Excursion Guide 18th IMCG Field symposium and General Assembly 2018

The Dutch experiences in restoration ecology and challenges for the future

August 20–31 /2018 Colofon

Excursion guide and book of abstracts of the IMCG symposium (22nd Aug 2018 Texel)

18th IMCG Field Symposium and General Assembly 2018 20 – 31 August 2018.

Authors (in alphabetical order) Ab Grootjans Andre Jansen Annemieke Kooijman Boudewijn Beltman Casper Cusell Christan Hoetz Enno Bregman Gert Jan Baaijens Gert Jan Van Duinen Henk Everts Iwan Mettrop Jos Verhoeven

Editor Ab Grootjans

Design Gert Jonker BNO, Apeldoorn

Sponsors OBN/VBNE ERA Foundation

Organizers Jos Verhoeven Ab Grootjans Andre Jansen Annemieke Kooijman Gert-Jan van Duinen Ella de Hullu

2 Contents

The Dutch experiences in restoration ecology and challenges for the future...... 4

th Schedule of the 18 IMCG Field excursions...... 5

nd IMCG Symposium 22 August NIOZ Texel...... 7

The Hors on the Wadden Sea Island of Texel...... 8

The Wyldlannen Friesland...... 16

The Drentsche Aa National Landscape Park...... 20

The Dwingelderveld...... 28

The National Park Weerribben-Wieden...... 35

Restoration of bog relics in the Netherlands...... 41

Bargerveen...... 44

Wierdense Veld...... 48

Aamsveen...... 51

The Polder Westbroek...... 55

Peat meadows in The Netherlands...... 63

3 ThTeh De uDtucthc hex epxepreierniecnecse isn inre rsetostroartaiotnio nec eoclogloyg ayn adn dch cahllaelnlegnegse fso fro trh teh feu ftuutruer e

TheThe Netherlands Netherlands is isdensely densely populated populated and and the the development development of of infrastructures infrastructures and and intensificationintensification of of agriculture agriculture in inthe the sixties sixties and and seventies seventies of of the the last last century century has has destroyeddestroyed or or damaged damaged most most of of our our existing existing nature nature reserves. reserves. But But this this has has also also triggeredtriggered much much practical practical and and fundamental fundamental research research aimed aimed at at understanding understanding the the mechanismsmechanisms of of wet wet ecosystem ecosystem decline decline in inthe the Netherlands. Netherlands. During During the the eighties eighties andand nineties nineties the the government government decided decided to to buy buy out out farmers farmers in inareas areas where where modern modern agricultureagriculture was was no no longer longer viable viable and and hand hand them them over over to to nature nature conservation conservation The Dutch experiencesorganizations.organizations. in restoration Such Such areas areas were were then then transformed transformed into into nature nature areas, areas, by by raising raising waterwater levels levels and and decreasing decreasing nutrient nutrient availability. availability. During During that that period period much much ecology and challengesknowledgeknowledge for was wasthe developed developed future in inrestoring restoring wet wet ecosystems ecosystems that that had had been been in informer former agriculturalagricultural use. use.

WeWe aim aim to to show show the the participants participants areas areas where where scientific scientific and and practical practical knowledge knowledge in inDutch Dutch restoration restoration areas areas has has developed. developed.

he Netherlands is densely populated and the development of infrastructures and intensification of agriculture in the sixties T and seventies of the last century has destroyed or damaged most of our existing nature reserves. But this has also triggered much practical and fundamental research aimed at understanding the mechanisms of wet ecosystem decline in the Netherlands. During the eighties and nineties the government decided to buy out farmers in areas where modern agriculture was no longer viable and hand them over to nature conservation organizations. Such areas were then transformed into nature areas, by raising water levels and decreasing nutrient availability. During that period much knowledge was developed in restoring wet ecosystems that had been in former agricultural use.

We aim to show the participants areas where scientific and practical knowledge in Dutch restoration areas has developed.

AreaAreaArea thatthat that willwill will be be visitedbe visited visited during during during the excursion. the the excursion. excursion.

