Translator Disclaimer
2 February 2021 Effect of Buffer Zone Structure on Diversity of Aquatic Vegetation in Farmland Water Bodies
Author Affiliations +
Abstract

Farmland water bodies (FWBs) are marginal habitats in the agricultural landscape. However, regardless of their small size, they are refuges for natural vegetation and species-rich animal assemblages. They are especially important in areas where the intensification of agricultural activities reduces the ecological quality of the landscape. The study aimed to evaluate the effect of the habitat structure of the FWB buffer zone on macrophyte association diversity in the FWBs (n = 49). As many as 175 species of vascular plants, three stoneworts, and 40 (41–49 according to Chao2 estimator) plant associations were recognized, including 19 of high conservation priority. The occurrence of threatened associations (0–5 per FWB) was not correlated with the number of common (unthreatened) ones. The most important predictors of plant diversity (at the species level and the association level) were the connection of the FWB with a watercourse (positive effect), and the distance from the FWB to the nearest built-up areas (positive effect). The results suggest that even though the high percentage cover of perennial vegetation mitigates the effect of diffused pollutions of agricultural origin on FWB biota, its significance may become low when the distance from a built-up area to FWBs is small.

INTRODUCTION

Due to the intensification of farming practices which among others results in changes in land management (landscape uniformization), agriculture is considered one of the main threats responsible for the impoverishment of biodiversity (Ryszkowski et al. 1998, Boix et al. 2012). Studies on the ecological effects of intensive agriculture have proven that living resources in rural areas depend strongly on the landscape structure. It is commonly accepted that diverse, heterogenic landscape structure is a key factor for preserving high biodiversity in farmland (Benton et al. 2003) as well as maintaining its ecosystem services of desired quantity and quality (Tscharntke et al. 2005).

Besides tree lines or strips of other semi-natural vegetation, which enhance farmland biodiversity level most markedly (Marshall et al. 2002, Wuczyński et al. 2014), also small water bodies are valuable natural or semi-natural elements of the landscape. The role of such water bodies in regulation of hydrological regime, nutrients cycles, and their supporting function for ecosystem services have been strongly emphasized by many authors (Céréghino et al. 2014, Novikmec et al. 2016), and presumably the most essential among their many ecological functions is the enrichment of biodiversity (Declerck et al. 2006, Gioria et al. 2010, Lewis-Phillips et al. 2019).

Due to their small size, the farmland water bodies are highly sensitive to natural and anthropogenic disturbances which can reduce their ecological role. However, this attribute is a driver of spatial and temporal variability which is a characteristic feature of farmland water bodies. The variability results in developing numerous, diverse and in most cases impoverished phytocoenoses. The zonal vegetation pattern, typically found in eutrophic lakes, is only rarely observed in such small water bodies. In most cases, the plant communities form a mosaic of aquatic and rush plant associations that appear in the form of interpenetrating irregular patches (Gołdyn 2010). Despite the above-mentioned sensitivity to some disturbances, such water and wetland ecosystems belong to the most species-rich, natural elements of the landscapes that have been altered as a result of long-term human activities, including the agricultural ones. Among the freshwater habitats, the number of species observed per unit area in such water bodies equals or even exceeds that in lakes or rivers (Williams et al. 2003). Therefore, despite their small size, water bodies are broadly recognized as landscape elements that significantly contribute to the local and regional aquatic biodiversity (Williams et al. 2003, Nicolet et al. 2004, Scheffer et al. 2006, Bilton et al. 2006, Downing 2010) and often harbor many rare or endangered species (Oertli et al. 2002, Gołdyn et al. 2005, Davies et al. 2008).

Freshwater macrophyte species diversity depends on many local factors, such as the physicochemical quality of water, water body area or disturbances (Bornette and Puijalon 2011). The most important threat to macrophyte diversity is agricultural intensification. As a result of intensive soil cultivation and increase of mineral and organic fertilization, significant amounts of nutrients are not fully used in crop production, and a surplus of fertilizers moves to the surface waters and increases their trophic state and accelerates eutrophication (Gamrat et al. 2012, Steffen et al. 2013). Nutrients are taken up by the organisms developing in water bodies and retained in elements of the water body ecosystems (Gałczyńska and Kot 2010). This process leads to the shallowing of the water bodies during the process of deposition of organic matter and in consequence their gradual overgrowing. Another negative driver of changes in macrophyte diversity is urbanization of catchment, which: (a) increases nutrient and sediment loadings causing or accelerating the eutrophication rate, (b) is a source of contaminants, and (c) favours increase in human disturbance, e.g. recreational activity. Sometimes, the water bodies are used as waste stabilization facilities to treat domestic wastewater for irrigation of public spaces (e.g. parks) or crop irrigation (Gruchlik et al. 2018). Currently, ecosystems of small water bodies are listed among those most vulnerable to the effects of anthropogenic landscape changes associated with intensive farming. In Poland, the number of natural small water bodies is constantly decreasing, which reflects worldwide trends (Boix et al. 2012).

A properly managed agricultural landscape should consist of diverse habitat patches, which are changeable in time and space (cultivated fields), and stable, non-farmed, fragmented semi-natural habitats. These include, among others, various farmland shelterbelts, meadows, small water bodies, and wetlands. Such a diversified structure of the agricultural landscape can mitigate many unfavourable effects associated with modern farming practices, contributing to high biodiversity (Kędziora et al. 2012). The above-mentioned impact of nonpoint source pollution is a threat to the quality of water in agricultural areas. Diffused contamination in the agricultural landscape can be effectively mitigated by the above-mentioned non-farmed, marginal habitats that serve as biogeochemical barriers (Ryszkowski and Kędziora 2007, Gerke et al. 2010). The efficiency of this process depends primarily on the type, location, and density of barriers in the catchment area.

The aim of the study was to recognize how the hydrophyte and helophyte plant diversity in small water bodies located in farmland (for simplicity hereinafter referred to as FWBs) is related to their connection with watercourses, and to the land-use in FWB buffer zone. We assumed that the higher proportion of biological filters in the buffer zone, the more diverse plant communities in the FWBs. The mechanism underlying these relationships is that the probability of contamination of water bodies with diffused or point source pollutions, which exceeds the resistance of some plant species, is low when the proportion of biogeochemical barriers in buffer zone is high. As a result, a more diverse buffer zone around FWB would increase the number of rare and/or vulnerable plant communities.

