Restoration of plant species diversity of ditch banks : ecological constraints and opportunities

(1)

Restoration of plant species diversity of ditch banks : ecological

constraints and opportunities

Blomqvist, M.M.

Citation

Blomqvist, M. M. (2005, February 3). Restoration of plant species diversity of ditch banks :

ecological constraints and opportunities. Retrieved from https://hdl.handle.net/1887/592

Version:

Not Applicable (or Unknown)

License:

Licence agreement concerning inclusion of doctoral thesis in the

_{Institutional Repository of the University of Leiden}

Downloaded from:

https://hdl.handle.net/1887/592

(2)

Chapter 4 The effects of agri-environment schemes on productivity

and plant species diversity in ditch banks

M. M. Blomqvist & W. L. M. Tamis

Abstract

(3)

Chapter 4 70

Introduction

Agri-environment schemes (AES) are implemented in many European countries to conserve and improve the diversity of the fauna and flora of agricultural landscapes (Beintema & Müskens 1987; Ovenden et al. 1998; Kleijn et al. 2001; Musters et al. 2001; Carey 2001). Yet, the effectiveness of these schemes remains controversial, particularly for improving botanical values (Cameron 2001; Kleijn & Sutherland 2003; Critchley et al. 2003). Botanical AES are frequently applied in (wet) grasslands, hayfields and ditch banks. Although the prescribed botanical management has proved to be effective under experimental conditions in nature reserves (Bakker 1989) and on farmland (van Strien 1991; Melman 1991; van der Linden & de Jong 1994), implementing this management in practice within AES has not resulted in a consistent improvement of plant diversity (Kleijn and Sutherland 2003). The ecological causes behind these successes and failures, however, still remain unclear.

In the Netherlands, botanical AES are most commonly applied in ditch banks, which function as an important refuge for many formerly common grassland, wetland and hayfield species (Clausman & van Wijngaarden 1984; Westhoff & Weeda 1984; Melman 1991; Blomqvist et al. 2003b). The recommended management was similar within all types of AES until 2000: no fertilisation, ‘careful’ ditch cleaning (little disturbance, no ditch sediment or plant parts in the ditch bank) and extensive grazing and later first mowing dates (van Strien 1991; Melman & van Strien 1993; LNV 1995, but see below). The newest schemes (DLG 2000) still focus on nutrient reduction, but do not have any recommendations with regard to the timing of mowing (and grazing). All in all, the main critical ecological elements of these management recommendations has thus been a reduction in productivity and, especially in the earlier schemes, allowing later flowering species to set seed.

(4)

Agri-environment schemes, productivity and species richness 71

spring and summer, which may enhance competition for light and reduce germination opportunities with low species richness as a result (Bakker 1989). Thirdly, colonisation or regeneration constraints caused by the lack of a seed bank or nearby source populations, results in a time-lag in the recovery of the species richness (Oomes & van der Werf 1996; Bakker & Berendse 1999).

In the present study, we focus on the ecological limitations of current management recommendations. Since the main critical ecological element in ditch bank agri-environment schemes is the reduction of nutrients, we focus on productivity. Firstly, we report changes in productivity and diversity in study ditch banks on six dairy farms (situated in six different areas) in the Western Peat District in the Netherlands after 10 years of AES. In five areas, we compare these results with data from surrounding reference ditch banks mostly without AES. Secondly, we investigate productivity-diversity patterns and the role of a high starting productivity in space and time. Finally, we discuss alternative explanations including dispersal constraints, differences between the recommended and the actual applied management, and the possible effects of regional and national factors.

We answer the following research questions:

1. Is the vegetation diversity increasing in the study ditch banks with AES in comparison with the surrounding reference ditch banks?

2. Is the applied agri-environment management successful in reducing productivity compared with the surrounding reference?

3. Are productivity and diversity negatively correlated as predicted by theory? Is an actual change in productivity in time followed by a change in diversity? Does starting productivity affect the changes in diversity?

Material and methods Research area

The research area (51°51’N - 52°07’N and 4°45’E - 5°08’E) is situated mainly in the Province of South-Holland in the Western Peat District in the Netherlands. The dominant soil types are peat and peat with clay. Most land is utilised as pasture for dairy cattle and sheep. Typically, the pastures have 0.5-1.5 m wide field edges (ditch banks) of varying species richness. Ditch water levels in the permanently water-filled ditches are artificially controlled. Winter levels are normally some 10-15 cm below summer levels. Dataset

(5)

Chapter 4 72

AES since 1992 (Kruk et al. 1994). Within these schemes a ‘no cure, no pay’ system is applied, where the farmer is free to choose type of ditch bank management, but encouraged to follow the recommended management (see introduction). Data from these study ditch banks were collected in 1993-95 (July-August) and in 2000-2003 (mainly in May, Table 1). To check for possible differences in the actual management of these study ditch banks, we made an inventory of the general management strategy (since 1992) using questionnaires in 2000. Actual management was checked monthly during field work in the period 2000-2003. Information about changes in water levels were obtained from water boards in each area (Table 2).