5 5

4 Schedule of the 18th IMCG Field excursions

Detailed planning (28th July 2018): Thursday 23 August 08.30-9.30: Bus Travel from Den Burg to Texel Monday 20 August Harbour, Ferry to Den Helder Harbour. Participants travel to Texel on their own. Participants 10.00-11.00: Travel by bus from Den Helder Harbour to arriving at Amsterdam Schiphol Airport take the train Eernewoude (Friesland). to Den Helder (close to Texel). From there take the 12.00-16.00: Boat trip (lunch on board) to wet meadow shuttle bus to the Harbour and take the ferry boat to reserve surrounded by deeply drained polders. Guidance Texel. After arrival on Texel take the local bus to the by local managers and experienced researchers that town of Den Burg, which stops opposite of Stayokay (our have studied the area. accommodation on Texel). Adress: Haffelderweg 29, 1791 16.00-17:00: Travel to Havelte (NIVON Hunehuis; AS Den Burg. We will later provide a detailed travelling Hunebeddenweg 1, 7973 JA Darp). scheme. Travellers by car can park the car at the harbour 19.00: Dinner (close to Hunehuis). for free. 19.00: Dinner at Stayokay Den Burg. Friday 24 August 08.30-09:30: Travel to village Oudemolen (Drentsche Aa Tuesday 21 August National Park). 08.30: Departure with local busses (Texelhopper) to 09.30-10.30: Coffee stop at OudeMolen regional office De Slufter, a near natural salt marsh and dune area. We of Staatsbosbeheer. Introduction into Geohydrology by will have a three hours walk through the area guided by Enno Bregman. experienced researchers and managers. We will take a 10.30-13.00: Two stops in the National Park Drentsche bagged lunch from Stayokay. Aa (De Heest). Guidance by local managers and 12.30-13.30: Coffee at Restaurant De Slufter, eat lunch. experienced researchers that have studied the area. 13.30: Departure with local busses to De Hors (De Geul). 13.00-14.00: Travel to National Park Dwingelderveld 14.00- 17.00: Excursion through De Hors with natural (Spier). Lunch packet in the bus. dune formation and dune slacks of different age. The 14.00-17.30: Three-hour walk in the N. P. excursion will be guided by experienced researchers Dwingelderveld, the large wet heathland reserve in that have studied the area. Europe. Discussion on hydrology of small heathland 17.00: Local bus back to Stayokay. bogs. Guidance by local managers and experienced 19.00: Dinner at Stayokay. researchers that have studied the area. 17.30- 18.00: Travel from Dwingeloo to Havelte Wednesday 22 August (Hunehuis). 08.15: Local bus to Harbour, 10 minutes-walk to 19.00: Dinner (close to Hunehuis). Research Institute NIOZ. 8.45: Start IMCG-symposium at NIOZ. Program will Saturday 25 August be provided later. 08.30-9.00: Travel to Ossenzijl/Kalenberg (Overijsel). 12.30-13.30: Lunch at NIOZ. (address: Hoogeweg 27, 8376 EM Ossenzijl). 13.30-17.00: IMCG-symposium. 09.00-11.00: Two-hour walk at Stobberibben a well- 18.00: Local bus from Harbour to Stayokay in Den Burg. studied terrestrialization fen in NW-Overijsel. Guidance 19.00: Dinner in Restaurant De Lindenboom, Den Burg. by local managers and experienced researchers that have studied the fens here. 11.30-12.00: Travel to Wanneperveen.

