Flash in the pan: how grassland renewal affects reproduction of Northern Lapwings Vanellus vanellus

ABSTRACT

Capsule: Grassland renewal is associated with short-term beneficial effects on density and reproduction of Northern Lapwings Vanellus vanellus followed by long-lasting reduced breeding numbers after sward establishment.

Aims: To assess short- and long-term effects of grassland renewal on the reproduction of Northern Lapwings.

Methods: In the years 1999–2011 we counted breeding Northern Lapwings on 18 reseeded grassland plots in a chronosequence ranging from one year before until 3 years after reseeding. We also monitored nesting dispersion and hatching success in different habitats, particularly tillage, permanent grasslands and reseeded grasslands during sward establishment.

Results: Grassland that underwent reseeding possessed several-fold more Northern Lapwings during the phase of sward establishment than before. In the subsequent years, breeding numbers significantly declined and remained low throughout the study period. In recently reseeded grassland, Northern Lapwings tended to breed in larger colonies. Although habitat-specific differences in hatching success were not significant, we found significantly higher nest predation rates in agriculturally improved grasslands than in tillage and reseeded grasslands during sward establishment.

Conclusions: Due to its long-term adverse effects on density and reproduction of Northern Lapwings, grassland renewal should not be practised in areas with conservation responsibilities for grassland-breeding waders.

Reseeding is a common farming practice that is widely promoted to increase the content and quality of grass plants used for grazing, haymaking and silage and to control agricultural weeds. Like many other factors typical of intensified grassland management, such as fertilizer use, mowing regimes and impacts of drainage, reseeding increased in its frequency of occurrence over recent decades (Conijn et al. 2002, Creighton et al. 2011). In the Netherlands, for instance, only 3% of grasslands were renewed within five years in the 1950s, while today this proportion is 23.5% (Roodbergen & Teunissen 2014).

There is agreement that intensification of agricultural practices, such as the renewal of grassland, are key factors in causing the severe declines of grassland-breeding shorebirds (Charadrii) in Western Europe. They have led to loss and degradation of habitats and were found to reduce the breeding productivity by decreasing clutch and chick survival directly or via their influences on prey availability and predators (Chamberlain et al. 2000, Donald et al. 2001, Vickery et al. 2001, Wilson et al. 2004, Schekkerman & Beintema 2007, Schekkerman et al. 2009, Kleijn et al. 2010, Kentie et al. 2013, Kentie et al. 2015).

The impact of grassland management on the nesting performance of Northern Lapwing Vanellus vanellus (hereafter Lapwing) was studied in detail by Baines (1988, 1989, 1990) in upland grassland habitats in northern England. He found that increased fertilizer application, drainage and reseeding resulted in declines of breeding densities by 74% in pastures and 56% in meadows. These declines in population size were associated with negative impacts on the Lapwings’ productivity. Hatching success, for instance, dropped significantly due to increasing failure rates caused by farmwork and predation. Similar results were obtained in a long-term study on upland farmland in southern Scotland (Taylor & Grant 2004). However, McCracken & Tallowin (2004) estimated that the open nature of the swards in the phase of establishment soon after reseeding might attract farmland birds because of a better accessibility to soil and surface-active invertebrates. Similarly, some studies reported high densities of breeding Lapwings on grasslands prepared for reseeding by application of herbicides and on recently reseeded grasslands in the establishment phase (Schifferli et al. 2006, Müller et al. 2009).

Since the effects of grassland renewal on breeding Lapwings might differ between the short and the long term, it needs further study to understand the impacts more completely. Especially with regard to the conservation of Lapwings, it is important to know whether reseeding of grasslands in breeding areas should continue to be practised or even promoted. In the present study, we therefore investigated the influence of grassland renewal on the breeding density of Lapwings in a chronosequence of five years, ranging from one year before until three years after reseeding. Since agricultural management of grassland was found to affect nest survival, we additionally checked for habitat-specific differences in hatching success and causes of nesting failure.