MATERIAL AND METHODS

Study area

The study was conducted in the General Dezydery Chłapowski Landscape Park (Western Poland, 52°03′42.8″N, 16°49′29.5″E). The Park is a part of the ground moraine formed during the Leszczyński Baltic glaciation, characterized by a relatively poor relief (Kondracki 2011). The bedrock consists of light drifted clay on which a layer of light and slightly clayey/clay sands is deposited (up to a depth of 30–80 cm). Clay sands are here initial podzolic soils which are the autogenous soils (Marcinek et al. 2011), a dominant type in the Park area.

Their characteristic feature is the elution of carbonates into deeper parts of the soil profile and the vertical mixing of clay minerals and hydroxides of iron and aluminum (Bartoszewicz 1998).

Agriculture is the dominant activity in the area (73.5% of the area) which has been under intense human pressure for many centuries. The settlement density amounts to 20 per 100 km2 and is higher than mean density in Poland by 28%. All the terrestrial ecosystems are anthropogenic. The environmental impact of agriculture in this region (Wielkopolska) is strong, as reflected, for example, by the usage of mineral fertilizers and yields. In 2007, the mineral nitrogen doses amounted to about 90 kg ha-1, compared to 80–120 kg ha-1 in Ireland, Belgium, Netherland in Germany (Eurostat 2020).The yield of common wheat, cereals for grain, and winter oilseed rape (averaged for 2005–2007) in the region was higher than in other regions of Poland (averaged) by about 10–11% (Eurostat 2020 and Wielkopolska Provincial Office 2020).

However, the area is characterized by a mosaic, diverse structure of landscape due to the presence of various shelterbelts, small forest islands, water bodies and wetlands. Lakes, peat bogs, and other water bodies and watercourses are of natural (post-glacial) or anthropogenic origin. The density of FWBs is 2.4 per km2 and they cover about 0.5% of the total area of the Park (Juszczak and Arczyńska-Chudy 2003). A total of 49 water bodies (Fig. 1) that keep water during the whole vegetation period (even in the years with very low precipitation), were selected for the study. The area of a single FWB usually does not exceed 2 hectares and the mean depth is mostly less than 1 m. All the studied FWBs were located within an agricultural landscape or within rural settlements.

Fig. 1.

Localisation of 49 water bodies in the General Dezydery Chłapowski Landscape Park.

img-z3-7_263.jpg

In the context of the study aims, the area offers favourable spatial landscape arrangement for studying the role of landscape structure in the mitigation of the impact of diffused agricultural pollutions, as well as point anthropogenic impacts related to settlements and roads, on the plant diversity in small FWBs.

Survey of plant species and plant association composition

The effect of habitat structure on plant diversity was analysed at the species- and association level. The latter was performed for two reasons. Firstly, due to a classification of plant communities according to the degree of their threat (Brzeg and Wojterska 2001), it is possible to understand the importance of environmental threats better then recognized with the use of species-level only. Secondly, the number of plant associations is a proxy for spatial heterogeneity of water bodies.

The field surveys were carried out twice a year – in April and throughout July and August in the years 2005–2007. The survey in each water body was conducted with the use of pontoon, and by walking around the water body. The data on plants originate from three sources: (i) sampling with the use of phytosociological relevés, (ii) viewing the entire surface of the reservoir, and (iii) sampling of submerged vegetation using the anchor. The plant associations were distinguished based on the 304 phytosociological relevés that were taken according to the Braun-Blanquet method (Fukarek 1967, Matuszkiewicz 2005). The relevés were taken in each of visually distinguished homogeneous vegetation patches. The number of relevés ranged from 2 to 20 per water body, and the size of a single relevé amounted to 4–30 m2. The wider and/or larger vegetation patch, the larger relevé size. The lists of plant species occurring in the FWBs were complemented with all other species noticed or collected during the field survey. Submerged vegetation was surveyed with the use of 3–5 samples per FWB per visit. The larger FWB and/or less visible submerged plants, the higher number of samples. Most of the observed or collected specimens of plants were taxonomically determined in the field, in some cases (about 15% of the species) – in the laboratory, with the use of optical equipment.

The plant species, that are bioindicators for high water quality (i.e. for mesotrophic water), were distinguished following Zarzycki et al. (2002).The plant species were grouped into ten group of species (guilds), that prefers one habitat type or have given ecological trait (submerged vs floating leaves): ‘aquatic-submerged', ‘aquatic-floating', ‘marsh-rush', ‘meadow-marsh', ‘meadow', ‘meadow-fallow', ‘sporadically flooded areas', ‘ruderal-fringe', ‘crop field’, and ‘forest' species (by Matuszkiewicz 2005).The hydrophyte (‘aquatic-submerge’ and ‘aquatic-floating') and helophyte (‘marsh-rush') species were considered typical species to FWBs and were the subject of the analyses of the relationships between habitat structure and plant diversity.

The plant species associations were distinguished and classified regarding their threat level and frequency of occurrence based on the classification by Brzeg and Wojterska (2001). The associations were divided into three groups: threatened, unthreatened and bioindicators of high water quality. Species names of vascular plants are used according to Mirek et al. (2002) and for stoneworts, the nomenclature by Dąmbska (1964) is accepted.

Habitat structure in buffer zones

The measurements of buffer zone habitat structure were carried out using QGIS software. The following materials were used for analysis: the service of viewing (WMS) hydrological data for the area of Poland in the reference system 1992 and ortophotomaps provided by Geoportal 2 (size of the water body) and Google Earth (aerial photographs).

As earlier studies show, the effects of land use is strongest at the local scale (Declerck et al. 2006), therefore the buffer zone habitat structure was quantified within a distance of 100 meters from FWB shoreline. A total of six variables (predictors of number of species and number of associations) were used to quantify the FWB and buffer features (Table 1):

  1. The area (ha) of the FWB (FWB_area), measured with the aid of Google maps. The FWB border line was identified as outer water surface edge and/or outer rush vegetation edges.