5 12.00-16.00: Field excursion to Veldweg, Kiersche Wednesday 29 August Wijde and Meppelerdiep, restored terrestrialization fens 08.00-10.00: Travel from Zwartemeer to Tienhoven and flood meadows. Guidance by local managers and (Utrecht). experienced researchers that have studied the fens. 10.00-10.30: Coffee break in Cafe Het Olde Regthuys, 16.00-16.30: Travel from the town of Meppel to Tienhoven Havelte. 10.30-13.30: Three-hour walk in Westbroek mires, 18.00: Dinner (close to Hunehuis). starting from Bert Bos-pad. The results of restoration works in fen area will be shown. Guidance by local Sunday 26 August managers and experienced researchers that have 09.30-10.30: Travel to the village of Zwartemeer. studied the fens. 10.30-11.30: Coffee-stop at the visitors centre of the 13.30-14.30: Trip by bus to see the surrounding national nature conservation agency (SBB). landscape including 6 meter deep former agricultural 11.30-15.30: Four-hour walk through the restored bog polder area (Betune Polder) that has been flooded complex Bargerveen, which is one of the best studied and and is now a drinking water reservoir for the city of most expensive restoration objects in the Netherlands. Amsterdam. Lunch in the bus. Guidance by local managers and experienced 14.30-16.30: Boat tour (two boats) at Nieuwkoopse researchers that have studied this bog. Plassen. 15.30-16.00: Travel to accommodation near 16.30-17.30: Travel to accomodation in Utrecht Zwartemeer (Sportlandgoed; Verlengde van (Stayokay, Neude 5, Utrecht). Echtenkanaal NZ 2, 7894 EA Zwartemeer). 18.30: Dinner at Stayokay in Utrecht. 18.30: Dinner at Sportlandgoed. Thursday 30 August Monday 27 August 08.30-9.30: Travel from Utrecht to the town of Zegveld 08.30-09.30: Travel to Wierdense Veld. where we visit an experimental research station on land 09:30-12.00: Field excursion to another large bog use of peat soils (VIC; Oude Meije 18, Zegveld). remnant (Wierdense Veld) that is well studied and where 09.30-12.30: Mini-symposium on soil subsidence restoration measures have been taken since several due to agricultural drainage and possible solutions. decades. Guidance by local managers and experienced 12.30-13.30: Lunch at experimental station Zegveld. researchers that have studied this bog. 13.30-14.30: Short visit to experimental rewetting site 12.00-13.00: Travel to Glanerbrug. 14.30 Travel back to Accommodation at Stayokay 13.00-16.30: Field excursion to Aamsveen, a partly Utrecht (Neude 5). restored bog at the border with Germany. Guidance by 18.00: Dinner at Stayokay Utrecht. local managers and experienced researchers that have 20.00: Evening walk through the inner city of Utrecht. studied this bog. Guides: Peter Grootjans, Llewellyn Bogaers, Jos 16.30-17.30: Travel back to accommodation at Verhoeven. Zwartemeer. 18.30: Dinner at Sportlandgoed. Friday 31 August 08.30-9.00: Walk to venue General Assembly IMCG, Tuesday 28 August Academiegebouw, Domplein Utrecht. 08.30-09.00: Travel to first field excursion stop. 09.30-12.30: General Assembly IMCG. 09.00-16.00: Field excursion to Reest and Hunze river 12.30- 13.30: Lunch. valleys, with species-rich meadows and fen vegetation 13.30-14.30: Back to Stayokay Utrecht or drop of (Reest valley; Schrapveen) and re-development of fen participants at Central (train) Station Utrecht. This is meadows and mires from intensive agriculture lots in a for participants that will travel home today. Utrecht river valley (Hunze valley).Visit to LOFAR telescopes in Central Station has a direct connection to Schiphol restoration area. Airport Amsterdam. 16:00: Travel back to accommodation at Zwartemeer. 18.00: Dinner at Sportlandgoed. 14.30: End of IMCG field trip and General Assembly

6 IMCG Symposium 22nd August NIOZ Texel

Time Speaker Title 8.45-9.00 Opening Symposium 9.00-9.20 Stephan Glatzel Ecosystem dynamics in the Pürgschachen Bog, Austria 9.20-9.40 Tanya Lippmann Vegetation and vegetation management play a role on the net greenhouse gas budget of rewetted peatlands 9.40-10.00 Weier Liu Estimating Greenhousgas emission from vegetation maps after rewetting (Drentsche Aa valley, NL) 10.00-10.20 Samer Elshehawi Natural isotopes identify groundwater flows in the Drentsche Aa Brook Valley, The Netherlands 10.20-10.40 Timofey Orlow Regional peat depth estimation measured with GPR 10.40-11.00 coffee break 11.00-11.20 Zhao-Jun Bu Effect of simulated herbivory on the performance of two peat mosses 11.20-11.40 Beverley Clarkson Resilience of a raised bog remnant impacted by lowered water tables and nutrient inputs 11.40-12.00 Anders Lyngstad From raised bogs to razed bogs – threatened lowland mires in Norway 12.00-12.20 Ulrich Graf & Effects of cattle exclosure and partial rewetting on bog vegetation: the case of Angéline Bedolla Schwaendital Bog 12.20-12.40 Marina Abramchuk Water regime of Zvaniec mire - implications for nature conservation management 12.40-13.00 Agata Klimkowska/ Microbial communities in the soils in natural and restored fen peatlands – relevance Wiktor for restoration 13.00-14.00 lunch 14.00-14.20 Piet-Louis Grundling Building resilience into South Africa's peatlands to couple with climate change: is it an option? 14.20-14.40 Eric Munzhedzi Progress towards peatland conservation in NW Province, South Africa: setbacks and successes 14.40-15.00 Jason Le Roux The wetland inventory of Swaziland: reporting on the first peatland discoveries 15.00-15.20 Siew Yan Lew Peatland Management in Southeast Asia 15.20-16.00 thea break 16.00-16.20 Jan Peeters & Paludiculture in the Baltics Andreas Haberl 16.20-16-40 Tapio Lindholm The Finnish-Russian Working Group on Nature Conservation; cooperation in mire protection 16.40-17.00 Monica Sofia Peruvian High Andean Peatlands Maldonado 17.00-17.20 Aaron Pérez-Haase Pyrenean mires in protected areas: main threats and conservation actions 17.20-17.40 Eulàlia Pladevall- Response of engineering plant species to experimental conditions in pryenean fens Izard 18.00 Depature Poster presentations Country Author Title South Africa Althea Grundling Peatlands under pressure Latvia Mara Pakalne Peatlands in Latvia Germany Michael Trepel Challenges for water Management in German Peatlands Germany Tatiana Minayeva Indication of peatland restoration success: examples and challenges Netherlands Gert-Jan van Duinen Paludiculture in the Deurnsche Peel