In five areas, data from the study ditch banks were compared with ditch bank data in their respective surroundings (reference ditch banks) for the period 1984-2002 (Provincie Zuid-Holland 1993). The time-span 1984-1992 was included to get a better picture of long-term developments in the reference ditch banks (assumed to be representative of the general development in each area). Soil type and water management was similar in study ditch banks and in the respective reference. Depending on the area, between 3 and 25% (Table 2) of the area surrounding the study ditch banks was under some kind of management scheme. The reference ditch banks therefore functioned as a partial control to the effects of AES (as a system) on the vegetation.

The data were divided into five sampling periods with approximately the same number of years in each period: 1) 1984-1986 (‘starting situation’), 2) 1989-1992, 3) 1993-1995 (‘begin study period’), 4) 1996-1998 and 5) 1999-2003 (‘end study period’). The number of sampled ditch banks varied per period (Table 1). Of the study ditch banks, 42 sampled in the begin period (93-95) were sampled at least once in the end period (99-03; repeated measures). Of the reference ditch banks 78 were repeatedly sampled in 93-95 and 99-03.

Field measurements

The vegetation was recorded in 50 m long plots, their width varied with the ditch bank. This plot size is most commonly used for ditch bank analyses as it represents the total ditch bank species richness (Melman et al. 1991). Nomenclature follows (van der Meijden 1996). To ensure a proper comparison between datasets, some taxa were combined. Per ditch bank plot, we recorded presence and % cover of each taxon (referred to as ‘species’ for convenience).

Biomass samples were taken during the second or third week in May in the study ditch banks in 2000-2003 (Table 1) in those vegetation plots that had not been mown or grazed yet. In each 50 m plot, five replicate samples were taken at regular intervals and dried at 70 °C for at least 3 days. Weights were expressed as g dry weight / m2_{. To investigate the biomass development throughout the season, additional samples}

(6)

Table 1. Structure of the dataset. Sampling period, the years involved and total numbers of sampled ditch banks and sampling month

indicated. Study db = ditch banks on six farms practising ditch bank management according to agri-environment schemes, vegetation and biomass data indicated; Reference db = ditch banks mostly without AES (see Table 2) in the surrounding of the study db for five areas (no data available for area 5).

Sample Years Number of ditch banks sampled per month

period Study db Reference db

Vegetation Biomass Vegetation

1 Starting situation 1984-86 - - 172 May-Oct

2 1989-92 - - 192 Jun-Sep

3 Begin study period 1993-95 156 (52x3) Jul-Aug* - 354 May-Sep**

4 1996-98 - - 177 May-Sep

5 End study period 1999-02 293 May-Sep**

2000 12 May* 9 May

2002 44 May*, 12 Jul 34 May, 12 Jul, 12 Oct

2003 73 May*, 12 Jul 72 May, 12 Jul, 12 Sep

* 42 study ditch banks in six areas were sampled in the period 1993-95 and at least once in the period 1999-03. ** 78 reference ditch banks in five areas were repeatedly sampled in the period 1993-95 and 1999-03.

(7)

Table 2. Management characteristics for study ditch banks with AES in six areas (one farm per area). Ditch cleaning once a year on every

farm. Num = area number; Changes in water levels: cm change and year indicated; Soil type: P = peat, P+C = peat + clay; Stocking rate = measure of number of ‘standard’ grazers per ha; Grazers: C = cattle, S = sheep; First mowing = timing of first mowing of ditch banks; Fertilization = fertilizer input in ditch banks; Deposition = deposition of ditch plant parts and moderate amounts of ditch sediment in ditch bank; Ditch cleaning disturbance = disturbance caused by ditch cleaning: ranked 1 = minimal – 4 = severe; Ref db = reference ditch banks; % area with AES = approximate % of ditch banks with an agri-environment scheme in the reference area surrounding the study ditch banks (visually estimated using maps provided by various agricultural organisations, see Acknowledgements). Estimated overall intensity of management on study farms: 4 < 6 < 1 ≤ 5 ≤ 3 < 2.

Area Study db Ref db

Num Changes in

water levels 1992-2003

Soil

type Stocking rate Gra-zers First mowing Fertili-sation Deposition Ditch cleaning disturbance Other modifications to recommended management % area with AES 1 -2 to -5 cm

(1998) P 1.8 C begin June No High edge 3 - 25%

2 -3 to -5 cm

(1998)

P+C 2.5 C+S mid May No Ditch bank

(thick layer) 4 Db grazed short in winter by sheep 3% 3 -3 to -8 cm (1998)

P+C 1.8 C begin June No High edge /

ditch bank

2 grazing in early April 25%

4 0 cm P+C 2.0 C end June No High edge 1 “drinking tub” 10%

5 +5 cm (2002) P 2.5 C+S begin June No Ditch bank

(plant parts) 2 Db grazed short in winter by sheep No data 6 -8 cm (1996) -2 cm (2002)

P 2.0 C+S begin July No High edge /

ditch bank

3 Db grazed short in

winter by sheep

7%

Recommendations low - end June /

begin July

No High edge 1 -

74 Chap

ter

(8)