Methods

Study area

Field data were collected in the Mittelradde-Niederung (Lower Saxony, Germany), a widely open landscape of approximately 2400 ha with several hedges and some small forests. Although most of the area is farmed intensively, it still harbours high numbers of breeding waders especially Eurasian Curlews Numenius arquata and Lapwings (compare Oertzen & Düttmann 2006). With regard to the agricultural use, the complete brook valley was farmed as permanent grassland until the early 1970s. Today, approximately 30% of the area is arable, with maize and winter cereals as the dominant crops. The remaining area is still permanent grassland on pied soil but also underwent dramatic management changes in the course of the last decades. Facilitated by extensive drainage and increasing levels of fertilizer application, traditional haymaking and cattle grazing have been replaced by early and frequent silage cuttings. These intensively farmed grasslands dominate the Mittelradde-Niederung. Non-intensively farmed grasslands, with traditional agricultural practices like haymaking and cattle grazing, are restricted to just a few areas.

Since 2007, the Mittelradde-Niederung has been part of a Special Protection Area under the European Birds Directive 79/409/EEC. Subsequently, conservation measures for breeding waders were implemented or extended. They predominantly consisted of a reimbursement to farmers to protect wader clutches during farming operations. In later years, the nest protection programme was extended by conservation measures for chicks, by leaving strips uncut on early-cut fields and reduced driving speed during mowing. However, farmers are free to join this conservation programme, and the conversation programme itself is not aiming to reduce the intensity of farming. Reseeding, application of high levels of fertilizer and early mowing dates, for instance, are still common practices. An adaptive management for grassland-breeding waders is implemented only on approximately 80 ha of permanent grassland owned by the government.

Data collection

Mapping of territories and clutches

In the years 1999–2011 territories of breeding Lapwings were mapped on 18 reseeded grasslands in the Mittelradde-Niederung. Although the size of the study area changed with time and the occurrence of reseeding of grassland plots was not predictable, we detected breeding territories of Lapwings on the 18 reseeded grasslands over a period of at least 5 years ranging from one year before until three years after the reseeding. The area of the reseeded pastures varied between 1.0 and 6.7 ha. According to standard methods applied for the monitoring of Lapwing territories (Hustings et al. 1989, Bibby et al. 1992, Südbeck et al. 2005) we visited the fields at least 5 times between the end of March and the end of June. Mapping was predominantly carried out in the early morning soon after sunrise by using binoculars and telescopes to scan the fields for nesting birds, adults with chicks or birds showing territorial behaviour.

In addition to the mapping of territories we counted Lapwing clutches on 16 areas of reseeded grassland in the years 2003–11. Again, we followed the changes in clutch numbers over 5 years ranging from one year before until three years after reseeding.

In this study, we exclusively focussed on short- and long-term effects of grassland reseeding on the setup of territories and the production of clutches of Lapwings. We did not check for dependent variables giving an explanation for changes in habitat selection of breeding Lapwings along the chronosequence of 5 years, for example, sward height and the presence of bare soil.

Nesting dispersion

Since Lapwings may breed in single pairs as well as in colonies of different size (Shrubb 2007), we also monitored the patterns of nesting dispersion in different habitats of the Mittelradde-Niederung. In particular, we recorded nesting dispersion of Lapwings in a sample of 222 occupied fields in the years 2005–09. Of these 222 occupied fields, 121 were tillage, 85 permanent grassland and 16 reseeded grassland in the phase of sward establishment. Permanent grassland was defined as all grasslands covered with grass plants for at least two years.

Nest survival and causes of nesting failure

In order to check for habitat-specific differences in nest survival and causes of nesting failure, we used data from a conservation programme that includes nest protection to prevent destruction by farming operations. Thus all data on nest survival and nesting failure due to a specific cause relate to protected nests.

Since visible marking of wader nests is an efficient tool to avoid damage by farmers, it influences the causation of nesting failure (Junker et al. 2006, Kragten et al. 2008). From the years 2003–09, data were available on 670 protected nests. With regard to nesting habitat we found 409 lapwing nests in tillage, 191 nests in agriculturally improved pastures and 70 nests in reseeded grasslands during the phase of sward establishment.

Nest protection involved marking the location of nests with bamboo canes of approximately one meter height, placed 2–4 m from the nest. For estimating daily nest survival rates and daily nest failure rates we checked the protected nests almost weekly from a distance (usually over 150 m) between mid of March and mid of July. Nests were assumed to have hatched successfully when: (a) newly hatched chicks were present in the nest, (b) small eggshell remnants were found at the bottom of the nest, and/or (c) parent birds produced alarm behaviour when nest sites were approached. Nest visits were kept to a minimum since field studies revealed an increase in predation risk with nest visits (Teunissen et al. 2006, Roodbergen & Teunissen 2014). Causes of nesting failure were identified as predation of eggs as indicated by empty nests, remains of yolk or characteristic eggshell fragments left by predators (Bellebaum & Boschert 2003). When a nest contained cold eggs, it was recorded as deserted. When there were signs of recent farming operations with typical remnants of destroyed nests, we recorded them as failed due to farming activities.

