RESEARCH ARTICLE

Restoration of Heterogeneous Disturbance Regimes for the Preservation of Endangered Species Steven D. Warren and Reiner Büttner

ABSTRACT Disturbance is a natural component of . All species, including threatened and endangered species, evolved in the presence of, and are adapted to natural disturbance regimes that vary in the kind, frequency, severity, and duration of disturbance. We investigated the relationship between the level of visible soil disturbance and the density of four endan- gered plant species on U.S. Army training lands in the German state of Bavaria. Two species, gray hairgrass (Corynephorus canescens) and mudwort (Limosella aquatica), showed marked affinity for or dependency on high levels of recent soil disturbance. The density of fringed gentian (Gentianella ciliata) and shepherd’s cress (Teesdalia nudicaulis) declined with recent disturbance, but appeared to favor older disturbance which could not be quantified by the methods employed in this study. The study illustrates the need to restore and maintain disturbance regimes that are heterogeneous in terms of the intensity of and time since disturbance. Such a restoration strategy has the potential to favor plant species along the entire spectrum of , thereby maximizing plant and stability.

Keywords: fringed gentian, gray hairgrass, heterogeneous disturbance hypothesis, mudwort, shepherd’s cress

cosystems and the species that create and maintain conditions neces- When the European Commission Einhabit them typically evolve in sary for survival. issued Directive 92/43/EEC (Euro- the presence of quasi-stable distur- The of northern Europe pean Economic 1992) bance regimes which are characterized lie within what would be mostly forest requiring all European Union nations by general patterns of perturbation, in the absence of disturbance. Prior to designate a certain percentage of including the types of disturbance and to recent anthropogenic alteration of their landmass as ‘special areas of their frequency, intensity, size, peri- disturbance regimes, the grasslands conservation’ in the NATURA 2000 odicity, and/or duration. Significant were maintained for approximately network, many nations recognized the alteration of the disturbance regimes 1.8 million years by wild ungulate her- ecological value of their military train- has the potential to negatively affect bivores (Pärtel et al. 2005). During the ing areas. As a result, they nominated native biodiversity. The restoration of last millennium, wild ungulates have large tracts of their training lands. For biodiversity is a common goal of res- been largely replaced by domestic her- example, Denmark nominated 45% of toration ecologists around the globe. bivores. More recently, the abandon- its military training landmass, while However, the process is often more ment of traditional practices the Netherlands included 50%, and complicated than it initially appears, and increased use of agricultural fer- Belgium 70% (Gazenbeek 2005). and may well involve the restoration tilization have resulted in a dramatic In Germany, military training areas of the disturbance regime in addition decrease in the spatial extent and bio- provide suitable for a greater to the reintroduction of native spe- diversity of the grasslands ( Jentsch and density of Red List (the European cies ( Jentsch 2007). This can be par- Beyschlag 2003, Butaye et al. 2005, counterpart of U.S. ‘threatened and ticularly important for threatened and Isselstein et al. 2005). endangered’) animal species than pro- endangered species that are dependent During the Cold War, military vided by most other ‘special areas of on a particular disturbance regime to training areas represented some of the conservation’ (Warren et al. 2007). largest tracts of in northern In the post-Cold War era, numerous Ecological Restoration Vol. 32, No. 2, 2014 Europe. Disturbance created by mili- military training areas in Europe have ISSN 1522-4740 E-ISSN 1543-4079 tary training maneuvers provided an been abandoned or have experienced ©2014 by the Board of Regents of the effective surrogate for the disturbance a significantly diminished frequency University of Wisconsin System. under which the grasslands evolved. and intensity of military-related