7 The Hors on the Wadden Sea Island of Texel; dune slackThes with rare fen species Hors on the Wadden. Sea Island of Texel; dune slacks with rare fen species Ab Grootjans, Annemieke Kooijman

Tuesday 21stAb Aug Grootjans,ust 08.30-17.0 Annemieke0 Kooijman Tuesday 21st August 08.30-17.00

The Hors on the island of Texel he Hors on the south-eastern tip of the island of Texel (52°59’N, 4°44’E) is a very young and almost Tnatural dune area with a very broad beach plain (Figure 1), enabling the natural formation of embryonic dunes (Figure 2). The newly formed dune ridges can enclose parts of the beach, thus preventing flooding with sea water. In groundwater discharge areas fresh water dune slacks develop, which have a vegetation type that resembles that of rich fens (Lammerts & Grootjans 1998). Some of these slacks are not entirely natural, as they are enclosed by several sand dikes that were created artificially by the coastal protection agency (RWS) in the mid-1970’s. But The Hors now has a series of parallel dune slacks of different ages, covering a total live span of a few hundred years (Ballarini et al. 2003, Oost et al. 2012, see also figure 3). Apart from an very interesting geomorphology

Figure 1. Impressions of The Hors on the island of Texel the Hors at the Wadden Sea island of Texel.

The Hors on the south-eastern tip of the island of Texel (52°59’N, 4°44’E) is a Upper left: view from very young and almost natural dune area with a very broad beach plain (Figure the dunes to the Hors 1), enabling the natural formation of embryonic dunes (Figure 2). with embryonic dune in the background. Upper The newly formed dune ridges can enclose parts of the beach, thus preventing right; primary slack flooding with sea water. enclosed by a young dune rich. Lower left: very young dune slack

(2 years old) with some precipitation of iron, 11 indicating groundwater influence. Lower right: flowering individuals of Liparis loeselii.

Figure8 1. Impressions of the Hors at the Wadden Sea island of Texel. Upper left: view from the dunes to the Hors with embryonic dune in the background. Upper right; primary slack enclosed by a young dune rich. Lower left: very young dune slack (2 years old) with some precipitation of iron, indicating groundwater influence. Lower right. Flowering individuals of Liparis loeselii.

Hors Lakes De Hors Dune slack

Beach

Figure 2. Overview of the Hors (on the left), the Hors lake, (background) and on the left the southern part of the main island of Texel (Photo: Staatsbosbeheer 2005).

12

Figure 1. Impressions of the Hors at the Wadden Sea island of Texel. Upper left: view from the dunes to the Hors with embryonic dune in the background. Upper right; primary slack enclosed by a young dune rich. Lower left: very young dune slack (2 years old) with some precipitation of iron, indicating groundwater influence. Lower right. Flowering individuals of Liparis loeselii.

Hors Lakes De Hors Dune slack

In groundwater discharge areas fresh water dune slacks develop, which have a vegetation type that resembles that of rich fens (Lammerts & Grootjans 1998).