Productivity and diversity

Productivity is often expressed as standing biomass, which can be seen as an overall measure reflecting competition for nutrients and light and levels of disturbance (Wilson & Keddy 1988). Since we only had biomass data from the study ditch banks in end period (99-03), we needed other productivity measures to be able to investigate changes in time and to compare study and reference ditch banks. We therefore selected two other productivity parameters that could be indirectly derived from the botanical data: Ellenberg N-values (Ellenberg et al. 1992; Wiertz 1992; Hill et al. 1999) and grass/forb ratios (no. of grass species / no. of forb species). N-values reflect overall nutrients available to plants (Hill & Carey 1997; Ertsen et al. 1998). As for grass/forb ratios, previous studies have shown that the graminoid life form often dominates (both in terms of cover and proportion of species) under conditions with higher productivity. This dominance seems to be related to the fact that forb species are inferior competitors under low light conditions (Kull & Aan 1997), and/or to a more homogeneous light regime in grassy swards, reducing germination opportunities (Willems et al. 1993; Bosy & Reader 1995; Willems & Nieuwstadt 1996). Productivity (per plot) was thus defined: 1) biomass in May (g / m2_{), 2) N-value, 3) grass/forb ratio (no. of grass species / no. of forb}

species). Average N-values and grass/forb ratios were calculated by taking the unweighted species values per plot rather than values weighted by % cover, since cover values of individual species are particularly susceptible to seasonal variation (Melman et al. 1988a; Hill & Carey 1997; Ertsen et al. 1998; Schaffers & Sykora 2000).

Diversity was defined as: 1) the number of species per plot and 2) the number of ‘target’ species per plot (Clausman & van Wijngaarden 1984, van Harmelen et al. 1997, modified by Blomqvist et al. 2003a; see also Chapter 2: Appendix 1). According to this definition approximately one third (38%) of the taxa in this dataset could be defined as ‘target species’. Hereafter, diversity refers to numbers of species and target species.

Statistical analyses

All analyses were performed using the statistical package SPSS 11.5.0. To ensure homogeneity of variances, all variables were log-transformed before analyses. This also means that values on differences in time reflect relative changes (% increase or decrease). Backtransformed values are given in the results.

Differences between May and July measurements

(9)

Chapter 4 76

July (in 2002 and 2003) in 12 ditch banks that were excluded from grazing. GLM with month as fixed and year, area and ditch bank number within area as random factors showed that there were no significant differences between May and July measurements in the following variables: number of species, number of target species, N-values and the grass/forb ratio. Changes in these variables between 1993-95 and 1999-03 were thus not related to differences in sampling month, but reflect real changes between the begin and end study period.

Patterns of change in plant species diversity and productivity

Study ditch banks with AES 1993-1995 vs. 1999-2003 and long term changes in reference ditch banks 1984-2002. In the study ditch banks, we analysed the 42

repeatedly sampled plots together with a GLM with period as fixed, area as random factor and (ditch bank) plot within area as random factor. The interaction period*area was included in the model. In reference ditch banks, all reference plots in all areas together with a GLM with period as fixed and area as random factor. A significant interaction would indicate that the development in time differed per area. Values were averaged per plot within each period before the analysis.

Direction and development of diversity and productivity in study and reference ditch banks between 1993-1995 and 1999-2003. We compared mean (per period) and changes

(difference in mean value 99-03 vs. 93-95) in diversity and productivity between study and reference ditch banks. Only those plots that had been repeatedly sampled in the begin and end period were used. Differences were tested with a GLM analysis with treatment (study vs. reference) and period (93-95 vs. 99-03) as fixed factors and area as random factor. All interactions were included in the model. Plot within treatment and area was included as random factor.

Diversity-productivity relationships in the study ditch banks

Spatial productivity-diversity patterns. In the study ditch banks, we investigated whether

(10)

Changes in time in productivity and diversity. Since we are investigating the effects of

AES we are also interested in actual changes in time between the begin and the end period. We wanted to know whether an actual change in productivity is followed by a change in diversity. We calculated the average change in diversity and productivity per year with linear regression in the 42 repeatedly sampled plots. This was followed by a linear regression of the change in diversity (∆D) as a function of the change in productivity (∆P): ∆D = a+ b*∆P. We were also interested in the possible effects of starting productivity on the change in diversity. As above, a clearly stronger linear relationship with log-transformed data compared to a linear relationship with untransformed data would indicate the presence of a ‘tail’ and would thus confirm such an effect.

Results

Patterns of change in vegetation diversity and productivity

Study ditch banks with AES 1993-1995 vs. 1999-2003 and long term changes in reference ditch banks 1984-2002. In the study ditch banks, diversity usually decreased,

while productivity increased. Across all study ditch banks, the number of species decreased significantly between the begin (93-95) and end (99-03) period (means 30.8 vs. 29.1; Table 3, Figs. 1a,e). The overall number of target species, on the other hand, did not differ significantly between the two periods (means 6.61 vs. 6.53); in two areas, however, a (non-significant) increase took place (Figs. 1b,f). The change in N-values differed between farms (significant interaction area*period, Table 3); N-values usually increased, but decreased in two areas (Figs. 1c,g). Overall means increased from 5.82 to 5.93. The grass/forb ratio increased in time in all areas (means 0.36 vs. 0.43; Figs. 1d,h).