  2. The connection of the FWB with watercourses (FWB_conn), categorized as absent (0) or present (1).

  3. The relative proportion of biogeochemical barriers/filters (Filter_rel) is the proportion of the total area of biogeochemical barriers that can filter surface and ground water, i.e. wooded area, grasslands, wetlands, hedgerows and water bodies, in relation to the total area of the analysed sample of FWB buffer zone, divided by WB area. We used the relative proportion, as the efficiency of biogeochemical barriers that cover e.g. 10% of buffer zone is expected to be different for the FWB with an area of 0.1 ha, and for the FWB with an area of 1.0 ha. The area of the biogeochemical barrier was measured with the use of Google Earth.

  4. The diversity of biogeochemical barriers (Filter_div), i.e. Shannon diversity index H', estimated with the use of the proportion of biogeochemical barrier types listed above in relation to the total area of the biogeochemical barrier.

  5. The distance (in meters) from the FWB border line to the nearest built-up area, i.e. a place with buildings (Built_dist) was measured with the aid of Google Earth. The measurement was not limited to 100 meters buffer zone because the distance from FWB to the nearest built-up area was larger than 100 meters in 26 cases.

  6. The presence of a paved or unpaved road (Road), categorized as absent (0) or present (1).

Table 1.

Type and summary statistics of habitat and plant variables. FWB_area – area of farmland water body, FWB_conn – connection of the FWB with watercourses, Filter_div – diversity of biogeochemical barriers, Filter_rel – relative proportion of biogeochemical barriers, Built_dist – distance (m) from the FWB to the nearest built-up area, N_spp_tot – total number of plant species, N_spp_ind – number of bioindicatory species, N_ass_tot – total number of associations, N_ass_thr – number of threatened or rare associations, N_ass_ ind – number of bioindicator associations.

img-z5-2_263.gif

Statistical analyses

The species richness and association richness were assessed with the aid of species richness estimator Chao2-bc (bias corrected) (Chao et al. 2016). The relationships between the numbers of common and rare communities were tested with the use of Gamma rank correlation which is recommended when a given variable includes many values with the same rank. The above analyses were done with the aid of R package ‘SpadeR’ (Chao et al. 2016), and Statistica 12 (Statsoft 2014).

Relationships between habitat variables (predictors) and plant variables were analysed with the use of the Multi-Model Inference package (MuMIn) (Bartoń 2016) in the R environment (R Core Team 2016). In the analysis of relationships between habitat and plant associations, all numeric variables were standardized to the variables with means equal to zero, and SD equal to one. In the case of the number of plant species, or number of associations, the assumption was accepted about Poisson or negative binomial error frequency distribution of explained variables with the log link function. Negative binomial frequency distribution was used for over-dispersed data, based on the overdispersion test (Cameron and Trivendi 1990), run with the aid of R package AER (Kleiber and Zeileis 2008).

Best model subsets were selected with a corrected Akaike Information Criterion (AICc). The best model subset contained the models for which the difference in AIC relative to AICmin is smaller than 2 (ΔAICc<2). When the best model subset contained more than one model, they were averaged using Akaike weights. The coefficients of the variables in best models were averaged unconditionally, i.e. irrespective of their presence or absence in each model. When assessing the significance of predictors, factors taken into account included averaged coefficients, their 95% confidence interval, statistical significance (P), importance (relative measure of predictors estimated by the MuMIn), and the number of models (in the best subset) containing given predictor. As the numeric variables used in the models were standardized, variable coefficients can be considered as a measure of the effect size.

RESULTS

Species richness and species guilds

As many as 175 species of vascular plants and three species of stoneworts were found in the studied FWBs, including species rare in Wielkopolska Region (Chara fragilis, Chara vulgaris, Utricularia vulgaris) and partially protected by law in Poland: Batrachium trichophyllum, Chara hispida, Menyanthes trifoliata, Nymphaea alba and Ranunculus linqua. Alien species were represented by Canadian waterweed (Elodea canadensis) and sweet flag (Acorus calamus). Species richness (Chao2-bc) amounted to 232, with 95% CI 206–285. The highest number of species per FWB amounted to 75. As it was outlier data, it was excluded from further analyses of total plant species number. Thus, the number of species per FWB varied from 9 to 49, the mean amounted to 21.1 ± 7.3 (SD), and the median was equal to 20.0.

The plant species that prefer meadows, marshes or rushes made up almost 60% of species (Appendix 1, Table 2). As many as 72 species of hydrophytes and helophytes were found (Appendix 1). The number of these species per FWB ranged from 4 to 25, the mean amounted to 11.9 ± 4.3 (SD), and median was equal to 12.0.

Table 2.

Number of species and domination (%) of plant species guilds (criterion: preferred habitat type, by Matuszkiewicz 2005).

img-ABcE_263.gif

Twenty species (Appendix 1) were bioindicators for mesotrophic water bodies, i.e. for high water quality. Their number per FWB amounted to 0–7, with mean 1.4 ± 1.8 (SD), and median 1.0.

Plant association diversity

There were 40 plant associations distinguished in the FWBs, and the Chao2's value amounted to 41 (95% confidence interval is 40 to 49). As many as 12 associations occurred only in 1–2 water bodies (Table 3). More than half of the identified syntaxa (21) belonged to rush associations from Phragmitetea class and about one quarter (11) to water plant associations from Potametea class (Table 3). Submerged vegetation and plants with floating leaves were represented by stonewort associations from Charetea class (3), and macrophyte associations from Lemnetea (5) and Potametea class (11). Among the seven most frequently observed associations (that is in ten or more FWBs), five belonged to Phragmitetea class, with most abundant Phragmitetum and Typhetum latifoliae associations, and two other (Ceratophylletum demersi and Lemnetum minorisone) belonged to other classes (Table 3). The number of plant associations per FWB amounted to 1–10, mean was equal to 4.6 ± 2.1 (SD), and median amounted to 4.0.

The most valuable with respect to their conservation status were 19 associations considered as ‘V’ or ‘I’, and 13 associations were recognized as bioindicators of high water quality (Table 3). The number of threatened associations ranged from 0 to 5 per FWB, mean amounted to 1 ± 1.3 (SD), and median – 1.0. The occurrence of threatened associations was not correlated with the number of common (unthreatened) ones.

