Monitoring High Nature Value Grassland in Transylvania, Romania

Laura SUTCLIFFE1 and Krystyna LARKHAM2

1 Dept Vegetation Analysis, Univ. Göttingen, Untere Karspüle 2, 37073 Göttingen, Germany
sutcliffe.laura@gmail.com
2 Imperial College London, South Kensington, London SW7 2AZ, UK

download pdf

KEYWORDS:

HNV farming, Romania, grassland biodiversity, agri-environment schemes, plant indicator species, monitoring.

ABSTRACT

Semi-natural grassland has in the last 50 years become increasingly rare in much of Europe, however, low levels of intensification and small-scale farming have led to the preservation of substantial areas of this valuable habitat in Romania. Nevertheless, here too land management practices have in recent years begun to change, and there is increasing threat to Romanian grassland biodiversity from abandonment and intensification. One means of countering this threat is to offer financial support for ‘biodiversity-friendly’ farming through agri-environment scheme payments, such as the scheme initiated in Romania for High Nature Value grassland in 2007. One important aspect of agri-environment schemes is the monitoring of its efficacy in maintaining or improving the ecological quality of the farmland. Therefore, this investigation uses multivariate analysis of vegetation survey data to identify a list of plant indicator species that can be used to monitor the quality of lowland grassland under agri-environment scheme, based on the study region of Southern Transylvania.

INTRODUCTION

Low-intensity agricultural habitats are one of the most valuable sources of biodiversity in Europe (Bignal & McCracken, 1996). This is particularly apparent in semi-natural grasslands, i.e. naturally occurring (not planted) vegetation that is regularly grazed or cut, resulting in a state that mimics natural habitats (Beaufoy, 2008), which can contain some of the highest plant densities in Europe, if not the world.

Nevertheless, semi-natural grassland is under threat due to its low economic viability in modern agriculture. A dramatic decline in the surface area of semi-natural grassland has been seen in the last century, particularly in northern and western European countries such as the UK, where some 97% of semi-natural grassland has been lost since 1930 (Fuller, 1987). The major drivers of this loss have been intensification to increase yields of more fertile areas, and associated abandonment of more marginal land.

In recognition of both the importance of areas of high agricultural biodiversity and the threat they face, the concept of High Nature Value (HNV) farmland was created to describe “those areas in Europe where agriculture is a major (usually the dominant) land use and where that agriculture supports or is associated with either a high species and habitat diversity, or the presence of species of European conservation concern, or both.” (Andersen et al., 2003). Eligible farmland is identified on the basis of 1) a high proportion of semi-natural vegetation; 2) the dominance of low-intensity agriculture or a mosaic of semi-natural and cultivated land and small-scale features; 3) the presence rare species or a high proportion of European or world populations.

The conservation of HNV farmland is now an EU objective and measures such as agri-environment schemes (AES) to provide support for these areas have been built into the 2007-2013 rural development plans of the Member States (EEA, 2004). As part of its accession to the EU in 2007, Romania now provides agri-environment payments for areas identified as HNV (34% of the utilisable agricultural area: Paracchini et al., 2008: see Figure 1).

Figure 1: Current distribution of HNV grassland (shaded grey) in Romania according to CORINE Land Cover data (adapted from: NRDP 2007). Study area outlined with a black box.

To qualify for the payments, landowners must follow certain management prescriptions designed to prevent the effects of agricultural intensification. The basic HNV package 1 forbids, for example, any ploughing, rolling, seeding and the use of chemical fertilisers and pesticides during the 5-year contract period, as well as delaying mowing until after the 1st of July and restricting stocking rates to below 1 Livestock Unit ha-1 (NRDP, 2007). The effectiveness of these measures for all areas under the scheme is, however, not guaranteed. Monitoring of the quantity and quality of HNV grassland is therefore necessary to determine whether the quality of grassland is being maintained, or if adjustment to the measures is needed. In Romania, this monitoring is intended to be carried out through regular surveys of the relative numbers of vascular plant indicator species present in sample areas of HNV grassland (NRDP, 2007). The preliminary list of 22 plant indicator species given in the 2007–2013 National Rural Development Plan of Romania is incomplete (NRDP, 2007). Therefore in order to contribute the improvement of this list, this study used a multivariate analytical approach to identify appropriate plant indicator species based on the study area of the lowland grasslands of Southern Transylvania. It was assumed that the monitoring procedure will take the form of a line transect or similar, which will provide a rapid, representative overview of the parcel. Thus, highly visible and rapidly identifiable vascular plant taxa with a preference for low-intensity HNV grassland were sought.