Some of these slacks are not entirely natural, as they are enclosed by several sand dikes that were created artificially by the coastal protection agency (RWS) in the mid -1970’s. But The Hors now has a series of parallelBeach dune slacks of different ages, covering a total live span of a few hundred years (Ballarini et al. 2003, Oost et al. 2012, see also figure 3). Figure 2. Overview of the Hors (on the left), the Hors lake, (background) and on the left the southern part of the main island of TexelFigure (Photo: 2. Staatsbosbeheer Overview 2005).of the Hors (on the left), the Hors lake, (background) and on the left

the southern part of the main island of Texel (Photo: StaatsbosbeheerFigure 3. Overview 2005). of the southern part of Texel, showing the dune ridges of various age, starting in the 13th century until now (the 12 Hors in the south. After Jager & Kikkert (1998) Staatsbosbeheer 2014). In the 17th century the southern part of Texel was a natural harbour, where the bulk of the Dutch fleet was stationed.

this dune area is also famous for the occurrence of United States and Canada, to northern and Central thousands of individuals of the typical fen species European countries, Russia and even several localities Figure 3. Overview of the southern part of Texel, showing the dune ridges of various age, Liparis loeselii, which is a small orchid which occurs on in Siberia. It has been listed in Annex II and Annex IV starting in the 13th century until now (the Hors in the south. After Jager & Kikkert (1998) Staatsbosbeheerall Dutch Wadden 2014). Sea In islands the 17 andth century also on the the German southern partof the of Council Texel Directivewas a natural 92/43/EEC on the Conservation harbour,island where of Borkum. the bulk It thrivesof the inDutch wet dune fleet slacks was onstationed. of natural habitats, thus making it a priority species for mineral soils (Jones & Etherington 1992, Lammerts & conservation in most European countries. Apart fromGrootjans an 1998,very Oostermeijer interesting & Hartmangeomorphology 2014). this dune area is also famous for the occurrenceIt is much of better thousands known as of a veryindividuals rare and highly of the typicalDune fen slackspecies development Liparis loeseliiprotected, which orchidis a smallof low-productive, orchid which rich fens occurs (Wheeler on all DutchIn order Wadd to estimateen Sea the islandsage of dunes and dune slacks and alsoet al. on 1998, the Wotavová German et al.island 2004, Pawlikowskiof Borkum. 2008, It thrives inin thewet Hors dune area, slacks we used on information of aerial photos mineralOostermeijer soils (Jones & Hartman & Etherington 2014). Liparis 1992, loeselii Lammerts has a &(Figure Grootjans 4), as well 1998, as topographic -and vegetation maps Oostermeijerwide distribution, & Hartman ranging 2014). from the northeast of the (Sharhudin 2014, Kooijman et al. 2016).

It is much better known as a very rare and highly protected orchid of low- productive, rich fens (Wheeler et al. 1998, Wotavová et al. 2004, Pawlikowski 9 2008, Oostermeijer & Hartman 2014). Liparis loeselii has a wide distribution, ranging from the northeast of the United States and Canada, to northern and Central European countries, Russia and even several localities in Siberia. It has been listed in Annex II and Annex IV of the Council Directive 92/43/EEC on the

13

Conservation of natural habitats, thus making it a priority species for conservation in most European countries.

Dune slack development

In order to estimate the age of dunes and dune slacks in the Hors area, we used information of aerial photos (Figure 4), as well as topographic -and vegetation maps (Sharhudin 2014, Kooijman et al. 2016).

able to colonize newly formed slacks at very early stages. Two years after a slack had become vegetated, L. loeselii was already able to establish a small population.