Weak relationships between both numbers are reflected by the small and statistically significant value of correlation coefficient (Gamma = -0.001, P = 0.99). Also the number of plant associations (0–4 per FWB) that are considered as indicators of high water quality was not a predictor of the number of other associations (Gamma = 0.10, P = 0.45).

Multivariate analysis of the relationships between the habitat structure and the number of plant species

The total number of plant species per FWB was related to five predictors (Appendix 2A), however two of them were most important. In the averaged model, the coefficients, the relative variable importance (RVI), and the statistical significance (P) of the distance from the nearest built-up area (Built_dist), as well as the connection of FWBs with watercourse (FWB_conn), were markedly higher than the other ones, and they affected the total number of plant species positively (Table 4). Both predictors were also present in all the six best models (Appendix 2A). The best models' Goodness-of-Fit was relatively high as reflected by adjusted R2 that ranged from 0.40 to 0.45.

Table 3.

The plant associations found in the study area, their status assessment and occurrence frequency in Wielkopolska region (by Brzeg and Wojterska 2001) and in the study area. (Frequency in the study area is expressed as a number of water bodies inhabited). Bioindicators of high water quality (by Szoszkiewicz et al. 2010) are marked in bold.

img-z8-2_263.gif

Continued

img-z9-11_263.gif

The distance from the built-up area and relative proportion of biological filters were most important predictors of the number of bioindicator plant species (Table 5), however the statistical significance of both predictors was very low (P = 0.59 and P = 0.82, respectively), and the Goodness-of-Fit of the models was low (0.00–0.04) (Appendix 2B). That means that the number of these species was affected by some other factors, not considered here.

Table 4.

Averaged best models (see Appendix 2A) of the relationships between water body and habitat structure in the buffer zone and the total number of helophyte and hydrophyte species. CI-Min and CI-Max – 95% confidence interval, P – significance, RVI – relative variable importance. FWB – farmland water body, Built_dist – distance (m) from the FWB to the nearest built-up area, FWB_conn – connection of the FWB with watercourses, FWB_area – area of farmland water body, Filter_rel – relative proportion of biogeochemical barriers. For the details – see section ‘Statistical analyses'.

img-z9-20_263.gif

In the case of the total number of plant associations, five best models were selected, built of four variables. Their adjusted R2 values ranged between 0.16 and 0.23 (Appendix 2C). Based on the averaged model (Table 6), it can be interpreted that the connection of a FWB with watercourse (FWB_conn) and the distance from the nearest built-up area (Built_dist) were the most important predictors of the total number of plants. They were components of five and three (respectively) of five best models, their RVI and coefficient were highest, and their statistical significance, too. However the statistical significance of Built_dist was much weaker than 0.05 (P = 0.34) (Table 6).

Table 5.

Averaged best models (see Appendix 2B) of the relationships between water body and habitat structure in the buffer zone and the number of plant species, that are bioindicators of high water quality. CI-Min and CI-Max – 95% confidence interval, P – significance, RVI – relative variable importance. Built_dist – distance (m) from the FWB to the nearest built-up area, Filter_rel – relative proportion of biogeochemical barriers. For the details – see section ‘Statistical analyses'.

img-z10-2_263.gif

Table 6.

Averaged best models (see Appendix 2C) of the relationships between water body and habitat structure in the buffer zone and the number of all plant associations. CI-Min and CI-Max – 95% confidence interval, P – significance, RVI – relative variable importance. FWB – farmland water body, FWB_conn – connection of the FWB with watercourses, Built_dist – distance (m) from the FWB to the nearest built-up area, FWB_area – area of farmland water body. For the details – see section ‘Statistical analyses'.

img-z10-5_263.gif

The variability in the number of threatened associations in relation to FWB and buffer zone structure was best described by three models built of four predictors (Appendix 2D). The adjusted R2 ranged between 0.27 and 0.32. Averaging the best models (Table 7) clearly indicates that the distance from the nearest built-up area (Built_dist) was the most important predictor, that affected the number of threatened associations positively. It was component of all the best models, and its significance (P < 0.001), relative importance (RVI) and coefficient value was the highest among the predictors.

Table 7.

Averaged best models (see Appendix 2D) of the relationships between water body and habitat structure in the buffer zone and the number of threatened plant associations. CI-Min, CI-Max – 95% confidence interval, P – significance, RVI – relative variable importance. Built_dist – distance (m) from the FWB to the nearest built-up area, Filter_rel – relative proportion of biogeochemical barriers. For the details – see section ‘Statistical analyses'.

img-z10-12_263.gif

The dominance of Built_dist also was clearly expressed in the variability of the number of associations, that are indicators of high water quality (Appendix 2E). This variable was the only one which was present in all four best models. Adjusted R2 of the best models ranged from 0.10 to 0.12. In the averaged model, Built_dist had the highest coefficient value, RVI and statistical significance (Table 8).

Table 8.

Averaged best models (see Appendix 2E) of the relationships between water body and habitat structure in the buffer zone and the number of plant associations which are bioindicators of high water quality. CI-Min and CI-Max – 95% confidence interval, P – significance, RVI – relative variable importance. FWB – farmland water body, Built_dist – distance (m) from the FWB to the nearest built-up area, FWB_conn – connection of the FWB with watercourses, Filter_rel – relative proportion of biogeochemical barriers. For the details – see section ‘Statistical analyses'.

img-z11-2_263.gif

DISCUSSION

The total marsh-aquatic plant species richness (gamma diversity) in the studied FWBs (a total of 72 species) was similar to the species richness in other regions of Europe. As many as 60 species occurred in Slovakia (Bubíková and Hrivnák 2018), 47 species in Norway (Edvardsen and Økland 2006), 70 species in Japan (Usio et al. 2017), and Svitok et al. (2018) found 127 species in Slovakia. The number of species usually depends on sampling site, however the number of compared water bodies in most of these surveys was similar (64–69). Svitok et al. (2018) studied much larger sample (n = 92) of diverse (e.g. oxbow, glacial, temporary ponds, gravel, sand pits) water bodies. The alpha diversity of aquatic plant species in small water bodies usually amounts to several species per water body. The mean number of these species per water bodies in studied FWBs was similar to the mean number of species, found by Svitok et al. (2018) (mean 8.1), and Davies et al. (2008) in Denmark, Germany and UK (mean ranged from 7.8 to15.3). I was about three times larger than in other surveys, where it amounted 3.9 (mean) in Slovakia (Bubíková and Hrivnák 2018), 4 (median) in Norway (Edvardsen and Økland 2006), and 3.5 (median) in Japan (Usio et al. 2017).