Table 3. Results of GLM analyses (F-values) of effect of period (93-95 vs. 99-03), area

and interaction on diversity and productivity in study ditch banks with agri-environment schemes in six areas. F = fixed factor, R = random factor. Plot within area was included as a random factor. A significant interaction period*area indicates that the development in time differs between areas. Significance levels: +_{p < 0.1, * p < 0.05, ** p < 0.01, ***} p < 0.0001, n = 42. See fig 1. for direction of change.

Variable df Diversity Productivity

(11)

(12)

The direction of change in time in diversity and productivity of the reference ditch banks differed between the five areas (interaction factor period*area (near-) significant for all variables; Figs. 2a-d). All areas started with a similar number of species (1984-86), but the development in time differed: species numbers were decreasing in three and increasing or remaining stable in two areas. Numbers of target species fluctuated considerably in time: generally numbers were maintained or increased in three areas and decreased in two. Areas differed in their productivity in the starting situation (1984-86). N-values tended to increase or to be maintained, converging towards the end period (1999-03). Grass/forb ratios generally remained stable in the long term.

Study vs. reference ditch banks 1993-1995 vs. 1999-2003. For total number of species

we found a significant effect of treatment (Table 4); the mean number of species was on average 15% higher in the study ditch banks than in the reference (Fig. 1a). Overall, areas also differed significantly in species richness. The interaction period*treatment was significant at a 0.1-level, indicating that the differences between study and reference ditch banks tended to become smaller in time as a consequence of declining species richness in the study plots (Fig. 1e). For the number of target species we found a significant interaction between area and treatment (Table 4), which indicates that study ditch banks did not always have more target species than the reference. The number of target species was higher in the study ditch banks in three and lower in two areas compared with the reference (Fig. 1b). The fact that period*treatment was not significant, means that no significant differences were found between study and reference ditch banks in the pattern of change through time in target species (Fig. 1f).

For N-values we found a significant three-way interaction (Table 4), indicating that the effects of period and treatment differed per area. N-values were usually lower in study ditch banks than in the reference (Fig. 1c). N-values increased more on three and decreased more in two study areas (3 and 4) compared with the reference (Fig. 1g). For grass/forb ratios we found a significant treatment effect; on average, the grass/forb ratios were 19% lower in the study ditch banks than in reference (Fig. 1d). This was mainly due to higher numbers of forbs, rather than lower numbers of grasses in study ditch banks. Grass/forb ratios always increased in study ditch banks, but decreased in the

Fig. 1. (opposite page). Differences between study and reference ditch banks in five

(13)

Chapter 4 80

reference in three areas (Fig. 1h). A near-significant period*treatment effect for the grass/forb ratio reflected the fact that differences between study and reference ditch banks tended to become smaller in time.

Overall, we thus found that diversity tended to be higher and productivity lower in study ditch banks compared with the reference. As for change in productivity and diversity in time, we did not find consistent patterns in all areas. However, on the whole, the desired effects, i.e. a decrease in productivity and an increase in diversity in study ditch banks compared with the reference, did not materialise.

Fig. 2. Development of diversity: (a) number of species, (b) number of target species

(14)

Table 4. Results of GLM analyses (F-values) of effect of period (93-95 vs. 99-03),

treatment (study with AES vs. reference without AES), area and interactions on diversity and productivity in five areas. F = fixed factor, R = random factor. Plot within area and treatment was included as a random factor. Significance levels: +_{p < 0.1, * p < 0.05, **} p < 0.01, *** p < 0.0001, n = 112. See fig 1. for differences in levels and direction of

change.

Variable df Diversity Productivity

No. of species No. of target species N-value Grass/forb ratio Period (F) 1 0.231 0.050 0.525 0.451 Treatment (F) 1 12.970* 0.984 3.441 8.534* Area (R) 4 9.843* 2.280 3.486 1.238 Period*treatment 1 4.445+ _2.271_0.413_7.249+ Treatment*area 4 1.126 3.306* 0.325 1.859 Period*area 4 1.753 3.489 1.351 4.020 Period*treatment*area 4 1.443 1.482 4.306** 1.450 Plot*area*treatment (R) 102 3.040*** 4.358*** 4.272*** 2.406*** Diversity-productivity relationships

Spatial productivity-diversity patterns. The numbers of species and target species were

lower at higher N-values (N), higher grass/forb ratios (GF) and larger above-ground biomass (BIO; Table 5, Fig. 3). The (partial) r2_{-values were very similar for the linear}

relationships with untransformed data and with transformed data indicating that models were essentially linear. In other words, we found no indication of a Grime ‘tail’ in the spatial patterns. Both number of species and target species showed the strongest relationship with the grass/forb ratio. N-values were strongly related to the number of target species but less strongly to the number of species. Biomass in May, on the other hand, was more strongly related to the number of species than to number of target species. Productivity parameters tended to be weakly positively correlated (partial r2 -values after correction for years: N-GF (93-95): r2_{= 0.002, p = 0.594, N-GF (99-03): r}2 = 0.025, p = 0.074, N-GF (total): r2_{= 0.009, p = 0.114, BIO-N (99-03): r}2_{= -0.000, p =} 0.893 and BIO-GF (99-03): r2_{= 0.064, p = 0.007).}

Changes in time in productivity and diversity. An increase in N-values was accompanied

(15)

Chapter 4 82

these log-transformed data gave very similar results to untransformed data, starting productivity did not appear to have an effect on the change in diversity in the present study.