June 2014 ECOLOGICAL RESTORATION 32:2 • 189 disturbance. Many, with extensive Act. It grows erect to 25 cm and occurs FFTA is located approximately and diverse grasslands, have experi- on dry grasslands with soils similar 35 km northeast of HTA and 5 km enced encroachment of woody species to mudwort, but without the addi- southeast of the town of Amberg. It and concomitant reductions in the tional moisture (Rothmaler 1994, comprises approximately 138 ha. Due populations of species that depend on Rosenbauer 1996). Shepherd’s cress to intensive training activities during grasslands maintained by disturbance (Teesdalia nudicaulis) is a small annual the 1960’s, large areas were converted (IUCN 1996, Chojnacka et al. 2010, forb that is listed as endangered on into barren sand. While large sandy Tschöpe et al. 2011). the Bavarian Red List. It occurs on meadows and open sandy soil can Our hypothesis, based on years sandy, acidic grasslands (Schulze- still be found, the open character is of personal observation on the U.S. Motel 1986, Oberdorfer 1990). currently threatened by shrub inva- Army training areas in Germany, was sion and increasing crown density of that numerous Red List plant species Study Area reforested areas. were favored by the disturbance regime We sampled plant populations at four KTA comprises approximately related to military training activity. active U.S. Army training facilities 1,088 ha and is situated northeast The objective of this effort was to test in the state of Bavaria: Grafenwöhr of the town of Kitzingen. It lies in the hypothesis and determine the Training Area (GTA); Hohenfels a large Pleistocene sand dune area. degree to which it was valid. Assum- Training Area (HTA); Freihölser Mixed layers of sand and gravel depos- ing the hypothesis was supported by Forst Local Training Area (FFTA); its from the nearby Main River and the research data, the information will and Klosterforst Training Area (KTA). former sand dunes characterize the assist restoration ecologists in identi- GTA is approximately 23,000 ha in subsoil. The area has been used for fying and quantifying an appropri- size. It was first used by the Royal military training since before World ate disturbance regime to maintain Bavarian Army in 1910 and has been War I, first by the German Army these threatened and endangered plant used continuously since that time, and, since World War II, by the U.S. species. with control ceded to the U.S. Army Army. Prior to military usage, the area at of World War II (Burkhardt was mostly forested with a mixture Methods 1994). The training area is located of hardwood and coniferous forest. between the towns of Vilseck and Approximately 100 ha were cleared for Study Species Grafenwöhr, about 88 km northeast military maneuvers and have remained We evaluated four Red List plant spe- of Nürnberg. The western third of the open since that time. training area is characterized by calcar- cies occurring on U.S. Army training Data Collection areas in the German state of Bavaria. eous soils; the eastern part is composed Gray hairgrass (Corynephorus canes- mainly of sandy soils. It is situated in a Field methods varied based on the cens) is a densely tufted, glaucous low-lying area with numerous streams, species of interest. Gray hairgrass was perennial grass that grows to 30 cm. small lakes and swampy areas. Average found and surveyed at GTA, KTA and It is listed as endangered on the Bavar- annual precipitation is approximately FFTA. We used a plot size of 5 m × 630 mm. 5 m, or 25 m2. One hundred plots ian Red List. It occurs most frequently HTA is approximately 16,000 ha on sandy, acidic grasslands with loose were established; 29 at GTA, 21 at in size. It has been used almost con- sand (Oberdorfer 1990, Rothmaler KTA, and 50 at FFTA in areas where tinuously since it was established in 1994). Mudwort (Limosella aquatica) the species was found. Because the is a small annual forb that prefers 1938 by the German Army (Bur- objective was to ascertain the rela- loamy or clayey soils and occurs most khardt 1994). It is located approxi- tionship, if any, between disturbance commonly in moist areas along the mately 84 km southeast of Nürnberg. and species , the plots were margins of ephemeral water puddles The training area is characterized by equally distributed among areas rep- hilly terrain with softly rounded rock resenting 5 levels of physical ground (Rothmaler 1994, Rosenbauer 1996). formations, karst topography, mixed It can grow to a height of 12 cm and disturbance (i.e. 0–20, 21–40, 41–60, coniferous and deciduous forest, and is listed as endangered on the Bavar- 61–80, and > 81%). A conservative dry valleys. The soils in the training ian Red List. Fringed gentian (Gen- definition of disturbance was used, tianella ciliata, syn. Gentiana ciliata area are mostly clay loams derived labeling as disturbed only the por- and Gentianopsis ciliata), is a small, from weathered limestone or dolo- tion of the plots where recent, defini- tive ground disturbance was evident late-flowering biennial that is listed mite. Dolomitic sands are also pres- ent. Average precipitation is between (e.g., visible vehicle tracks, damaged as endangered on the German Red plants, displaced soil, etc.); the mere List, is on the early warning list for 650 mm and 700 mm per year. Drain- absence of was considered Bavaria, and requires protection under age is predominantly internal, with Germany’s Federal Nature Protection much of the water entering the karst insufficient to definitively indicate system. recent disturbance. Six transects were