Data analysis

Breeding pairs of Lapwings, counted by territory mapping as well as by searching for clutches, and colony size on reseeded grasslands were tested for significant differences between the years of the reseeding chronosequence. We applied generalized linear mixed models (GLMM) with a year of reseeding chronosequence as a fixed effect, field size as an offset, and year of data collection and field identification as random terms to account for data points collected on the same field or in the same year. Mixed models were applied after a likelihood ratio test showed a better fit than the corresponding linear model without random terms, following Zuur et al. (2009). For models with Lapwing territories or clutches as the dependent variable, we chose a Poisson distribution, because the variables were count data. In the model where differences in colony size of Lapwings between the years of reseeding chronosequence were analysed, we chose a negative binomial distribution because a likelihood ratio test showed a better performance compared to a Poisson model. To test the significance of the fixed effects, a likelihood ratio test of the model with and without the fixed effect was carried out. The models were tested for overdispersion and checked graphically to ensure model assumptions were met (Zuur et al. 2009). GLMMs were analysed in R (R Core Team 2016) using packages MASS (Venables & Ripley 2002), lmtest (Zeileis & Hothorn 2002), lme4 (Bates et al. 2015) and blmeco (Korner-Nievergelt et al. 2015). For nesting dispersion, we additionally tested for differences between different breeding habitats, particularly between tillage, permanent grassland and reseeded grassland in the phase of sward establishment. We used a non-parametric Kruskal-Wallis analysis of variance (ANOVA), which in case of significance was followed by post-hoc tests to locate the significant difference(s).

To compare differences in hatching success between the breeding habitats we calculated daily nest survival rates with standard deviations (Mayfield 1961, Beintema & Müskens 1987). In the same way, we analysed nesting failure as a result of a particular cause, for example predation, desertion or farming activities. Nests were assumed to have been lost between the last two visits. Failure dates were rounded up to the nearest day (Hazler 2004). Habitat-specific differences in nest survival and failure rates were analysed by Mayfield logistic regression, an approach which removes bias caused by unrecorded failed nests and the stage at which nests were found. Mayfield logistic regression assumes binomially distributed data with the logit function as a link function (Aebisher 1999). In particular, by using the statistical package GenStat (VSN-G-ED; Site reference: AD5GA6) we explored the response variables of daily nest survival rate and daily risk of nesting failure due to a specific cause and used breeding habitat (3 level factor) as an explanatory variable.

Results

Breeding density

Territory mapping and counts of clutches showed the same pattern in the chronosequence of grassland reseeding (Table 1), both differed significantly between years of reseeding chronosequence (likelihood ratio test of GLMM with and without fixed effect, territories: χ²= 41.6, df = 4, P < 0.001; clutches: χ²= 35.9, df = 4, P < 0.001). In particular, breeding density of Lapwings significantly increased in the phase of sward establishment immediately after reseeding. The subsequent growth and intensive management of the swards reduced the abundance of the birds to values before the reseeding. In particular, one year after sward establishment, sixteen of eighteen plots housed no breeding Lapwings at all. In the following two years, the number of reseeded fields colonized by Lapwings slowly increased (two years after sward establishment: 5 of 18 fields; three years after sward establishment: 9 of 18 fields), but the breeding density remained low compared to the phase of sward establishment.

Table 1. Chronosequence of changes in territory and nest densities (mean/10 ha ± se) of Lapwings in the Mittelradde-Niederung, Lower Saxony, Germany, due to reseeding of grassland. Different letters refer to statistical differences (P < 0.05) obtained for each series from a GLMM with year of reseeding chronosequence as fixed effect.