Considering the results of the analyses carried out at the association level, the study confirms that despite the small area and depth of the FWBs, and their location in a highly-productive agricultural landscape, they can host a large diversity of water and wetland plants, similar (or even higher) to that observed in lakes. For example, 27 plant associations were found in diverse lakes in the Wielkopolski National Park (Pełechaty and Sugier 2002), which is adjacent to our study area, 7–25 – in the phytolitoral of shallow lakes in the Mazurski Landscape Park (Ciecierska 2004), and 28 – in the oxbow lakes of the middle Bug river valley (Lorens 2006). The importance of small FWBs for plant diversity in farmland is strongly highlighted by an unexpectedly large proportion (19/40) of associations recognized as threatened in the Wielkopolska Region (10% of Poland, intensive agriculture). Among these 19 associations there are also those from Charetea class that are rarely found in small water bodies, and have been recorded less and less frequently in Wielkopolska in the last decades (Gąbka 2009). Unfortunately, the comparisons with the use of species associations cannot be performed for other regions in Europe, as this approach is only rarely applied outside of Central European countries.

Despite a relatively large range of water body size (0.01–1.24 ha), this factor weakly affected the diversity (both at species and association level) of plant associations, which is reflected by the absence or low importance of water body area in the averaged models of the habitat-plant diversity relationships (Tables 48). Weak relationships between the water body size and aquatic plant diversity seem to be common in case of small, astatic water bodies. The lack of such relationship was reported also by Oertli et al. (2002), Declerck et al.(2006), and Hassall et al. (2011). Apparently, although habitat area usually is a key environmental factor of taxonomic diversity, its importance in the analysed habitat types is smaller than the importance of other predictors.

The plant species diversity at both considered levels (species and association) was positively affected by the connection of the water body with a watercourse, especially in the total number of aquatic species (Table 4), and the total number of associations (Table 6). This connection could be beneficial to species dispersion along watercourses and presumably improve the water condition due to water circulation, cleaning and oxygenation. Such a feature supposedly also decreases isolation of water bodies which can contribute to a higher taxonomic plant richness in small farmland water bodies (Møller and Rørdam 1985, Bosiacka and Pieńkowski 2012). This is also in line with the study by Akasaka and Takamura (2012), who discovered, that both alpha diversity (number of plant species per water body) and gamma diversity (total number of plant species in a group of water bodies), were higher in the interconnected ponds compared to isolated ones. However, in our study, when threatened associations or bioindicators (both species and associations) of water quality are considered, the significance and the relative importance of the connection of FWBs with watercourses decreases strongly (Table 8) or even this predictor is not the component of the model any more (Tables 5 and 7). Thus, the hydrological connection of the studied FWBs did favour a high aquatic plant diversity, however the mechanism (likely to enhance a more efficient dispersion) worked only with common species and associations.

We also recognized a significant positive effect of the distance to the nearest built-up area. It can be considered a proxy for both direct and indirect impact of inhabitants and settlements on the FWBs. This predictor affected all analysed plant variables, and it was the first or second most important predictor in the averaged models (Tables 48). Their relative importance was especially high for threatened associations and for the communities of bioindicative value.

Contrary to our expectation that the higher proportion of biological filters in the buffer zone, the more diverse plant communities in the FWBs, the predictors representing the amount (Filter_rel) and heterogeneity (Filter_div) of biological barriers were either not the components of the models (for the number of all associations, Table 6) or their significance, effect size, and the relative variable importance were very low (Tables 45 and 78). It suggests that even though the role of habitat diversity in buffer zone in the study area was positive in mitigation of indirect farming impact, it was insufficient for mitigation of the direct impact of humans and settlements. This is a similar conclusion to that of Akasaka et al. (2010), who found that the proportion of urban areas affected macrophyte diversity negatively. Also in the study by Novikmec et al. (2016), the proportion of intensively exploited land (arable land, urban areas) on the catchment scale was positively correlated with the deterioration of physicochemical properties of the ponds, which adversely affected plant diversity. This suggests that the presence of farm infrastructure and housing estates can be the main obstacle for the occurrence of valuable and diverse aquatic and rush plants in FWBs. Only Lemnetum minoris, Typhaetum latifolie and Acoretum calami show resistance to the impact of human pressure in intensively managed rural landscapes.

Usually, the high proportion of biogeochemical barriers favours the overall diversity of plant associations (Declerck et al. 2006, Akasaka et al. 2010, Novikmec et al. 2016), as such linear landscape elements, which act as highly efficient biogeochemical barriers, uptake significant amount of chemical compounds dissolved in surface and ground waters. Apparently, not only buffer zones in the form of trees and shrubs growing next to a water body edge are efficient in improving water quality (Joniak et al. 2017), but tree or shrub belts or lines located at a greater distance from it, too. Due to this function, the linear habitats in the buffer zone can reduce the amount of nutrients from mineral fertilizers used in agriculture that cyclically flow into the water body, and can mitigate the impact of some random incidents of water pollution. The results of our study suggest, however, that in case of a high anthropogenic impact, the efficiency of natural filters is strongly limited, especially when rare and threatened species are considered. It is in concordance with a conclusion from the paper by Novikmec et al. (2016), that effective conservation of farmland water bodies cannot be achieved merely through the management of narrow buffer zones around them but should also involve an appropriate management of the whole catchment. The authors paid attention to the maintenance of a less intensive land-use, that can be considered as a source of diffused pollutions. However our study shows, that the presence of built-up areas, as point sources of pollutions or as a driver of direct human impact on small water bodies, has to be taken into account too, when planning actions for biodiversity protection in these elements of agricultural landscape.