190 • June 2014 ECOLOGICAL RESTORATION 32:2 established at 1m intervals, and per- Table 1. The degrees of freedom (df), mean squares, F-ratios and p-values pendicular to one end of each plot. associated with the analyses of variance on the differences between study At 1 m intervals along each transect, sites and levels of disturbance for each of the four species evaluated in this study. the presence or absence of disturbance was recorded (36 points per plot) as a df Mean square F p quantitative estimate of the percent Gray hairgrass disturbance in the plot. All individuals Study sites 2 33.89 2.27 0.110 of gray hairgrass were counted by plot. Levels of disturbance 4 65.98 4.41 0.003 Circular plots with a radius of 3 m Interaction 8 20.28 1.36 0.227 and an area of 28.3 m2 were used to Error 82 14.93 measure fringed gentian, mudwort, Mudwort and shepherd’s cress. Fringed gentian Study sites 1 0.02 0.03 0.870 and mudwort were surveyed in 50 Levels of disturbance 4 0.87 17.29 0.000 plots each at GTA and in an additional Interaction 4 0.17 0.28 0.895 Error 143 0.63 50 plots each at HTA. For shepherd’s Fringed gentian cress, fifty survey plots were estab- Study sites 1 0.01 1.93 0.168 lished in suitable habitat at GTA; 50 Levels of disturbance 4 0.03 4.97 0.001 additional plots were established at Interaction 4 0.01 1.26 0.293 FFTA. Plots for all species were dis- Error 93 0.01 tributed evenly among areas represent- Shepherd’s cress ing the same five levels of relatively Study sites 1 23.16 2.23 0.139 uniform ground disturbance as used Levels of disturbance 4 28.63 2.76 0.033 for gray hairgrass. Interaction 4 2.07 2.03 0.097 Error 84 10.37 Statistics To maximize comparability with exist- ing and future studies, all count data disturbance significantly affected all The density of fringed gentian did were converted to density (numbers species (Table 1). not vary significantly among distur- per m2) prior to statistical analyses. Across all 3 locations where it was bance levels below 61% (Figure 3). Disturbance levels in each plot were present, the density of gray hairgrass It declined precipitously thereafter, converted to one of the five distur- did not vary significantly between the and was absent when disturbance bance level categories. Analyses of two lower levels of disturbance, i.e. exceeded 80%. The density of shep- variance were computed for each spe- 0–20% and 21–40%, but the density herd’s cress trended down as the level cies using the SYSTAT 13 analyti- at these two levels of disturbance was of disturbance increased (Figure 4). cal software (SYSTAT Software, Inc., statistically less than the density in The density was significantly lower Chicago, IL). The analyses tested the the two highest levels of disturbance, at > 81% disturbance than at 0–20% effects of the level of disturbance and i.e. 61–80% and > 81% (Figure 1). and 41–60% disturbance. Because of study site on the density of the respec- Density in the mid-range disturbance high variability within disturbance tive species. Statistical outliers were category (41–60%) was intermediate levels, no other statistical differences removed when encountered to prevent between the lower and higher distur- in plant density by disturbance level effects on the results and conclusions. bance categories, and not statistically were evident. When the analyses of variance revealed different from them. significant differences between treat- Mudwort followed a similar trend Discussion ments or sites, post-hoc separation of to gray hairgrass, increasing in density means were computed using the Fish- at greater levels of disturbance (Figure The four plant species sampled in er’s LSD method with a probability 2). The density of this forb was great- this study are considered to be threat- level of 0.05. est at >81% surface disturbance, and ened or endangered (Red List) under significantly less or absent at lower German and/or European Union guidelines, and all have been reported Results levels of disturbance. Although the density of mudwort trended upward as occurring on disturbed sites. Our There were no statistical differences as disturbance increased among the data show that gray hairgrass and in the density of any of the examined lower disturbance levels, the differ- mudwort tend to be early succes- species that could be attributed to the ences were not statistically significant sional species that colonize sites that study sites; the percentage of the soil until reaching the greatest disturbance show clear visible evidence of physi- surface showing visible signs of recent level. cal disturbance. Fringed gentian and

June 2014 ECOLOGICAL RESTORATION 32:2 • 191

6 1.5 b b b

5 ) ) 2 2 - - m m

4 1.0 ab (no. (no.