	Year before reseeding	Year of reseeding	One year after reseeding	Two years after reseeding	Three years after reseeding
Territory densities	2.16 ± 0.69^a	13.37 ± 2.45^b	1.4 ± 0.82^a	1.52 ± 0.62^a	3.23 ± 1.33^a
Nest densities	1.74 ± 0.75^a	16.3 ± 3.54^b	1.76 ± 1.08^a	1.57 ± 0.69^a	2.85 ± 1.45^a
Colony size	1.5 ± 0.27^a	3.94 ± 0.92^b	2.67 ± 1.2^a	1.67 ± 0.42^a	1.33 ± 0.24^a

Nesting dispersion

Colony size differed between years of reseeding chronosequence (likelihood ratio test of GLMM with and without fixed effect: χ²= 40.9, df = 4, P < 0.001), with significantly larger colonies in the year of reseeding than in the years before and after (Table 1). Nesting dispersion of Lapwings in reseeded grasslands in the phase of sward establishment significantly differed from those in tillage and permanent grassland (Kruskal-Wallis ANOVA: H = 7.34, P = 0.017). In particular, in the former, Lapwings tended to breed in larger colonies (Table 3), while no such difference was found between tillage and permanent grassland (P = 0.416).

Hatching success and causes of nesting failure

Table 2 gives a habitat-specific overview of the number of successful nests and nests failed due to a specific cause. Daily nest survival rates varied between tillage, permanent grassland and recently reseeded grassland (Table 3). However, these habitat-specific differences were not significant (GLM, D = 4.3, df = 2, P = 0.118).

Table 2. Percentages of successful Lapwing nests and nests failed due to a specific causes in different breeding habitats (tillage: n = 409 nests, permanent grassland: n = 191 nests, reseeded grassland: n = 70 nests) of the Mittelradde-Niederung, Lower Saxony, Germany.

	Reseeded grasslands	Permanent grasslands	Tillages	Total
Successful	61.43 (n = 43)	40.84 (n = 78)	51.10 (n = 209)	49.25 (n = 330)
Predation	32.86 (n = 23)	57.07 (n = 109)	39.61 (n = 162)	43.88 (n = 294)
Desertion	4.29 (n = 3)	2.09 (n = 4)	4.89 (n = 20)	4.03 (n = 27)
Farming operations	1.43 (n = 1)	0.00	4.40 (n = 18)	2.84 (n = 19)

Table 3. Habitat-specific differences in daily nest survival rates and nest failure rates (means, ± se) due to different causes of protected Lapwing nests in the Mittelradde-Niederung, Lower Saxony, Germany. Additionally, we analysed habitat-specific differences in colony size (mean, se) in a sample of 222 occupied plots. For further details see text.

	Reseeded grasslands	Permanent grasslands	Tillages
Nest survival rate	0.9744 ± 0.0048	0.9634 ± 0.0035	0.97 ± 0.0021
Nest predation risk	0.0216 ± 0.0045	0.037 ± 0.0035	0.025 ± 0.0019
Risk of nest desertion	0.0028 ± 0.0016	0.0014 ± 0.0007	0.0031 ± 0.0007
Nest failure by farming operations	0.0009 ± 0.0009	0	0.0028 ± 0.0007
Colony size	4.0 ± 0.91	2.07 ± 0.16	2.4 ± 0.19

With regard to causes of nesting failure, predation was found to be the dominant factor over all breeding habitats (Table 3). Predation risk, however, significantly differed with habitat (GLM, D = 12.0, df = 2, P = 0.003). Estimates of parameters revealed that nest predation occurred more frequently in agriculturally improved grassland than in tillage and in recently reseeded grassland (both tests: t > 2.38, P < 0.018). Additionally, we found habitat-specific differences in nest destruction by farming activities (Table 3; GLM, D = 7.2, df = 2, P = 0.027). Despite the implementation of nest protection measures some nests were still destroyed by farming machines. In particular, estimates of parameters indicated a significant difference in nesting failure by farming operations between agriculturally improved grasslands and tillage, with higher rates in the latter (t = 1.97, P = 0.049). In contrast, we found no significant differences in nest desertion with habitat (Table 3; GLM, D = 2.7, df = 2, P = 0.26).

Discussion

Breeding density and dispersion

In the present study, we found evidence that reseeding of grassland attracts breeding Lapwings during the phase of sward establishment. In reseeded grassland, birds even tended to breed in larger colonies than in other habitats, such as tillage and intensively farmed pastures. However, the attractiveness of reseeding grassland for breeding Lapwings was just a short-term effect without sustainability. It was restricted to the short period of sward establishment which in the present study covered just one single breeding period. In line with Baines (1988) we subsequently observed an avoidance of these grasslands by the birds. Although we found a slow increase in breeding Lapwings in the years following sward establishment, breeding densities remained on a low level throughout the study period.