50

Figure 4. Map shows sampling locations in dune slack of 40 Figuredifferent 4 ages. Map at the shows southern sampling part of Texel. locations in dune slack of different ages at the southern part of Texel. 30 Occurrence of Liparis loeselii in dune 20 Occurrenceslack succession of seresLiparis loeselii in dune10 slack succession seres Number of individuals/m2 Oostermeijer & Hartman (2014) found that the 0 Oostermeijer & Hartman (2014) found that the population0 2 life12 span15 of the18 species34 population life span of the species in coastal dune Population age (years) in coastal dune slacks was generally very short, between 5-15 years. Favorable slacks was generally very short, between 5-15 years. 1-leaf 2-leaf Adults conditions recorded for L. loeselii populations include a high pH (6-7) combined Favorable conditions recorded for L. loeselii populations with low availability of nutrients (WheelerFigure 5 .Figure1998).Population 5. Population structure On theof structure L. loeselii Dutch ofaccording L. loeseliiand to ageaccordingGerman of population to age on the Hors, include a high pH (6-7) combined with low availabilityTexel. Errorof populationbars in the bar on graphthe Hors, indicate Texel. standard Error deviation.bars in the 1- leafbar representgraph young Waddenof nutrients Sea (Wheeler islands 1998). On such the Dutch conditions and German areimmature present individuals. when 2-leaf individual exfiltrating are also young, groundwater but have not yet flowered. is Adults have flowers.indicate standard deviation. 1-leaf represent young immature presentWadden Sea with islands relatively such conditions high are concentrations present when individuals. of calcium 2-leaf individualand bicarbonate are also young, but (Grootjans have not yet Sometimesflowered. the orchid Adults was haveable toflowers. establish its population even before the etexfiltrating al. 2015). groundwater is present with relatively highformation of a slack was completely finished. This indicates that L. loeselii is a concentrations of calcium and bicarbonate (Grootjanswell -disperseda well-dispersed species, and species, can colonize and cana new colonize site easily, a new as suggested site by Jones (1998). The short window of opportunity for L. loeselii during natural vegetation Ouret al. 2015).results showed that the window ofsuccession opportunityeasily, in the as Hors suggested areafor couldL. by Jonesloeselii only be (1998). extended to The establish modestlyshort window (ca. a 5 -10 years) by populationOur results showedwas relatively that the window short, of opportunity less thana management 20of opportunity years regime (Figureof for annual L. loeselii mowing 5) during, butin the natural thisoldest dunespeciesvegetation slacks. wasfor L. loeseliiable to establish take advantage a population was of relatively this short In comparison, window.succession populations inPopulations the onHors a beach area couldplain of thison only the beorchidGerman extended Wadden were Sea island short, less than 20 years (Figure 5), but this species of Borkummodestly survived (ca. for 5-10more years) than 30 by years a management without any regimemanagement of intervention (Petersen 2003). This species was first seen at this German site in was able to take advantage of this short window. 198514 and annual a population mowing with in many the oldest individuals dune still slacks. exists there today (3,000- Populations of this orchid were able to colonize newly10,00 0; GrootjansIn comparison, et al. 2017). populations on a beach plain on formed slacks at very early stages. Two years after a Hydrologicalthe German systems Wadden analyses Sea island of Borkum survived

slack had become vegetated, L. loeselii was already ableThe isohypsefor more map thanof August 30 years 2013 without for the southern any management part of Texel (Figure 6), to establish a small population. indicatesintervention that the dune (Petersen slacks in the 2003). Hors This areas species are supplied was first with groundwater from rather local hydrological systems, with rather thin groundwater lenses (5-10 Sometimes the orchid was able to establish its m). Thisseen can beat concludedthis German from site the inchemical 1985 and composition a population of the with groundwater at population even before the formation of a slack was differentmany depths individuals (Figure 7). still exists there today (3,000-10,000; completely finished. This indicates thatL. loeselii is The chlorideGrootjans and calcium et al. 2017).concentrations have a similar pattern. Low values were found in the young dunes and high values at greater depths and under the high

10 15 dunes at the right of figure 7. The low values in the young dunes can be explained by the scarce vegetation cover of these dunes (see Figures 1 and 2).

Figure 6. Isohypse map of the phreatic groundwater table August-September 2013) at the south-east part of Texel (the Hors). The yellow dotted line represents one of the transect studies in the hydrological research