3 a 2 0.5 Density Density a a 1 a a a 0 0.0 0-20 21-40 41-60 61-80 81-100 0-20 21-40 41-60 61-80 81-100

Percent disturbance Percent disturbance

Figure 1. The density of gray hairgrass in plots repre- Figure 2. The density of mudwort in plots represent-

senting different levels of visible surface disturbance at ing different levels of visible surface disturbance at Grafenwöhr Training Area, Klosterforst Training Area, Grafenwöhr Training Area and Hohenfels Training Area, and Freihölser Forst Training Area, Germany. Bars with Germany. Bars with different letters differ significantly different letters differ significantly p( < 0.05). (p < 0.05).

6.0 a a 0.150 a

) ) 2 2 -

a - m m

4.0 ab 0.100 ab (no. (no.

a b

2.0 Density

0.050 Density bc 1

c 0.000 0.0 0-20 21-40 41-60 61-80 81-100 0-20 21-40 41-60 61-80 81-100 1

Percent disturbance Percent disturbance

Figure 3. The density of fringed gentian in plots rep- Figure 4. The density of shepherd’s cress in plots rep- resenting different levels of visible surface soil distur- resenting different levels of visible surface soil distur- bance at Grafenwöhr Training Area and Hohenfels bance at Grafenwöhr Training Area and Freihölser Forst Training Area, Germany. Bars with different letters Training Area, Germany. Bars with different letters differ significantly p( < 0.05). differ significantly p( < 0.05). shepherd’s cress occur somewhat later successional gradient. And, indeed, it 2006). It is commonly reported on in successional progression, after clear would likely be impossible on homog- dry, nutrient-poor grasslands derived evidence of physical disturbance may enously managed landscapes. from coastal (Łabuz and Grunewald no longer be readily apparent. Exam- Gray hairgrass exhibited affinity for 2007, Lomba et al. 2008) and inland ples of disturbance-dependent threat- physical soil disturbance, occurring (Ujházy et al. 2011) sand dunes. Such ened and endangered amphibians and significantly more frequently where grasslands generally support a sparse have also been documented on disturbance was plainly evident, and sward of vascular plants and numerous military training areas (Warren and being significantly less frequent where lichens ( Jentsch and Beyschlag 2003, Büttner 2008a, b). The situation pres- disturbance was less. Gray hairgrass is Hasse and Daniëls 2006), and progress ents somewhat of a quandary for land known as a pioneer species, colonizing toward woody vegetation where dis- managers who are expected to manage recently disturbed ground (Tschöpe turbance is insignificant or absent for for biodiversity and for the preserva- and Tielbörger 2010, Ödman et al. long periods ( Jentsch and Beyschlag tion of threatened and endangered 2011). It benefits, in terms of seedling 2003). Various forms of physical dis- 1 1 species, as such species may occur , establishment, and fruit- turbance help maintain conditions variously along a successional gradi- ing success, from physical soil distur- favorable for gray hairgrass, includ- ent. It may be considered illogical or bance ( Jentsch et al. 2002, Eichberg ing excavation by ants and mammals even impossible to manage simulta- et al. 2007, Jentsch et al. 2009) and ( Jentsch et al. 2002), burial by wind- neously for species along an entire sand deposition (Hasse and Daniëls blown sand (Marshall 1965, Forey et