Although we did not check for an effect of independent variables on the habitat choice of breeding Lapwings through the chronosequence of grassland reseeding, it is likely that the attractiveness and the following avoidance of reseeded grasslands can be explained by well-known habitat preferences of this wader species: Nesting Lapwings select open landscapes and grassland habitats with short, open swards. Further factors favouring the breeding of Lapwings are high water tables and the presence of open shallow waters (Klomp 1954, Berg et al. 1992, Liker 1992, Milsom et al. 2002, Verhulst et al. 2007). During the phase of sward establishment, reseeded grassland often meets the habitat requirements of breeding Lapwings. In the Mittelradde-Niederung, for instance, the growing of reseeded grass was frequently hampered by high water tables, especially after heavy rain. As a result, these grasslands possessed heterogeneous swards, areas of bare soil without vegetation and temporary shallow water pools. Since Lapwings are highly responsive to sward and surface water conditions during the period of settlement, it is likely that the habitat conditions at the phase of sward establishment enhanced the set up of breeding territories. In line with the present study McCallum (2012) reported high densities of breeding Lapwings in fields that underwent fodder crop management for two consecutive years followed by reseeding with grass. The reseeded and subsequently grazed pastures harboured high percentages of bare ground, likely to attract Lapwings.

In the present study, not all reseeded grassland was colonized by the birds. In these cases, we assume that other factors, such as short distances to neighbouring hedges and trees or the fast establishment of swards, prevented Lapwings from nesting. In that respect, we observed a faster establishment of swards at drier sites than at wet ones.

Even at wet sites sward establishment took place within a year after reseeding. At that stage, intensively farmed grassland lost the former patchy structure. In particular, the rapid growth of swards which was almost always accelerated by fertilizer application resulted in a uniform and dense vegetation structure. In line with Baines (1988), who documented a significant decline in breeding Lapwings when old pastures were turned into intensively farmed grassland, we observed the same effect in reseeded grassland after sward establishment. The result that low densities of Lapwings occurred in the grassland of the Mittelradde-Niederung even before they were reseeded is easily explained by an almost complete disappearance of old pastures within the last 40 years (unpublished data).

Although not significant, we observed an increase in Lapwing density in reseeded grassland with age of sward establishment. With regard to the underlying mechanisms, it is well known that changes in plant species composition and vegetation structure take place with ageing of the (reseeded) grassland. However, these processes – even when accelerated by conservation measures such as cessation of fertilizer application – take place on a time scale of decades or more (Bakker 1989, Kapfer 1988, Bakker & Berendse 1999). The changes in plant composition and vegetation structure are known to be related to changes in animal communities including birds (Siemann et al. 1998, Bekker et al. 2006, Haddad et al. 2009). Therefore, we assume that the observed increase in Lapwing density with age of sward establishment was also caused by changes in plant composition and vegetation structure.

Hatching success and nesting failure

Differences in habitat quality are well known to influence reproductive behaviour and breeding success in birds (Krebs 1971, Pierotti 1982, Ens et al. 1992, Lambrechts et al. 2004). Several studies of the Lapwing have revealed significant differences in hatching success and the causes of nesting failure between arable land and intensively farmed grassland. In particular, in line with the present study, nest predation rates were found to be higher in grassland than in arable land, especially when the former were reseeding, drained and received applications of high levels of fertilizer (Galbraith 1988, Baines 1990, Shrubb 1990, Sheldon et al. 2007, Mac Donald & Bolton 2008a, but see also Salek & Smilauer 2002). It is likely that this habitat-specific difference in nest predation rate is due to differences in predator communities. However, in the present study, we observed such differences within the same study site. Since arable fields are less attractive foraging areas than grasslands because of lower densities of prey items (Jacob et al. 2014), we suggest that the present difference might reflect habitat use of the predators.

Apart from patterns of the environment and resulting differences in predator communities, nest predation risk in Lapwings is also affected by the social behaviour of the birds. Several studies revealed a significant decline in nest predation rate with colony size (Berg et al. 1992, Seymour et al. 2003, Mac Donald & Bolton 2008b, Laidlaw et al. 2016). In the present study, we obtained a similar result: The lowest nest predation rate was found in recently reseeded grassland which possessed larger breeding colonies than tillage and intensivelely farmed grassland.