Figure 7. Ca and Cl concentrations in the shallow groundwater in transect S-N on Texel, de Hors), measured in August 2013. MSL = Mean sea water level. Conceptual model of groundwater flow in the Hors area (Texel), based on the calcium and chloride concentrations in the Figure 6. Isohypse map of the phreatic groundwater table August-September 2013) at Hydrological systems analyses shallow groundwater. the south-east part of Texel (the Hors).The yellow dotted line represents one of the transect Thestudies isohypse in the map hydrological of August 2013 research.for the southern part Ca (mg/l) of Texel (Figure 6), indicates that the dune slacks in Hors Lakes the Hors areas are supplied with groundwater from rather local hydrological systems, with rather thin Non-vegetated dunes trap less salt spray than vegetated ones. That could groundwater lenses (5-10 m). This can be concluded explain the low chloride concentrations. Also the production of CO2 in the root from the chemical composition of the groundwater at zone is very low and therefore, little calcium can be dissolved in the calcareous different depths (Figure 7). < 50 dunes. The chloride and calcium concentrations have a similar 51-100 > 100 2m The hydrologicalpattern. Low system values were analysis found in thealso young showed dunes. andthat several dune slacks in the Clay 200m Hors areahigh functioned values at greater as depths“flow -andthrough under the lakes” high dunes (Stuyfzand & Moberts 1987; Figs. 7 and 9). atThe the westernright of figure Hors 7. The lake, low valuesfor instance in the young is permanently flooded, but is receivingdunes groundwater can be explained from by the the scarce northern vegetation and cover north -westernCl (mg/l) dunes, while it suppliesof the these eastern dunes (see Hors Figures lake 1 and and 2). several southern dune slacks with groundwaterNon-vegetated for most dunes of the trap year less salt (Figure spray than 7 and 8). vegetated ones. That could explain the low chloride A flow-through lake is an open water wetland that receives groundwater from concentrations. Also the production of CO2 in the root one side and loses (surface-) water at the other end (see also Stuyfzand 1993, zone is very low and therefore, little calcium can be < 50 Grootjansdissolved et al. in 2 the014 calcareous). Figure dunes. 9 illustrates the functioning of51- 100a through-flow dune slack. The hydrological system analysis also showed that > 100 Clay several dune slacks in the Hors area functioned as ‘flow-

through lakes’ (Stuyfzand & Moberts 1987; Figs. 7 and 9). The western Hors lake, for instance is permanently 6 Height above flooded, but is receiving groundwater from the northern mean sea level (m)

and north-western dunes, while it supplies the eastern Hors lake and several southern dune slacks with 0 16 groundwater for most of the year (Figure 7 and 8).

A flow-through lake is an open water wetland that

receives groundwater from one side and loses (surface-) water at the other end (see also Stuyfzand 1993,

Grootjans et al. 2014). Figure 9 illustrates the functioning of a through-flow dune slack.

Figure 7. Ca and Cl concentrations in the shallow groundwater in 11transect S-N on Texel, de Hors), measured in August 2013. MSL = Mean sea water level. Conceptual model of groundwater flow in the Hors area (Texel), based on the calcium and chloride concentrations in the shallow groundwater.

17

E. Hors Hors lakesMere North Sea E. Hors Hors lakesMere North Sea

Figure 8. PhotographFigure 8. Photograph of a primary of a duneprimary slack dune covered slack covered with with common FigureReed 10.(Phragmites Photograph of a mini-dune slack (2 meters in australis). Thecommon yellow Reed dotted (Phragmites line shows australis). the samplingThe yellow sitesdotted of line the hydrologicaldiameter) of research one year old showing iron precipitation on Figure 8 . Photograph of a primary dune slack covered with common Reed (Phragmites shows the sampling sites of the hydrological research (figure the left side, pointing to exfiltration of anoxic and iron rich australis).(figure 6 and The 7 yellow). dotted line shows the sampling sites of the hydrological research 6 and 7). groundwater. (figure 6 and 7).

Why is anoxic groundwater important

for the stability of pioneer stages?

Anoxic groundwater in the root zone of plant can be very harmful for many plant species. Most species are not adapted to high concentrations of reduced

iron, manganese and sulfide (Lamers et al. 2015). Several pioneer species in dune slacks have special

adaptation for prolonged flooding in the wet period, which enable them to dominate for quite some time. Schoenus nigricans and Littorella uniflora, for instance O2 REDOX O2 REDOX have well-developed aerengyma, through which 0 0 0 0 20 O2 REDOX 20 20 O2 REDOX 20 oxygen is actively transported to the tip of the roots, 400 040 040 040 2060 2060 2060 2060 where leakage of oxygen occurs. This phenomenon 4080 4080 4080 4080 h ) ( th d 60100 60100 (mm) depth ) ( th d 10060 60100 (mm) depth 80120 80120 12080 80120 is called ‘radial oxygen loss’ (ROL; Armstrong 1975,