192 • June 2014 ECOLOGICAL RESTORATION 32:2 al. 2008), surface mining (Wiegleb and Felinks 2001, Felinks and Wie- gand 2008, Řehouková and Prach 2008), trampling by domesticated and wild ungulates (Kooijman and de Haan 1995, Daniëls et al. 2008, Tschöpe et al. 2011) and military training maneuvers ( Jentsch et al. 2009, Chojnacka et al. 2010). Mudwort also exhibited clear signs of disturbance-dependence, being essentially absent at visible distur- bance levels less than 81%. This may be due to the fact that most pools and puddles present on the military train- ing areas utilized in this study were caused by vehicle traffic on wet soil. Mudwort is a pioneer species known to occur preferentially in that are regularly and shallowly inundated (Matevski and Kostadinovski 2009, Figure 5. A typical location at Grafenwöhr Training Area, Germany where Fischer et al 2010), often on muddy fringed gentian was found. river banks (Csiky and Purger 2008). It is an annual or short-lived perennial that thrives in shallow water for up to a few weeks, and then persists in the mud after water has receded. It was historically known to occur in the ruts of farm carts (Salisbury 1970) and in porcine pastures where the rooting behavior of pigs created small barren areas that collected pools of water (Beinlich and Poschlod 2002). Due largely to the decline of pasturing and other disturbances on seasonally wet, muddy habitat, mudwort has declined markedly (Salisbury 1970) and has reached a state of endangerment in several European nations (Király et al. 2008). Evidence of dependence on distur- bance was less clear with fringed gen- tian. Because of the conservative field Figure 6. A typical location at Freihölser Forst Training Area, Germany definition of disturbance used in this where shepherd’s cress was found. study, points with a history of distur- bance, but which no longer showed allowing a few years for seeds to be same geographic area, Poschlod et al. irrefutable evidence at the ground transported and germinate, it could (2008) reported that fringed gentian level, were not labeled as disturbed. be estimated that the plot was dis- tends to occupy recent grasslands Figure 5 is an example of a fringed turbed at least 10 years previously. and old field borders. In a long-term gentian plot. It appears obvious that Husáková (1996) lists fringed gentian study near Göttingen, Germany, a military battle tank had traversed as a ruderal species while Jentsch and Bornkamm (2006) recorded the pres- the area at some point in the past. Beierkuhnlein (2008) suggest that ence of fringed gentian for the first Given that the species (Picea fringed gentian has low competitive time some 19 years following the ces- abies) pictured typically produces a capacity and requires open soil and sation of disturbance on grassland that single whorl of branches annually, and safe sites for germination. Near the had been historically grazed. It reached