A research of the changes in plant species and plant association diversity of aquatic ecosystems within the General Dezydery Chłapowski Landscape Park in the last 30 years, carried out simultaneously with this study (Gołdyn et al. 2005, Arczyńska-Chudy et al. 2008, Gołdyn and Arczyńska-Chudy 2010), revealed disappearance or decrease of aquatic plant communities, which nowadays are considered to be endangered and for this reason included in the red list. The most significant change is the decline of stoneworts and Nymphaeo alba patches. The main reason for those transformations was associated with the deterioration of water quality caused by a shift to a more intensive agricultural land use, and a change in the management of meadows, including their drainage.

Summary and conclusions

  1. Despite the unfavourable environmental conditions (strong agricultural impact), farmland water bodies in the study area form marginal habitats important for a broad spectrum of aquatic plant associations and species, including those of high conservation priority.

  2. Taking into account that individual farmland water bodies harbour high richness of hydrophyte and helophyte species (11.9 ± 4.3) as well high association richness (4.6 ± 2.1) independent on their area, even very small farmland water bodies deserve protection against damage and pollution.

  3. Among different factors, the connection of the FWB to a watercourse and the distance from the settlements were much more important for preserving the high number of plant species and aquatic plant associations than the cover and diversity of perennial vegetation forming natural filtering buffer and improving water quality. It indicates that the effectiveness of buffer zones around water bodies is limited when the water bodies are located close to settlements. Therefore, in addition to the proper structure of the environment in the buffer zone (i.e. presence of trees and shrubs, meadows or wetlands), some more effective measures aimed at reduction of built-up area impacts on farmland water bodies (e.g. disposing of communal wastes or recreational activities) should be undertaken.

ACKNOWLEDGMENTS:

The authors thank the anonymous reviewers for valuable comments and suggestions, and Marcin Sęk for proofreading the manuscript.

REFERENCES

1.

Akasaka M., Takamura N., Mitsuhashi H., Kadono Y. 2010 – Effects of land use on aquatic macrophyte diversity and water quality of ponds – Freshwater Biol. 55: 909–922. Google Scholar

2.

Akasaka M., Takamura N. 2012 – Hydrologic connection between ponds positively affects macrophyte α and γ diversity but negatively affects β diversity – Ecology, 93: 967–973. Google Scholar

3.

Arczyńska-Chudy E., Gołdyn H., Pińskwar P., Jezierska-Madziar M. 2008 – Phytocenotic diversity of aquatic and rush vegetation of peat pits and its role in agricultural landscape of southern Wielkopolska (Poland) (In: After wise use – The future of peat-lands, Eds: C. Farrell, J. Feehan) – Proceedings of the 13th International Peat Congress, Tullamore, Ireland, pp. 144–146. Google Scholar

4.

Bartoń K. 2016 – MuMIn: multi-model inference. R package. Version 1.15.6. Google Scholar

5.

Bartoszewicz A. 1998 – [Soil cover in General Chlapowski Landscape Park] – Biuletyn Parków Krajobrazowych Wielkopolski, 3: 131–135 (in Polish). Google Scholar

6.

Benton T.G., Vickery J.A., Wilson J.D. 2003 – Farmland biodiversity: is habitat heterogeneity the key? – Trends. Ecol. Evol. 18: 182–188. Google Scholar

7.

Bilton D.T., Mcabendroth L., Bedford A., Ramsay P.M. 2006 – How wide to cast the net? Cross-taxon congruence of species richness, community similarity and indicator taxa in ponds – Freshwater Biol. 51: 578–590. Google Scholar

8.

Boix D., Bigos J., Céréghino R., Hull A.P., Kaletka T., Oertli B. 2012 – Pond research and management in Europe: ‘Small is beautiful’ – Hydrobiologia, 689: 1–9. Google Scholar

9.

Bornette G., Puijalon S. 2011 – Response of aquatic plants to abiotic factors: a review – Aquat. Sci. 73: 1–14. Google Scholar

10.

Bosiacka B., Pieńkowski P. 2012 – Do biogeographic parameters matter? Plant species richness and distribution of macrophytes in relation to area and isolation of ponds in NW polish agricultural landscape – Hydrobiologia, 689: 79–90. Google Scholar

11.

Brzeg A., Wojterska M. 2001– [Plant assosiations of Wielkopolska, state of knowledge and treats] (In: Plants of Wielkopolska and the South Pomeranian Lake District, Ed: M. Wojterska) – Przewodnik Sesji Terenowych, Bogucki Wyd. Nauk. Poznan, pp. 39–110 (in Polish). Google Scholar

12.

Bubíková K., Hrivnák R. 2018 – Artificial ponds in Central Europe do not fall behind the natural ponds in terms of macrophyte diversity – Knowl. Manag. Aquat. Ec. 419: 8. Google Scholar

13.

Cameron A.C., Trivedi P.K. 1990 – Regression-based tests for overdispersion in the Poisson model – J. Econom. 46: 347–364. Google Scholar

14.

Céréghino R., Boix D., Cauchie H.M., Martens K., Oertli B. 2014 – The ecological role of ponds in a changing world – Hydrobiologia, 723: 1–6. Google Scholar

15.

Chao A., Ma K.H., Hsieh T.C., Chiu C. 2016 – SpadeR: Species-richness prediction and diversity estimation with R. – R Package Version 0.1.1., pp. 1–7. Google Scholar

16.

Ciecierska H. 2004 – Phytocenotic diversity of littoral in lakes of the Masurian landscape park, current state and changes – Teka Komisji Ochrony i Kształtowania Środowiska Przyrodniczego, 1: 32–38. Google Scholar

17.

Davies B., Biggs J., Williams P., Whitfield M., Nicolet P., Sear D., Bray S., Maund, S. 2008 – Comparative biodiversity of aquatic habitats in the European agricultural landscape. Agric. Ecosys. Environ. 125: 1–8. Google Scholar

18.

Dąmbska I. 1964 – [Charophyta. Freshwater flora of Poland] – PWN, Warszawa, 126 pp. (in Polish). Google Scholar

19.

Declerck S., De Bie T., Ercken D., Hampel H., Schrijvers S., Van Wichelen J., Keijers S. 2006 – Ecological characteristics of small farmland ponds: associations with land use practices at multiple spatial scales – Biol. Conserv. 131: 523–535. Google Scholar

20.