MATERIAL AND METHODS

Between May and July 2009, the vascular flora of lowland mesic (Molinio-Arrhenatheratea R. Tx. 1937, and Festuco-Brometea Br.-Bl. et Tüxen ex Soó 1947) grassland around the villages of Saschiz, Bunesti and Viscri in Southern Transylvania was surveyed. Three management types (mixed pasture, sheep pasture and hay meadow) and three intensities (‘intensive’, with overgrazing or previous use of inputs; ‘extensive’, with no inputs and low stocking rates; and ‘abandoned’ or very undergrazed) were distinguished, and three repeats were sampled for each combination, giving 27 separate grassland parcels. For each parcel, five 100 m2 plots were sampled with five randomly placed 0.25m2 relevés in each, recording each species and its abundance/coverage (Braun-Blanquet). In addition, a transect was walked in each parcel, recording all species within 2 m either side. Constrained multivariate analysis (Canonical Correspondence Analysis: Fig. 2) was used to identify those species which had the strongest relationship with extensive grassland (i.e. those situated closest to the ‘extensive’ variable on the ordination plot). These species were screened for suitability (ease and speed of recognition in the field) to create a final list of indicator species. The relative occurrences of these species at each intensity level were tested with the species lists recorded from the transect (the anticipated monitoring methodology) using an independent two-tailed t-test.

RESULTS AND DISCUSSION

Figure 2 shows the distribution of species relative to the use intensity. Those selected for inclusion in the final list are displayed in Table 1.

Figure 2:
i) Simple ordination plot of a CCA of all relevé data (135 samples) using focus on interspecies distances and biplot scaling (excluding the lower 3% of species weight and fit ranges, names removed for clarity). Species selected for inclusion in the preliminary list are displayed as black circles, all other species displayed as empty triangles. The variables 'Intensive', 'Extensive' and 'Abandoned' are displayed as filled triangles.
ii) Simple ordination plot of a CCA of all transect data (27 samples) treated as above, and excluding the lower 6% of species weight and fit ranges.

1: List of 28 indicator species for lowland mesophilic grassland

Family Species Family Species
Apiaceae Pimpinella saxifraga Fabaceae Trifolium alpestre
Asteraceae Centaurea jacea   Trifolium montanum
  Centaurea nigrescens Gentianaceae Centaurium rythraea
  Centaurea pseudophrygia Lamiaceae Betonica officinalis
  Leucanthemum vulgare   Clinopodium vulgare
Campanulaceae Campanula patula   Prunella laciniata
Caryophyllaceae Dianthus armeria   Salvia pratensis
  Dianthus carthusianorum   Thymus pulegioides
  Stellaria graminea Poaceae Anthoxanthum odoratum
Colchicaceae Colchicum autumnale   Briza media
Dipsacaceae Knautia arvensis Primulaceae Primula veris
Euphorbiaceae Euphorbia cyparissias Rosaceae Filipendula vulgaris
Fabaceae Genista tinctoria   Fragaria viridis
  Ononis repens Rubiaceae Galium verum

When testing against the transect data, there was a significantly higher number of indicator species in ‘extensive’ than ‘intensive’ (P < 0.005, t = 5.15, two tailed, df = 7) or ‘abandoned’ transects (P < 0.05, t = 2.52, two tailed, df = 7), as shown in the boxplot in Figure 3.

Figure 3. Boxplot of the number of indicator species recorded in transects on different use intensities (N=9 for each intensity). Different letters indicate significant differences (t-test, 7 d.f., p<0.05).

Regarding the applicability of the results, out of the 25 grassland habitat types listed for the Romanian HNV grassland area in Sârbu et al. (2004), this investigation includes only ‘Hill mountain mesophilous meadows’ (Cynosurion cristati), ‘Hill mountain mesophilous manured meadows’ (Arrhenatherion) and ‘Hill and plateau xero-mesophilous grasslands’ (Cirsio-brachypodion pinnati). Although these are by far the most common grassland types, in order to use indicator species at the national scale modifications of the list for other grassland types based on similar studies will be necessary. In their analysis of plant indicator species for extensive grassland at a national level, Oppermann et al. (2009) distinguished six geographical zones in Germany, with separate lists for each zone. A similar approach could be considered using the five biogeographical zones of Romania.

A further outcome of this study is the confirmation that a statistical approach can be used to successfully identify plant indicator species. An alternative approach to identifying plant indicator species is that of collecting biotope and species data for the region, from which species are selected or rejected based on criteria such as distribution, frequency and ecological preferences, as well as professional experience. This method was successfully used in Germany to identify plant indicator species for grassland biodiversity (Matzdorf et al., 2008). However, this relies heavily on comprehensive, accurate and up-to-date botanical and habitat information of the kind not yet fully available for Romania.