Table 5. Relationship between productivity and diversity in study ditch banks for

different time periods. Productivity: N = N-values, GF = grass/forb ratios, Bio = Biomass in May g/m2_{; Diversity: No. species = number of species, no. target species =}

number of target species, n = no. observations. Explained variance (GLM: partial r2 -values) given after correction for year as random factor. Significance: ** p < 0.01, *** p < 0.0001.

Year & Productivity Diversity (partial r2_-values)1 _n

Month No. of species No. of target species

93-95 June N 0.016 0.243*** 156 GF 0.430*** 0.336*** 156 99-03 May N 0.144*** 0.240*** 130 GF 0.336*** 0.355*** 130 Bio 0.127*** 0.091** 115 Total N 0.059*** 0.241*** 286 GF 0.387*** 0.357*** 286

1_{All relationships were negative.}

Discussion

Changes in diversity in ditch banks with AES and in the surrounding reference ditch banks

As a rule, the plant diversity decreased or remained unchanged in ditch banks on farms with AES. Although the number of target species increased in study ditch banks with AES in two areas, the changes in diversity in time tended to be similar to the changes found in the reference ditch banks (Figs. 1e,f, Table 4). In other words, according to our findings, although diversity was sometimes maintained in ditch banks with AES compared with the reference, these schemes (as a system) do not (consistently) lead to an increase in diversity. What is the role of productivity in explaining these patterns? Productivity-diversity relationships in ditch banks

(16)

Fig. 3. Relationship

(17)

Chapter 4 84

(1991); ter Heerdt et al. (1991); see review by Grace (1999)). Since an actual decrease in productivity was (usually) coupled with an actual increase in diversity (Fig. 4), we infer that we are dealing with a causal relationship (Aarssen 2001). A reduction in productivity should therefore have a positive effect on diversity.

Fig 4. Changes in productivity (N-values and grass/forb ratios) vs. changes in species

(18)

What does the parameter ‘productivity’ mean in an ecological sense? All three productivity parameters were positively related to one another. However, the low explained variances indicate that they represent different aspects of productivity. As explained in the methods, biomass can be seen as an overall measure reflecting the effects of competition and disturbance; N-values reflect nutrients available to plants, and grass/forb ratios reflect light conditions and germination opportunities. How were these productivity parameters related to diversity?

The number of species and number of target species reacted differently to different productivity parameters. In accordance with the findings of Willems & Nieuwstadt (1996), the number of species was more strongly related to (changes in) grass/forb ratios than to N-values (or biomass, see Table 5, Fig. 4). Under the assumption that grass/forb ratios reflect germination opportunities rather than ‘available nutrients’, the present study lends support to earlier findings that productivity affects the species richness more via the indirect effects of nutrients on germination than a direct effect of nutrients themselves (Blomqvist et al. 2003a; Blomqvist et al. 2003b). Target species, on the other hand, were about equally strongly related to N-values and grass/forb ratios, which may be an indication that nutrients per se are also important for this plant group.

We also see that the changes in productivity are more strongly related to changes in the number of target species than changes in the number of species (higher regression coefficients: Fig. 4). It thus seems that target species are more sensitive overall to more productive conditions and that other factors are relatively more important for determining the number of species.

The influence of high starting productivity on (changes in) diversity

Since management recommendations do not only advocate a reduction in productivity, but often also recommend late mowing to promote flowering and seed set, an ecological conflict could arise in situations with high productivity (Melman 1991). The humpback curve predicts that, in highly productive situations (the ‘tail’ of the Grime curve (Grime 1973, Al-Mufti et al. 1977)), large changes in productivity will not result any clear change in diversity. However, in the present study, we found that (both spatial and time) relationships between productivity and diversity were linear, indicating that a decrease in productivity was accompanied by an increase in diversity and vice versa. This was surprising, because previous studies have found clear increases only below certain threshold values. Species diversity in grassland usually starts decreasing when yearly biomass production exceeds 600-700 g / m2_{(Oomes 1992) and tends to ‘level off’ above}

a yearly biomass production of 900-1000 g / m2_{(Melman 1991; Oomes & van der Werf}

2003). Since the mean total biomass production in our study ditch banks was around 900 g / m2_{, with maximum values up to 1700 g / m}2_{(Appendix 1), we had expected to find a}

(19)

Chapter 4 86

75% of these plots have total biomass production values lower than 1000 g / m2_(roughly

translating to a production of 200-300 g / m2_{in May). The fact that most of our other}

ditch banks lie below the ‘critical limit’ (Fig. 3) may therefore account for the lack of a clear effect of high starting productivity.

Actual productivity in ditch banks with AES and in the surrounding

Productivity is important for determining the plant species diversity in ditch banks, but what changes in productivity are actually taking place in the study ditch banks with AES? Contrary to the objectives of the recommended management, the productivity (both grass/forb ratios and N-values) was increasing in ditch banks with AES in most areas (Figs. 1g,h). The changes in time in study ditch banks, in comparison with the reference, differed per area (Table 4), but values usually followed the same pattern or increased more in study ditch banks. In other words, ditch bank agri-environment schemes as a system are not able to effectively reduce productivity. This at least partly explains why the expected increase in diversity has failed to materialise in these ditch banks.