June 2014 ECOLOGICAL RESTORATION 32:2 • 193 maximum density about 25 years natural disturbances regimes. Only psammophilous grasslands within the post-disturbance, before disappearing recently have researchers and land artillery range near the of Toruń. altogether, presumably due to com- managers begun to advocate deliber- Ecological Questions 12:87–125. Csiky, J. and D. Purger. 2008. Monitoring petition. Fringed gentian became a ate implementation of heterogeneous of plant species along the Drava river. significant component of the flora of land disturbing activities as a means Pages 13–56 in J.J. Purger (ed), Bio- fallow land in Poland approximately of restoring biodiversity following the diversity Studies along the Drava River. 10–15 years after agricultural fields cessation or disruption of natural dis- Pécs, Hungary: University of Pécs. had been abandoned (Woch 2011). turbance regimes (e.g., Jentsch et al. Daniëls, F.J.A., A. Minarski and O. Lep- Shepherd’s cress, like fringed 2009, Ödman et al. 2012). ping. 2008. pattern gentian, did not show evidence of changes of a lichen-rich Corynepho- dependence on disturbance, given rus grassland in the inland of the Acknowledgements Netherlands. Annali di Botanica Roma our conservative definition of distur- We express appreciation to Debra Dale, 8:9–19. bance. Figure 6 is typical of the shep- Martin Elyn and Wolfgang of the Eichberg, C., C. and A. Schwabe. herd’s cress habitat we encountered. Installation Management Command in Hei- 2007. Endozoochorous dispersal, Although obviously disturbed at some delberg, Germany for their financial and seedling emergence and fruiting suc- point in the past, the methodology logistical support for the effort, and to the cess in disturbed and undisturbed suc- employed prevented labeling points commanding officers of theth 7 Army Train- cessional stages of sheep-grazed inland as being definitively disturbed. Shep- ing Command and the individual military sand ecosystems. Flora 202:3–26. installations for site access. European Economic Community. 1992. herd’s cress has been reported on aban- Council Directive 92/43/EEC on the doned agricultural land (Valachovič conservation of natural habitats and of 2012), agricultural set asides (Berger et References wild flora and fauna. eur-lex.europa. al. 2003), periodically disturbed gravel Beinlich, B. and P. Poschlod. 2002. Low eu/LexUriServ/LexUriServ.do? uri= bars (Gilvear and Willby 2006), aban- intensity pig pastures as an alterna- OJ:L:1992:206:0007:0050:EN:PDF. doned military training areas ( Jentsch tive approach to habitat manage- Felinks, B. and T. Wiegand. 2008. Explor- et al. 2002), and clandestine graves ment. Pages 219–226 in B. Redecker, ing spatiotemporal patterns in early stages of primary succession on (Caccianiga et al. 2012). P. Finck, W. Härdtle, U. Riecken, and E. Schröder (eds), Pasture Landscapes former lignite mining sites. Journal of On militarized landscapes, inher- and Nature Conservation. Heidelberg, Vegetation Science 19:267–276. ent landscape heterogeneity, coupled Germany: Springer. Fischer, M., M. Burkhart, V. Pasqualetto with military training doctrine, serves Berger, G., H. Pfeffer, H. Kächele, and M. van Kleunen. 2010. Exper- to create highly heterogeneous dis- S. Andreas and J. Hoffman. 2003. iment meets : Plants turbance patterns (Herl et al. 2005, Nature protection in agricultural land- of river corridor distribution are Warren and Herl 2005). Because of scapes by setting aside unproduc- not more stress tolerant but bene- the heterogeneous nature of distur- tive areas and within arable fit less from more benign conditions fields (“Infield Nature Protection elsewhere. Journal of Plant bances occurring on military train- Spots”). Journal of Nature Conservation 3:149–155. ing areas, they are among the most 11:221–233. Flather, C.H., L.A. Joyce and C.A. Bloom- biodiverse habitats in regions where Bornkamm, R. 2006. Fifty years vege- garden. 1994. Species endangerment they occur (Flather et al. 1994, tation development of a xerother- patterns in the United States. Gen- IUCN 1996, Groves, et al. 2000, mic calcareous grassland in Central eral Technical Report RM-241. U. S. Warren et al. 2007). In accordance Europe after heavy disturbance. Flora Forest Service, Rocky Mountain with the heterogeneous disturbance 201:249–267. Forest and Range Experiment Station, Burkhardt P. 1994. The Major Train- Fort Collins, Colorado. hypothesis (Warren et al. 2007), ing Areas, 4th ed. Amberg, Germany: Forey, E., B. Chapelet, Y. Vitasse, others have likewise concluded that Druckhaus Oberpfalz. M. Tilquin, B. Touzard and heterogeneous disturbance is vital Butaye, J., O. Honnay, D. Adriaens, L-M. R. Michale. 2008. The relative impor- for conserving biodiversity on non- Delescaille and M. Hermy. 2005. Phy- tance of disturbance and environmen- militarized landscapes (e.g., Vandvik tosociology and phytogeography on tal stress at local and regional scales in et al. 2005, Fuhlendorf et al. 2006, the calcareous grasslands on Devo- French coastal sand dunes. Journal of Massada et al. 2009). The challenge, nian limestone of southwest Belgium. Vegetation Science 19:493–502. Fuhlendorf, S.D., W.C. Harrell, D.M. for practitioners of ecological resto- Belgian Journal of Botany 38:24–38. Caccianiga, M., S. Bottacin and C. Catta- Engle, R.G. Hamilton, C.A. Davis ration, is to develop and implement neo. 2012. Vegetation dynamics as a and D.M. Leslie Jr. 2006. Should het- land management strategies that tool for detecting clandestine graves. erogeneity be the basis for conserva- create and maintain a heterogeneous Journal of Forensic Science 57:1–6. tion? Grassland bird response to landscape composed of a mosaic of Chojnacka, J., W. Cyzman, A. Nienar- and grazing. Ecological Applications patches of varying age and resulting towicz and M. Deptuła. 2010. Vari- 16:1706–1716. from a variety of different kinds and ability of the structure and directions Gazenbeek, A. 2005. LIFE, Natura 2000 intensities of disturbance that mimic in the development of heaths and and the Military. Brussels, Belgium:

194 • June 2014 ECOLOGICAL RESTORATION 32:2 European Commission, Environment and rabbit disturbances for seedling Ödman, A.M., T.K. Schnoor, J. Ripa and Directorate General. establishment in dry acidic grasslands P.A. Olsson. 2012. Soil disturbance Gilvear, D. and N. Willby N. 2006. dominated by Corynephorus canescens. as a restoration measure in dry sandy Channel dynamics and geomor- Phytocoenologia 32:553–580. grasslands. Biodiversity Conservation phic variability as controls on gravel Jentsch, A., S. Friedrich, T. Steinlein, 21:1921–1935. bar vegetation; River Tummel, Scot- W. Beyschlag and W. Nezadal. 2009. Pärtel, M., H.H. Bruun and M. Sammul. land. River Research and Applications Assessing conservation action for 2005. Biodiversity in temperate Euro- 22:457–474. substitution of missing dynamics pean grasslands: Origin and con- Groves, C.R., L.S. Kutner, D.M. Stoms, on former military training areas in servation. Pages 1–14 in R. Lillak, M.P. Murray, J.M. Scott, M. Shafale, central Europe. R. Viiralt, A. Linke and V. Geherman A.S. Weakley and R.L. Pressey. 2000. 17:107–116. (eds), Integrating Efficient Grassland Owning up to our responsibilities: Király, G., A. Mesterházy, A. Király, Farming and Biodiversity. Proceedings Who owns lands important for biodi- G. Pinke and R. Pál. 2008. Auftre- of the 13th International Occasional versity? Pages 275–300 in B.A. Stein, ten von Nanocyperion-Arten in Wes- Symposium of the European Grassland L.S. Kutner and J.S. Adams (eds), Pre- tungen—die Rolle der Feuchtäcker Federation. Tartu, Estonia: Estonian cious Heritage: The Status of Biodiver- in ihrer Erhaltung. Journal of Plant Grassland Society. sity in the United States. New York, Diseases and Protection, Special Issue Poschlod, P., P. Karlík, A. Baumann and NY: Oxford University Press. 21:413–418. B. Wiedmann. 2008. The history of Hasse, T. and F.J.A. Daniëls. 2006. Species Kooijman, A.M. and M.W.A. de Hann. dry calcareous grasslands near Kall- responses to experimentally induced 1995. Grazing as a measure against münz (Bavaria) reconstructed by the habitat changes in a Corynephorus grass encroachment in Dutch dry application of paleoecological, his- grassland. Journal of Vegetation Science dune grassland: effects on vege- torical and recent-ecological meth- 17:135–146. tation and soil. Journal of Coastal ods. Pages 130–143 in P. Szabo and Herl, B.K., W.W. Doe III and D.S. Jones. Conservation 1:127–134. R. Hedi (eds), Human Nature: Studies 2005. Use of military training doc- Łabuz, T.A. and R. Grunewald. 2007. in Historical Ecology and Environmen- trine to predict patterns of maneu- Studies on vegetation cover of the tal History. Pruhonice, Czech Repub- ver disturbance on the landscape. I. youngest dunes of the Świna Gate lic: Institute of Botany of the Czech Theory and methodology. Journal of Barrier (western Polish coast). Journal Academy. Terramechanics 42:353–371. of Coastal Research 22:16–172. Řehouková, K. and K. Prach. 2008. Husáková, J. 1996. Ruderal vegeta- Lomba, A., P. Alves and J. Honrado. 2008. Spontaneous vegetation succes- tion on the Šumava National Park: Endemic sand dune vegetation of the sion in gravel-sand pits: A poten- Preliminary results. Silva Gabreta northwest Iberian peninsula: Diversity, tial for restoration. Restoration Ecology 1:115–121. dynamics, and significance for bio- 16:305–312. Isselstein, J., B. Jeangros and V. Pavlu. indication and monitoring of coastal Rosenbauer, A. 1996. Gentianaceae. 2005. Agronomic aspects of biodiver- landscapes. Journal of Coastal Research Pages 16–42 in O. Sebald, S. Seybold sity targeted management of temper- 24:113–121. and G. Philippi (eds), Die Farn- und ate grasslands in Europe—A review. Marshall, J.K. 1965. Corynephorus cane- Blütenpflanzen Baden-Württembergs, Agronomy Research 13:139–151. scens (L.) P. Beav. as a model for vol. IV. Stuttgart, Germany: Ulmer. IUCN (International Union for the Con- the Ammophila problem. Journal of Rothmaler, W. 1994. Exkursionsflora von servation of Nature). 1996. Tanks and Ecology 53:447–463. Deutschland, Gefäßpflanzen. 15th edi- Tyme—Biodiversity in Former Soviet Massada, A.B., Y. Carmel, G. Koniak and tion. Jena, Germany: Verlag Gustav Military Areas in Central Europe. I. Noy-Meir. 2009. The effects of dis- Fischer. Gland, Switzerland: IUCN. turbance based management on the Salisbury, E. 1970. The pioneer vegeta- Jentsch, A. 2007. Disturbance man- dynamics of Mediterranean vege- tion of exposed mud and its biologi- agement—Application of ecolog- tation: a hierarchical and spatially cal features. Philosophical Transactions ical knowledge to habitat restora- explicit modeling approach. Ecological of the Royal Society of London. Series B, tion. Pages 144–160 in W. Barth- Modelling 220:2525–2535. Biological Sciences 259:207–255. lott, K.E. Linsenmayr and S. Poremb- Matevski, V. and M. Kostadinovski. 2009. Schulze-Motel, W. (ed). 1986. Crucife- ski (eds), Biodiversity: Structure and New locality of species Limosella rae. Illustrierte Flora von Mitteleuropa. Function, vol. II. Oxford, UK: EOLSS aquatica L. in the flora of the Republic Vol. IV, 3rd edition. Berlin, Germany: Publishers. of Macedonia. Hladnikia 23:77–80. Parey Buchverlag. Jentsch, A. and W. Beyschlag. 2003. Vege- Oberdorfer, E. 1990. Pflanzensoziolgische Tschöpe, O. and K. Tielbörger. 2010. The tation ecology of dry acidic grasslands Exkursionsflora, 6th ed. Stuttgart, role of successional stage and small- in the lowland area of central Europe. Germany: Ulmer-Verlag. scale disturbance for establishment Flora 198:3–25. Ödman, A.M., L-M. Mårtensson, of pioneer grass Corynephorus canes- Jentsch, A. and C. Beierkuhnlein. 2008. C. Sjöholm and P.A. Olsson. 2011. cens. 2010. Applied Vegetation Science Research frontiers in change: Immediate response in soil chemis- 13:326–335. Effects of extreme meteorological try, vegetation and ground beetles to Tschöpe, O., D. Wallschläger, M. Burkart events on ecosystems. Geoscience soil perturbation when implemented and K. Tielbörger. 2011. Manag- 340:621–628. as a restoration measure in decal- ing open habitats by wild ungulate Jentsch, A., S. Friedrich, W. Beyschlag and cified sandy grassland. Biodiversity browsing and grazing: a case-study W. Nezadal. 2002. Significance of ant Conservation 20:3039–3058.

