater Wetlands and Their Sustainable Future. Parthenon Publishing, Boca Raton, FL, USA, pp. 231–241. Rejmánková, E., Macek, P., Epps, K., 2008. Wetland ecosystem changes after three years of phosphorus addition. Wetlands 28, 914–927. Rodgers, H.L., Day, F.P., Atkinson, R.B., 2003. Fine root dynamics in two Atlantic white-cedar wetlands with contrasting hydroperiods. Wetlands 23, 941–949. Ruˇzíˇcká, M. (Ed.), 1994. Ecological potential of floodplain area of the River Morava. Ekologia (Bratislava) 13, 1–216.

K.R. Edwards / Ecological Engineering 84 (2015) 325–335 Saggar, S., Hedley, C., Mackay, A.D., 1997. Partitioning and translocation of photosynthetically fixed 14 C in grazed hill pastures. Biol. Fert. Soils 25, 152–158. Ságová-Mareˇcková, M., Petrusek, A., Kvˇet, J., 2009. Biomass production and nutrient accumulation in Sparganium emersum Rehm. after sediment treatment with mineral and organic fertilisers in three mesocosm experiments. Aquat. Ecol. 43, 903–913. Sjöberg, K., Danell, K., 1983. Effects of permanent flooding on Carex-Equisetum wetlands in Northern Sweden. Aquat. Bot. 15, 275–286. Sokal, R.R., Rohlf, F.J., 1995. Biometry. WH Freeman, San Francisco. Sonderegger, D.L., Ogle, K., Evans, R.D., Ferguson, S., Nowak, R.S., 2013. Temporal dynamics of fine roots under long-term exposure to elevated CO2 in the Mojave Desert. New Phytol. 138, 127–138. Steingrobe, B., 2005. Root turnover of faba beans (Vicia faba L.) and its interaction with P and K supply. J. Plant Nutr. Soil Sci. 168, 364–371. Steingrobe, B., Schmid, H., Claassen, N., 2000. The use of the ingrowth core method for measuring root production of arable crops – influence of soil conditions inside the ingrowth core on root growth. J. Plant Nutr. Soil Sci. 163, 617–622. Sterner, R.W., Elser, J.J., 2002. Ecological Stoichiometry: The Biology of Elements from Molecules to the Biosphere. Princeton University Press, Princeton, NJ.

335

Tallowin, J.R.B., Jefferson, R.G., 1999. Hay production from lowland semi-natural grasslands: a review of implications for ruminant livestock systems. Grass Forage Sci. 54, 99–115. Tilman, D., 1987. Secondary succession and the pattern of plant dominance along experimental nitrogen gradients. Ecol. Mono 57, 189–214. van Vuuren, M.M.I., Berendse, F., de Visser, W., 1993. Species and site differences in the decomposition of litters and roots from wet heathlands. Can. J. Bot. 71, 167–173. Vogt, K.A., Vogt, D.J., Bloomfield, J., 1998. Analysis of some direct and indirect methods for estimating root biomass and production of forests at an ecosystem level. Plant Soil 200, 71–89. Waldrop, M.P., Zak, D.R., 2006. Response of oxidative enzyme activities to nitrogen deposition affects soil concentrations of dissolved organic carbon. Ecosystems 9, 921–933. ˇ zková, H., Edwards, K., Santr ˇ ˚ cková, H., 2008. Soil CO2 efflux in Zemanová, K., Cíˇ uˇ three wet meadow ecosystems with different C and N status. Community Ecol. 9, 49–55. Zhang, Q.H., Zak, J.C., 1998. Effects of water and nitrogen amendment on soil microbial biomass and fine root production in a semi-arid environment in West Texas. Soil Biol. Biochem. 30, 39–45.