If chosen well and recorded appropriately, regular monitoring of the relative number of indicator species in selected HNV grassland parcels throughout the country should detect changes in habitat quality. The success of the HNV scheme can thus be evaluated and, in the case of a decline in quality, changes made to the measures before potentially irreversible losses of biodiversity are suffered. Such a feedback system is vital, as the HNV management prescriptions as they stand are unlikely to be sufficient to conserve the existing level of biodiversity, in the face of the modernisation of Romanian agriculture. Although some traditional methods (e.g. the use of animal drawn machinery) are rewarded by a premium (‘package 2’), such labour intensive practices are likely to disappear as mechanisation becomes more widespread in the country. As Kleijn & Sutherland (2003) observe, even if an AES is financially beneficial, if there is little support, feedback or evaluation (as is currently the case in Romania), then agri-environment prescriptions will be viewed as merely an inconvenience, to be carried out with the least possible effort. The success of AES thus lies as much in the attitudes of the participating farmers, as in the measures themselves (Wilson & Hart, 2001), and more effort needs to be made across Europe to educate landowners and increase involvement in, and awareness of, achieving biodiversity goals.

Whilst they can be criticised for not going far enough, management prescriptions can also lead to the danger of inflexibility. Agri-environment schemes must involve some additional activity or change in practice on the part of the farmer. This conflicts with one of the central characteristics of traditional farming landscapes, i.e. the flexible responsiveness and adaptation of practices to the prevailing regional environmental conditions. Changes to this management pose a considerable risk of disturbing the natural rhythms of agricultural wildlife, which are intimately bound up in the annual farming cycle. One example from the Romanian HNV farming guidelines is that mowing may not occur before the 1st of July, in order to protect ground nesting birds – around two to three weeks later than it is traditionally carried out in the study area (E Ghilea pers. comm., 2009). This can impact the composition of both fauna and flora, as the timing of the cut affects which species can contribute ripe seed to the seed bank and which animals are disturbed by mowing. Whilst such targeted conservation management, when carried out sensitively, can benefit many other species (the so-called ‘umbrella’ effect: Andelman & Fagan, 2000), experience has shown that when it differs from the historical management, it may lead to a decline in populations of other species and unbalance ecosystems. A dramatic example of this is given by Konvicka et al. (2008), who found that changed mowing patterns in Czech meadows to conform to agri-environment regulations, directly caused the extinction of a large population of the endangered butterfly species, Colias myrmidone.

AES should thus ideally enable and encourage farmers to continue with the most ecologically friendly practices. This is claimed to be the advantage of ‘results-based’ schemes. Here, in contrast to the usual ‘management-based’ schemes, payments are directly linked to biodiversity, leaving the farmers free to manage their fields as they wish. This approach has been used since 1994 for a regional AES in the German state of Brandenburg (Kaiser et al., 2009) and has in recent years also been introduced to the states of Baden-Württemberg (Oppermann et al., 2002) and Lower Saxony (Keienburg et al., 2006). Under these schemes, fields are assessed using a standardised transect methodology, recording the number of plant indicator species for extensive management. Fields containing over a certain number of indicator species then qualify for payments. By linking payments to outcomes, not only does it guarantee effectiveness and efficiency of investment, but also increases the engagement of the landowners, giving them a greater degree of autonomy and sense of involvement with the conservation process (Matzdorf et al., 2008).

The results-based scheme is essentially a small-scale, targeted (or ‘deep and narrow’) approach to conservation, requiring substantial infrastructure and support. In contrast, the strength of the HNV scheme is that it encompasses large expanses of the most ecologically valuable agricultural land. This will combat the fragmentation effects that smaller-scale AES cannot prevent, allowing the natural migration of species and the support of large, resilient populations. However, as a ‘broad and shallow’ scheme, resources are thinly spread and results are difficult to measure. The current lack of detailed baseline ecological information in Romania (and in many other EU countries) also means that any evaluation of success will necessarily rely heavily on expert judgement, rather than hard figures. Thus, it is all the more critical that a monitoring scheme with clear goals is put in place quickly and effectively, so that future changes in biodiversity can be measured and declines in this natural resource prevented.

CONCLUSIONS

The analysis yielded 28 vascular plant indicator species (shown in Table 1). The significant differences in the number of species present in the different management intensities (Fig. 3) suggest that these species will decline in response to intensification or abandonment. Regular national monitoring of the number of indicator species in sample areas under HNV AES agreement (relative to control areas) will provide a measure of success of the Romanian AES, informing any improvements that may need to be made. This study covers the common lowland mesophilic grassland type, however, further studies of the 26 or so other grassland types in Romania will be necessary to provide specific lists for different areas.