June 2014 ECOLOGICAL RESTORATION 32:2 • 195 in North-Eastern Germany. Applied the landscape. II. Validation. Journal of Lower Lusatia, Germany. Applied Vegetation Science 14:200–209. of Terramechanics 42:373–381. Vegetation Science 4:5–18. Ujházy, K., J. Fanta and K. Prach. 2011. Warren, S.D. and R. Büttner. 2008a. Woch, M.W. 2011. Xerothermic veg- Two centuries of vegetation succession Active military training areas as refugia etation of fallow lands in west- in an inland sand dune area, central for disturbance-dependent endangered ern Małopolska. Annales Universita- Netherlands. Applied Vegetation Science insects. Journal of Conservation tis Mariae Curie-Skłodowska. Sectio C 14:316–325. 12:671–676. 66:105–120. Valachovič, M. 2012. Succession model Warren, S.D. and R. Büttner. 2008b. with Corynephorus canescens in aban- Relationship of endangered amphibi- doned sand fields (W Slovakia). ans to landscape disturbance. Journal Hacquetia 11:5–15. of 72:738–744. Vandvik, V., E. Heegaard, I.E. Mären and Warren, S.D., S.W. Holbrook, D.A. Dale, Steven D. Warren (corresponding author), P.A. Aarrestad. 2005. Managing het- N.J. Whelan, M. Elyn, W. Grimm US Forest Service, Rocky Mountain erogeneity: The importance of grazing and A. Jentsch. 2007. Biodiversity and Research Station, 735 North 500 East, and environmental variation on post- the heterogeneous disturbance regime Provo, UT 84606–1856, swarren02@ fire succession in heathlands. Journal on military training lands. Restoration fs.fed.us. of 42:139–149. Ecology 15:606–612. Warren, S.D. and B.K. Herl. 2005. Use of Wiegleb, G. and B. Felinks. 2001. Pre- Reiner Büttner, Institut für Vegeta- military training doctrine to predict dictability of early stages of primary tionskunde und Landschaftsökologie, patterns of maneuver disturbance on succession in post-mining landscapes Georg-Eger-Straße 1b, 91334 Hemhofen, Germany.

Gentianella ciliata. Sturm, J.G. 1796. Deutsch- lands Flora in Abbildungen. Nürnberg, DE: Gedruckt auf Kosten des Verfassers.

196 • June 2014 ECOLOGICAL RESTORATION 32:2