Why is productivity not decreasing? Is this pattern related to ‘faulty’ management recommendations, deviations in the local practical application of the recommendations at farm level, or a result of regional influences such as changes in water levels and atmospheric deposition? For grass/forb ratios, values in study ditch banks tended to converge with those of the reference ditch banks. This could indicate that regional factors are overruling any local management. Alternatively, later mowing in ditch banks with AES may favour grasses over forbs, in effect resulting in increasing competition for forbs (Ryser et al. 1995). For N-values, since the development of in time differs between study and reference ditch banks in different areas (Table 4, Figs. 1c,g), local management seems to be relatively more important than regional influences. The enormous variation in actual management in the study ditch banks in the present study (Table 2) makes it difficult to pinpoint any single cause. However, if we take the two most extreme study areas in terms of intensity, where management is least intensive (and most restrictive with nutrients) in area 4 and most intensive (highest input of nutrients via ditch sediment) in area 2, we see that they also represent extremes, especially in changes in target species and values. In study ditch banks in area 4, where the N-values decreased most, the largest increases in target species could also be seen and vice versa in area 2 (Figs. 1 f,g). This illustrates that the way agri-environment recommendations are carried out locally by farmers, will have important effects on the successes and failures as agri-environment schemes as a system.

(20)

variances (r2_{-values, Fig. 4) in the time pattern analysis dropped dramatically in} comparison with the spatial analyses (R2_{-values, Table 5). Previous research has} indicated that colonisation constraints often hamper restoration, see e.g. (Bakker et al. 1996; Bakker & Berendse 1999). In our study ditch banks with AES, colonisation constraints may also play an important role for determining diversity. Generally, diversity was much higher in the study ditch banks than in the surrounding reference ditch banks (Figs. 1 a,b). We infer that source populations are absent from the (immediate) surroundings of the study ditch banks, which indicates that dispersal constraints probably also hamper an increase in species diversity.

Management implications

Productivity often increased and diversity was maintained or decreased in ditch banks with agri-environment schemes, both in absolute terms and in relation to the reference. However, patterns were not consistent between areas; occasionally, ditch banks with AES did show a decrease in productivity and/or an increase in diversity. Since productivity and diversity were clearly negatively related both in space and time, AES in ditch banks should continue to focus on nutrient reduction. In most cases, more frequent mowing would probably be most appropriate. However, more research is required into the interactions between (high) starting productivity and mowing regime.

Moreover, as indicated by the fact that most study ditch banks started out with higher diversity and lower productivity than their reference ditch banks, a bias seems to exist in relation to the farmers that join agri-environment programmes. If this is the rule, rather than the exception, this will have implications for the goals of future programmes. It may be necessary, and perhaps even desirable, not only to focus on number of (target) species, but also on the expected changes in time. On farms with high diversity, the focus should be on the maintenance of diversity (‘refuge management’). However, more effort should also be put into recruiting farms with low diversity with an emphasis on long-term increase in diversity (‘improvement management’). After all, with source populations in the surrounding there is a real chance of improvement once productivity has been reduced - provided that the continuity of the schemes can be guaranteed for a longer period of time.

Acknowledgements

(21)

Chapter 4 88

Ministry of Agriculture, Nature and Food Quality; Lourien van de Hoek and Annemiek Kooij) and the agricultural nature and landscape associations Den Hâneker (Nell Brouwer) and Weidehof (Liesbeth van der Heuvel). The regional waterboards (Hoogheemraadschap Alblasserwaard en Vijfheerenlanden, De Stichtse Rijnlanden and Krimpenerwaard) provided information on changes in water tables.

Reference list

Aarssen, L. A. 2001. On correlations and causations between productivity and species richness in vegetation: predictions from habitat attributes. Basic and Applied Ecology 2: 105-114. Al-Mufti, M. M., Sydes, C. L., Furness, S. B., Grime, J. P. & Band, S. R. 1977. A quantitative

analysis of shoot phenology and dominance in herbaceous vegetation. J. Ecol. 65: 759-791.

Bakker, J. P. 1989. Nature management by grazing and cutting. Kluwer Publishing, Dordrecht, NL.

Bakker, J. P. & Berendse, F. 1999. Constraints in the restoration of ecological diversity in grassland and heathland communities. Trends. Ecol. Evol. 14: 63-68.

Bakker, J. P., Poschlod, P., Strykstra, R. J., Bekker, R. M. & Thompson, K. 1996. Seed banks and seed dispersal: important topics in restoration ecology. Acta Bot. Neerl. 45: 461-490. Beintema, A. J. & Müskens, G. J. D. M. 1987. Nesting success of birds breeding in Dutch

agricultural graslands. J. Appl. Ecol. 24: 743-758.