In conventionally managed grassland and tillage, Lapwings also often suffer from high nest losses by farming operations including trampling by stock (Beintema & Müskens 1987, Berg et al. 1992, Kragten et al. 2008). Nest protection measures are able to reduce these losses to low levels (Teunissen et al. 2006, Kragten et al. 2008). Nesting failure under the implementation of nest protection measures accidentally occur by overlooking the nest markers during farming operations. Although the frequency of occurrence of such failures was generally low we observed more such cases on tillage than on grasslands.

Nest protection measures as used in the present study are known to produce extra losses due to predation, especially at sites with generally high predation rates (Goedhart et al. 2010, Teunissen et al. 2006, but see also Zamecnik et al. 2017). Roodbergen & Teunissen (2014) recommend, therefore, the implementation of nest protection measures only under specific conditions, for example, the presence of sufficient chick foraging habitat under an adapted agricultural management scheme in an open landscape with high water tables.

Implications for conservation

According to our results reseeding offers one year of good conditions for Lapwings which breed in large colonies with low nest predation rates. Thus, a continuous mosaic of reseeding operations in wet soil conditions could be a promising conservation measure to maintain breeding populations of Lapwings. However, in the Mittelradde-Niederung the breeding population of Lapwings dropped by about 35% in the period 2009–16 despite reseeding of grassland being a common practise by local farmers (Bio-Consult & regionalplan & uvp 2009; Regioplan & uvp 2016, Degen 2017). This negative population development might be due to the current reseeding practice, which can be characterized as uncontrolled and exclusively driven by agronomical interests. A somewhat different result could be expected if reseeding events were carried out in a controlled way with some thought for conservation, for example, by defining the proportion of grassland reseeded in a rota system per year. However, for several reasons we cannot recommend such a conservation management for Lapwings: (1) Objective evidence that a continuous mosaic of reseeding operations is able to maintain breeding populations of Lapwings is still lacking. Although we found high hatching rates of clutches in recently reseeded grassland, it is unclear whether these high rates also result in a sufficient reproductive outcome. If not, just reseeded grassland would represent ecological traps which increase the attractiveness as breeding habitat disproportionately in relation to its value for reproduction; (2) Appropriate conditions for breeding Lapwings and other grassland-breeding waders can be achieved by raising and maintaining high water levels until mid-summer, which ensures the penetrability of the soil surface and provides sufficient prey items for adults and chicks (Ausden et al. 2003, Wilson et al. 2004, Eglington et al. 2010). In combination with an adaptive farming practice focussing on the survival of nests and chicks, for example, by reducing livestock densities, postponing the first cut of grass and applying less fertilizer, such a management has shown to attract Lapwings and to produce almost stable breeding populations over a long time and not for just one year. High water tables throughout the breeding period and an adaptive farming practice are predominantly found in nature reserves in the possession of the state or nature conservation charities (Wilson et al. 2005, Düttmann et al. 2006, Wilson et al. 2007, Blüml et al. 2012, Roodhart 2014). Conservation programmes which include the aforementioned patterns of habitat management and adaptive farming are preferable also under environmental aspects, since they reduce emissions of nitrogen and CO₂ which are known to occur to a great extent in the course of grassland renovation on drained pied soils (Oleszczuk et al. 2008, Seidel et al. 2009, Velthof et al. 2010). In summary, the renewal of grassland by ploughing and reseeding has short-term beneficial effects for breeding Lapwing, but it is followed by a long period of unfavourable conditions for all grassland-breeding waders including the Lapwing (Baines 1988, 1989, Kleijn et al. 2010, Kentie et al. 2013, 2015). Moreover, since up to now we have no evidence that the renewal of grassland is able to produce stable populations of Lapwings, we have no reason to recommend reseeding as a conservation measure. Due to its long-term negative effects on populations of grassland-breeding waders it should no longer be practised in meadow areas, especially when they are designed to protect these bird species.

Acknowledgements

We are grateful to all farmers, volunteers and students who participated in the nest site protection of Northern Lapwings in the Mittelradde-Niederung. We also wish to thank H. Hötker (Bergenhusen), H.-H. Bergmann (Bad Arolsen) and two anonymous referees for useful comments on the manuscript.

Additional information

Funding

The study was partly funded by Landkreis Cloppenburg and Landkreis Emsland.