ACKNOWLEDGEMENTS

The authors would like to thank the NGO Fundatia ADEPT (www.fundatia-adept.org) for their generous support throughout the fieldwork for this study, as well as Prof. Erwin Bergmeier.

REFERENCES

Andelman, S. J., Fagan, W. F., 2000 – Umbrellas and flagships: efficient conservation surrogates or expensive mistakes? Proceedings of the National Academy of Sciences of the United States of America 97:5954–5959.

Andersen, E., Baldock, D., Bennett, H., Beaufoy, G., Bignal, E., Brouwer, F., Elbersen, B., Eiden, G., Godeschalk, F., Jones, G., McCracken, D.I., Nieuwenhuizen, W., van Eupen, M., Hennekens, S. & Zervas, G., 2003 – Developing a High Nature Value indicator. Report for the European Environment Agency, Copenhagen, 76 pp.

Beaufoy, G., 2008 – HNV farming – explaining the concept and interpreting EU and national policy commitments. European Forum on Nature Conservation and Pastoralism. 15 pp.

Bignal E., McCracken D., 1996 – Low intensity farming systems in the conservation of the countryside. Journal of Applied Ecology 33:413–424.

EEA - European Environment Agency, 2004 – High Nature Value farmland - Characteristics, trends and policy challenges. European Environment Agency, Luxembourg: Office for Official Publications of the European Communities. 32 pp.

Fuller, R.M., 1987 – The changing extent and conservation interest of lowland grasslands in England and Wales ‐ a review of grassland surveys 1930‐84. Biological Conservation 40: 281‐300.

Kaiser, T., Rohner, M.-S., Reutter, M., Matzdorf, B., Schaepe, A., Hoffmann, E.. 2009 – Die Entwicklung einer Kennartenmethode zur Förderung von artenreichem Grünland in Brandenburg. Naturschutz und Landschaftspflege in Brandenburg 18:44–50 (in German).

Keienburg, T., Most, A., Prüter, J. (Eds.), 2006 – Entwicklung und Erprobung von
Methoden für die ergebnisorientierte Honorierung ökologischer Leistungen im
Grünland Nordwestdeutschlands. NNA-Berichte 19, 257 pp. (in German).

Kleijn, D., Sutherland, W. J., 2003 – How effective are European agri-environment schemes in conserving and promoting biodiversity? Journal of Applied Ecology 40:947–969.

Konvicka, M., Benes, J., Cizek, O., Kopecek, F., Konvicka, O., Vitaz, L., 2008 – How too much care kills species: Grassland reserves, agri-environmental schemes and extinction of Colias myrmidone butterfly from its former stronghold. Journal of Insect Conservation 12:519–525.

Matzdorf, B., Kaiser, T., Rohner, M.-S., 2008 – Developing biodiversity indicator to design efficient agri-environmental schemes for extensively used grassland. Ecological Indicators 8:256–269.

NRDP - Government of Romania, Ministry of Agriculture and Rural Development, 2007 – National Rural Development Programme 2007–2013. 812 pp.

Oppermann, R., Briemle, G., 2002 – Blumenwiesen in der landwirtschaftlichen Förderung. Erste Erfahrungen mit der ergebnisorientierten Förderung im baden-würt tembergischen Agrar-Umweltprogramm MEKA II. Naturschutz und Landschaftsplanung 34:203–209 (in German).

Oppermann, R., Krismann, A., Sonnberger, M., Weiß, B., 2009 – Bundesweites Biodiversitätsmonitoring zur Grünlandvegetation. Methodik und erste Erfahrungen.
Natur Landsch. 84: 62-70 (in German)

Paracchini, M.L., Petersen, J.E., Hoogeveen, Y., Bamps, C., Burfield, I., van Swaay, C., 2008 – High Nature Value Farmland in Europe - An estimate of the distribution patterns on the basis of land cover and biodiversity data. Report EUR 23480 EN. European Commission Joint Research Centre, Institute for Environment and

Sârbu A, Coldea G, Negrean G, Cristea V, Hanganu J, Veen P., 2004 – Grasslands of Romania. Final report on National Grassland Inventory 20002003. Bucharest: University of Bucharest & Royal Dutch Society for Nature Conservation. 71 pp.

Wilson, G. A., Hart, K., 2001 – Farmer participation in agri-environmental schemes: towards conservation-oriented thinking? Sociologia Ruralis 41:254–274.