Blomqvist, M. M., Bekker, R. M. & Vos, P. 2003a. The potential for restoration of plant species richness of grassland ditch banks from the soil seed bank . Appl. Veg. Sci. 6: 179-188. Blomqvist, M. M., Vos, P., Klinkhamer, P. G. L. & ter Keurs, W. J. 2003b. Declining plant species

richness of grassland ditch banks - a problem of colonisation or extinction? Biol. Conserv. 109: 391-406.

Bosy, J. L. & Reader, R. J. 1995. Mechanisms underlying the suppression of forb seedling emergence by grass (Poa pratensis) litter. Funct. Ecol. 9: 635-639.

Cameron, A. 2001. Remonitoring of agri-environment schemes in Northern Ireland. In: Bunce, R. G. H., Pérez-Soba, M., Elbersen, B. S., Prados, M. J., Andersen, E., & Smeets, P. J. A. M. (eds.) Examples of European agri-environment schemes and livestock systems and their influence on Spanish cultural landscapes, pp. 48-51. Alterra, Green World Research, Wageningen, NL.

Carey, P. D. 2001. Schemes are monitored and effective in the UK. Nature 414: 687.

Clausman, P. H. M. A. & van Wijngaarden, W. 1984. Het vegetatie onderzoek van de provincie Zuid-Holland: Verspreiding en ecologie van wilde planten in Zuid-Holland. Ia waarderingsparameters, 64 pp. Provinciale Planologische Dienst Zuid-Holland, Den Haag, NL.

Critchley, C. N. R., Burke, M. J. W. & Stevens, D. P. 2003. Conservation of lowland semi-natural grasslands in the UK: a review of botanical monitoring results from agri-environment schemes. Biol. Conserv. 115: 263-278.

DLG 2000. Subsidieregeling agrarisch natuurbeheer. Dienst Landelijk Gebied, Utrecht, NL. Ellenberg, H., Weber, H. E., Düll, R., Wirth, V., Werner, W. & Paulissen, D. 1992. Zeigerwerte

von Pflanzen in Mitteleuropa. Scripta Geobotanica 18: 1-258.

Ertsen, A. C. D., Alkemade, J. R. M. & Wassen, M. J. 1998. Calibrating Ellenberg indicator values for moisture, acidity, nutrient availability and salinity in the Netherlands. Plant Ecol. 135: 113-124.

Grace, J. B. 1999. The factors controlling species density in herbaceous plant communities: an assessment. Perspectives in Plant Ecology, Evolution and Systematics 2: 1-28.

(22)

Hill, M. O. & Carey, P. D. 1997. Prediction of yield in the Rothampstead Park Grass Experiment by Ellenberg indicator values. J. Veg. Sci. 8: 579-586.

Hill, M. O., Mountford, J. O., Roy, D. B. & Bunce, R. G. H. 1999. Ellenberg's indicator values for British plants. Technical Annex, Institute of Terrestrial Ecology / Department of Environment, Transport and Regions (NERC), Grange-over-Sands, UK.

Kleijn, D., Berendse, F., Smit, R. & Gillisen, N. 2001. Agri-environment schemes do not effectively protect biodiversity in Dutch agricultural landscapes. Nature 413: 723-725. Kleijn, D. & Sutherland, W. J. 2003. How effective are European agri-environment schemes in

conserving and promoting biodiversity? J. Appl. Ecol. 40: 947-969.

Kruk, M., Twisk, W., de Graaf, H. J. & ter Keurs, W. J. 1994. An experiment with paying for conservation results concerning ditch bank vegetations in the western peat district in the Netherlands. BCPC Monographs no. 58: Field margins: integrating agriculuture and conservation, pp. 397-402.

Kull, O. & Aan, A. 1997. The relative share of graminoid and forb life-forms in a natural gradient of herb layer productivity. Ecography 20: 146-154.

LNV 1995. Regeling beheersovereenkomsten en natuurontwikkeling, 41 pp. Ministerie LNV (Ministry of Agriculture, Nature Management and Fisheries), Landinrichting en beheer landbouwgronden, Utrecht, NL.

Melman, C. P., Clausman, P. H. M. A. & Udo de Haes, H. A. 1988a. The testing of three indicator systems for trophic state in grasslands. Vegetatio 75: 143-152.

Melman, D. 1991. Slootkanten in het veenweidegebied: mogelijkheden voor behoud en ontwikkeling van natuur in agrarisch grasland. PhD thesis, Leiden University, Leiden, NL (with English summary).

Melman, D. & van Strien, A. 1993. Ditch banks as a conservation focus in intensively exploited peat farmland. In: Vos, C. & Opdam, P. (eds.) Landscape ecology of a stressed environment, pp. 122-141. Chapman & Hall, London, UK.

Melman, P. J. M., Verkaar, H. J. & Heemsbergen, H. 1988b. Species diversity of road verge vegetation and mowing regime in the Netherlands. In: During, H. J., Werger, M. J. A. & Willems, J. H. (eds.) Diversity and pattern in plant communities, pp. 165-170. SPB Academic Publishing, Den Haag, NL.

Melman, Th. C. P., Udo de Haes, H. A. & van Wijngaarden, W. 1991. Size dependence of parameters for ecological factors and for nature conservation evaluation of grassland relevés. Biol. Conserv. 55: 347-354.

Moore, D. R. J. & Keddy, P. A. 1989. The relationship between species richness and standing crop in wetlands: the importance of scale. Vegetatio 79: 99-106.