Downing J. 2010 – Emerging global role of small lakes and ponds: little things mean a lot – Limnetica, 29: 9–24. Google Scholar

21.

Edvardsen A., Økland R. H. 2006 – Variation in plant species richness in and adjacent to 64 ponds in SE Norwegian agricultural landscapes – Aquat. Bot. 85: 79–91. Google Scholar

22.

Eurostat – Crop production in EU standard humidity by NUTS 2 regions [apro_cpshr] (Available 13.08.2020). Google Scholar

23.

Fukarek F. 1967 – [Phytosociology] – PWRiL, Warszawa, 218 pp. (in Polish). Google Scholar

24.

Gałczyńska M., Kot M. 2010 – Influence of anthropopression on concentration of biogenic compounds in small of waterponds in farmland – J. Elementol. 15: 53–63. Google Scholar

25.

Gamrat R., Gałczyńska M., Pacewicz K. 2012 – Spatial analysis of plant species distribution in midfield ponds in an agriculturally intense area – Pol. J. Environ. Stud. 21: 871–877. Google Scholar

26.

Gąbka M. 2009 – Charophytes of the Wielkopolska Region (NW Poland): Distribution, Taxonomy And Autecology – Bogucki Wydawnictwo Naukowe, Poznan, 110 pp. Google Scholar

27.

Gerke H.H., Koszinski S., Kalettka T., Sommer M. 2010 – Structures and hydrologic function of soil landscapes with kettle holes using an integrated hydropedological approach – J. Hydrol. 393: 123–132. Google Scholar

28.

Gioria M., Schaffers A., Bacaro G., Feehan J. 2010 – The conservation value of farmland ponds: predicting water beetle assemblages using vascular plants as a surrogate group – Biol. Conserv. 143: 1125–1133. Google Scholar

29.

Gołdyn H. 2010 – Changes in plant species diversity of aquatic ecosystems in the agricultural landscape in west Poland in the last 30 years – Biodivers. Conserv. 19: 61–80. Google Scholar

30.

Gołdyn H., Arczyńska-Chudy E. 2010 – Changes of plant associations and their succession processes course in peat pits as an effect of peat transformations and the intensification of meadow management (In: Physical, chemical and biological processes in soils, Eds: L.W. Szajdak, A.K. Karabanov) – Institute for Agricultural and Forest Environment P.A.S., Poznań, pp. 549–559. Google Scholar

31.

Gołdyn H., Kujawa A., Arczyńska-Chudy E., Jezierska-Madziar M., Sobczyk D. 2005 – Materials of flora of vascular plants of the general Dezydery Chłapowski Landscape Park – Rocz. Ar. Pozn., Bot.-Stec. 8: 55–71. Google Scholar

32.

Gruchlik Y., Linge K., Joll C. 2018 – Removal of organic micropollutants in waste stabilisation ponds: a review – J. Environ. Manage. 206: 202–214. Google Scholar

33.

Hassall C., Hollinshead J., Hull A. 2011 – Environmental correlates of plant and invertebrate species richness in ponds – Biodivers. Conserv. 20: 3189–3222. Google Scholar

34.

Joniak T., Kuczyńska-Kippen N., Gąbka M. 2017 – Effect of agricultural landscape characteristics on the hydrobiotic structure in small water bodies – Hydrobiologia, 793: 121–133. Google Scholar

35.

Juszczak R., Arczyńska-Chudy E. 2003 – [Ecological and water ditch system valorisation of small water bodies in the agricultural landscape of Wielkopolska] – Post. Nauk. Rol. 1: 145–160 (in Polish). Google Scholar

36.

Kędziora A., Kujawa K., Gołdyn H., Karg J., Bernacki Z., Kujawa A., Bałazy S., Oleszczuk M., Rybacki M., Arczyńska-Chudy E., Tkaczuk C., Łęcki R., Szyszkiewicz-Golis M., Pińskwar P., Sobczyk D., Andrusiak J. 2012 – Impact of land-use and climate on biodiversity in an agricultural landscape (In: Biodiversity enrichment in a diverse world, Ed: G.A. Lameed) – In-tech, pp. 281–336. Google Scholar

37.

Kleiber C., Zeileis A. 2008 – Applied Econometrics with R – New York: Springer-Verlag. URL https://CRAN.R-project.org/package=AER Google Scholar

38.

Kondracki J. 2011 – [Regional geography of Poland] – PWN, 468 pp. (in Polish). Google Scholar

39.

Lewis-Phillips J., Brooks S., Sayer C.D., Mc-Crea R. 2019 – Pond management enhances the local abundance and species richness of farmland bird communities – Agr. Ecosyst. Environ. 273: 130–140. Google Scholar

40.

Lorens B. 2006 – [Plants of river lakes and their phytocenotic and species diversity (In: River lakes of the Middle Bug Valley, Ed: W. Wojciechowska) – Różnorodność Biologiczna i Krajobrazowa, Wydawnictwo KUL, Lublin, pp. 55–94 (in Polish). Google Scholar

41.

Marcinek J., Komisarek J., Bednarek R., Mocek A., Skiba S., Wiatrowska K. 2011 – [Polish Soil Classification] – Soil Sci. Ann. 62: 1–193. Google Scholar

42.

Marshall J., Baudry J., Burel F., Joenje W., Gerowitt B., Paoletti M., Thomas G., Klein D., Le Cœur D., Moonen C. 2002 – Field boundary habitats for wildlife, crop and environmental protection (In: Landscape ecology in agroecosystems management, Ed: L. Ryszkowski) – CRC Press, Boca Raton, pp. 219–247. Google Scholar

43.

Matuszkiewicz W. 2005 – [Manual for the determination of plant communities in Poland] – PWN, Warszawa, 542 pp. (in Polish). Google Scholar

44.

Mirek Z., Piękoś-Mirkowa H., Zając A., Zając M. 2002 – Flowering plants and pteridophytes of Poland. Biodiversity of Poland – W. Szafer Institute of Botany, P.A.S., Kraków 442 pp. Google Scholar

45.