Mountford, J. O., Lakhani, K. H. & Kirkham, F. W. 1993. Experimental assessment of the effects of nitrogen addition under hay-cutting and aftermath grazing on the vegetation of meadows on a Somerset peat moor. J. Appl. Ecol. 30: 321-332.

Musters, C. J. M., Kruk, M., de Graaf, H. J. & ter Keurs, W. J. 2001. Breeding birds as a farm product. Conserv. Biol. 15: 363-369.

Olff, H. & Bakker, J. P. 1991. Long-term dynamics of standing crop and species composition after the cessation of fertilizer application to mown grassland. J. Appl. Ecol. 28: 1040-1052. Oomes, M. J. M. 1992. Yield and species density of grasslands during restoration management. J.

Veg. Sci. 3: 271-274.

Oomes, M. J. M. & van der Werf, A. 1996. Restoration of species diversity in grasslands: the effect of grassland management and changes in ground water level. Acta Botanica Gallica 143: 451-461.

Oomes, T. & van der Werf, A. 2003. Management of species rich hay pastures: must the application of fertilizer always be rejected? De Levende Natuur 104: 192-196.

Ovenden, G. N., Swash, A. R. H. & Smallshire, D. 1998. Agri-environmental schemes and their contribution to the conservation of biodiversity in England. J. Appl. Ecol. 35: 955-960. Provincie Zuid-Holland 1993. De vegetatie van Zuid-Holland 1976-1991. De wilde plantengroei

(23)

Chapter 4 90

Onderzoeksrapport en Kaartenbijlage. Provincie Zuid-Holland, Dienst Ruimte en Groen, Den Haag, NL.

Ryser, P., Lagenhauer, R. & Gigon, A. 1995. Species richness and vegetation structure in a limestone grassland after 15 years management with six biomass removal regimes. Folia Geobot. Phytotax. 30: 157-167.

Schaffers, A. P. & Sykora, K. V. 2000. Reliability of Ellenberg indicator values for moisture, nitrogen and soil reaction: a comparison with field measurements. J. Veg. Sci. 11: 225-244.

Silvertown, J. W. 1980. The dynamics of a grassland ecosystem: botanical equilibrium in the park grass experiment. J. Appl. Ecol. 17: 491-504.

Tallowin, J. R. B. 1996. Effects of inorganic fertilizers on flower-rich hay meadows: a review using a case study on the Somerset Levels, UK. Grassland and Forage Abstracts 66: 147-152.

ter Heerdt, G. N. J., Bakker, J. P. & de Leeuw, J. 1991. Seasonal and spatial variation in living and dead plant material in a grazed grassland as related to plant species diversity. J. Appl. Ecol. 28: 120-127.

van der Linden, M. & de Jong, F. M. W. 1994. Inrichting en beheer van slootkanten in het veenweidegebied. CML report 106, Leiden, NL.

van der Meijden, R. 1996. Heukels' Flora van Nederland. Wolters-Noordhoff bv., Groningen, NL. van Harmelen, W., van Heerden, A. & Kruk, M. 1997. Waardevolle oeverplanten voor het

agrarisch natuurbeheer: een hulpmiddel bij de herkenning. Provincie Zuid-Holland, Den Hâneker, Dienst Landelijk Gebied, Voorburg, NL.

van Strien, A. 1991. Maintenance of plant species diversity on dairy farms. PhD thesis, Leiden University, Leiden, NL.

Vermeer, J. G. & Berendse, F. 1983. The relationship between nutrient availability, shoot biomass and species richness in grassland and wetland communities. Vegetatio 53: 121-126. Westhoff, V. & Weeda, E. 1984. De achteruitgang van de Nederlandse flora sinds het begin van

deze eeuw. Natuur en Milieu 6: 8-17.

Wheeler, B. D. & Shaw, S. C. 1991. Above-ground crop mass and species richness of the principal types of herbaceous rich-fen vegetation of lowland England and Wales. J. Ecol. 79: 285-301.

Wiertz, J. 1992. Schatting van ontbrekende vocht- en stikstofindicatiegetallen van Ellenberg (1979). Instituut voor Bos- en Natuuronderzoek (IBN-DLO), Wageningen, NL.

Willems, J. H. & Nieuwstadt, M. G. L. 1996. Long-term after effects of fertilization on above-ground phytomass and species diversity in calcareous grassland. J. Veg. Sci. 7: 177-184. Willems, J. H., Peet, R. K. & Bik, L. 1993. Changes in chalk-grassland structure and species

richness resulting from selective nutrient additions. J. Veg. Sci. 4: 203-212.

(24)

Appendix 1. Most biomass samples were collected in May, before grazing commenced.

To get an impression of the development in time, we also measured biomass in July (peak values) and September / October (total cumulative values) before the ditch banks were mown in 2002 and 2003: mean, minimum and maximum values for 12 ditch banks excluded from grazing are indicated.

2002 2003

Biomass May July Sept / Oct May July Sept / Oct

Mean 195.1 644.0 918.3 125.8 717.1 900.5

Minimum 78.4 431.2 626.2 75.7 340.9 533.0

(25)