Møller T.R., Rørdam C.P. 1985 – Species numbers of vascular plants in relation to area, isolation and age of ponds in Denmark – Oikos, 45: 8–16. Google Scholar

46.

Nicolet P., Biggs J., Fox G., Hodson M.J., Reynolds C., Whitfield M., Williams P. 2004 – The wetland plant and macro-invertebrate assemblages of temporary ponds in England and Wales – Biol. Conserv. 120: 261–278. Google Scholar

47.

Novikmec M., Hamerlik L., Kočický D., Hrivnák R., Kochjarová J., Ot'ahel'ová H., Pal'ove-Balang P., Svitok M. 2016 – Ponds and their catchments: size relationships and influence of land use across multiple spatial scales – Hydrobiologia, 774: 155–166. Google Scholar

48.

Oertli B., Joye A.D., Castella E., Juge R., Cam-bin D., Lachavanne J.B. 2002 – Does size matter? the relationship between pond area and biodiversity – Biol. Conserv. 104: 59–70. Google Scholar

49.

Pełechaty M., Sugier P. 2002 – [Specificity of macrophyte vegetation of the Wielkopolski and Poleski National Park, state of preservation and problems of protection] – Acta. Agrophys. 67: 237–244 (in Polish, English summary). Google Scholar

50.

R Core Team 2016 – R: A Language and environment for statistical computing – R Foundation For Statistical Computing, Vienna, Austria. Google Scholar

51.

Ryszkowski L., Gołdyn H., Arczyńska-Chudy E. 1998 – Plant diversity in mosaic agricultural landscapes: a case study from Poland – Planta Europa. Proceedings of the Second European Conference on the Conservation of Wild Plants, Upsala, Sweden, pp. 281–286. Google Scholar

52.

Ryszkowski L., Kędziora A. 2007 – Modification of water flows and nitrogen fluxes by shelterbelts – Ecol. Eng. 29: 388–400. Google Scholar

53.

Scheffer M., Van Geest G.J., Zimmer K., Jeppesen E., Søndergaard M., Butler M.G., Hanson M.A., Declerck S., De Meester L. 2006 – Small habitat size and isolation can promote species richness: second order effects on biodiversity in shallow lakes and ponds – Oikos, 112: 227–231. Google Scholar

54.

Steffen K., Becker T., Herr W., Leuschner C. 2013 – Diversity loss in the macrophyte vegetation of northwest German streams and rivers between the 1950s and 2010 – Hydrobiologia, 713: 1–17. Google Scholar

55.

Svitok M., Novikmec M., Hamerlík L., Kochjarová J., Oťaheľová H., Paľove-Balang P., Senko D., Matúšová Z., Bubíková K., Hrivnák R. 2018 – Test of the efficiency of environmental surrogates for the conservation prioritization of ponds based on macrophytes – Ecol. Indic. 95: 606–614. Google Scholar

56.

Szoszkiewicz K., Kayzer D., Staniszewski R., Dawson H. 2010 – Measures of central tendency of aquatic habitat parameters: application to river macrophytes communities – Pol. J. Ecol. 58: 595–608. Google Scholar

57.

Tsharntke T., Klein A.M., Kruess A., Steffan-Dewenter I., Thies C. 2005 – Landscape perspectives on agricultural intensification and biodiversity - ecosystem service management – Ecol. Lett. 8: 857–874. Google Scholar

58.

Usio N., Nakagawa M., Aoki T., Higuchi S., Kadono Y., Akasaka M., Takamura N. 2017 –Effects of land use on trophic states and multi-taxonomic diversity in Japanese farm ponds – Agric. Ecosys. Environ. 247: 205–215. Google Scholar

59.

Wielkopolska Province Office – [Crop yields in Wielkopolska] –  http://www.poznan.uw.gov.pl/system/files/zalaczniki/plony_upraw_rolnych_wielkopolska.pdf(Available 14.08.2020). Google Scholar

60.

Williams P., Whitfield M., Biggs J., Bray S., Fox G., Nicolet P., Sear D. 2003 – Comparative biodiversity of rivers, streams, ditches and ponds in in agricultural landscape in southern England – Biol. Conserv. 115: 329–341. Google Scholar

61.

Wuczyński A., Dajdok Z., Wierzcholska S., Kujawa K. 2014 – Applying red lists to the evaluation of agricultural habitat: regular occurrence of threatened birds, vascular plants, and bryophytes in field margins of Poland – Biodivers. Conser. 23: 999–1017. Google Scholar

62.

Zarzycki K., Trzcińska-Tacik H., Różański W., Szeląg Z., Wołek J., Korzeniak U. 2002 – Ecological indicator values of vascular plants of Poland – W. Szafer Institute of Botany, P.A.S., Kraków, 184 pp. Google Scholar

Appendices

Appendix 1.

Plant species in farmland water bodies, their classification by preferred habitat or by ecological traits (Guild), and their frequency (n). Hydrophytes – “aquatic-submerged” and “aquatic-floating” species, helophytes - “marsh-rush” species. Bioindicators for high water quality (by Zarzycki et al. 2002) are marked bold.

img-z17-2_263.gif

Continued

img-z18-1_263.gif

Continued

img-z19-1_263.gif

Appendix 2.

Best model (generalised linear) subsets of relationships between the number of plant associations and water body features and land-use in the buffer zone. Assumptions: Poisson's (Poisson) or negative binomial (NB) error distribution of explained variable, log link function, criterion for best model selection – AICc. FWB – farmland water body, Built_dist – distance (m) from the FWB to the nearest built-up area, Filter_rel – relative proportion of biogeochemical barriers, FWB_area – area of farmland water body, FWB_conn – connection of the FWB with watercourses.

img-z19-3_263.gif

Continued

img-z20-1_263.gif
Krzysztof Kujawa, Ewa Arczyńska-Chudy, Krzysztof Janku, and Mostefa Mana "Effect of Buffer Zone Structure on Diversity of Aquatic Vegetation in Farmland Water Bodies," Polish Journal of Ecology 68(4), 263-282, (2 February 2021). https://doi.org/10.3161/15052249PJE2020.68.4.001
Received: 1 October 2020; Published: 2 February 2021
JOURNAL ARTICLE
20 PAGES


SHARE
ARTICLE IMPACT
Back to Top