h ) ( th d 100140 100140 (mm) depth ) ( th d 100140 100140 (mm) depth 0 200 400 600 -200 0 200 0 100 200 300 0 200 400 600 -200 0 200 0 100 200 300 120 120 120 120 Adema et al. 2005). Anoxic condition thus can prevent µmol l-1 mV µmol l-1 µmol l-1 mV µmol l-1 140 140 140 140 0 200 400 600 -200 0 200 0 100 200 300 0 200 400 600 -200 0 200 0 100 200 300 the establishment of later successional species, such µmol l-1 mV µmol l-1 µmol l-1 mV µmol l-1 Figure 9. Conceptual model of a through- flow lake (changed after Stuyfzandas grasses & (Calamagrostis Moberts epigeios and Phragmites Figure1987). 9 Groundwater,. Conceptual model rather of rich a through in iron -and flow with lake low (changed concentration after Stuyfzand of oxygen & and Moberts Figure 9. Conceptual model of a through- flow lake (changed australis) and Salix shrubs and pioneer species with sulphate (1) enters the dune slack on one site (2). When exposed to the air this water 1987). Groundwater,after Stuyfzand rather & rich Moberts in iron 1987). and Groundwater, with low concentration rather rich of oxygen and loses iron and phosphate and flows (only during wet periods) as surfaceROL watercan dominate to the the vegetation for a long time. sulphate (1) entersin iron andthe withdune low slack concentration on one site of oxygen (2). When and sulphate exposed to the air this water opposite site (closer to the sea). It infiltrates through organic layers Ademaand becomes et al. (2003a,b, deeply 2005) also discovered that loses iron and(1) phosphate enters the duneand flowsslack on (only one siteduring (2). When wet periods)exposed to as surface water to the anoxic due to reduction of sulphates. In the absence of iron in the upperLittorella layers uniflora sulfide and Schoenus nigricans can reduce the opposite site (closerthe air this to waterthe sea). loses Itiron infiltrates and phosphate through and flowsorganic (only layers and becomes deeply may accumulate in the topsoil (Adema et al. 2003a).The graphs in figure 9 illustrate anoxic due to duringreduction wet periods)of sulphates. as surface In water the absence to the opposite of iron site in the upperavailability layers sulof nutrientsfide for competitors, in particular these difference between the to sides of a dune slack. Measurements were done with may accumulate(closer in theto the topsoil sea). It ( infiltratesAdema et through al. 2003 organica).The layers graphs and in figurenitrate. 9 illustrate Dissolved ammonium that enters the slack via thesemicro difference-electrodes. between Note that the the to sidesdepth of is ain dune millimeters slack. Measurements below the soil weresurface. done with becomes deeply anoxic due to reduction of sulphates. In the groundwater and precipitation water is oxidized in the micro-electrodes.absence Note of ironthat in the the depthupper layersis in millimeterssulfide may accumulatebelow the soil surface. root zone of the ROL species and changes into nitrate. in the topsoil (Adema et al. 2003a).The graphs in figure 9 illustrate these difference between the18 to sides of a dune The low productive pioneer species only take up a slack. Measurements were done with18 micro-electrodes. Note little of this nitrate, while most of it is passing the oxic that the depth is in millimeters below the soil surface. rootzone and flows into an anoxic zone again, where denitrification occurs, which changes the nitrate in to

nitrogen gas (N2), which escapes into the atmosphere. Adema et al. (2005) hypothesized that this is a key mechanism to keep pioneer stages of dune slack stable

12 for up to 80 years, but only if the dune slack is supplied How sustainable are Liparis loeselii with sufficient anoxic groundwater, to allow flooding populations on the Dutch Wadden Sea for more than 6 month a year. islands Studies on the accumulation rate of organic On the Wadden Islands, decalcification of dunes once matter in the soil support this idea that pioneer fixed by vegetation, is a relatively rapid process, because

species in groundwater fed dune slack can keep the of the low primary CaCO3 content of 0.5% (Texel) to soil organic matter (SOM) very low for a long time. 1.2% (Schiermonnikoog). With a simple mass balance Sharhudin et al. (2104) sampled successional stages Stuyfzand (in: Grootjans et al. 2014) calculated for in dune slacks of the Dutch Wadden islands and Texel a decalcification rate of 0.56 m/century above showed that in most dune slack SOM accumulations the groundwater table and 0.15 m/century below it, increased linearly during the first 50-60 years and for a bare dune lacking aeolian inputs of calcareous then levelled off (Berendseet al.1998, Figure 11). sand, and 0.73 m/century above the groundwater table However, slacks with low productive species, such and 0.26 m/century below it, for a dune fixed by dune as Littorella uniflora, showed low accumulation rates grasses and mosses. (0.02-0.08 kg/m2/year), and persisted even over a This means that fen orchid, with a rooting depth to period80 years, of more but than only 90 ifyears the (Sharuhdin dune slack et al. is 2014). supplied withof about sufficient 0.1 m, will anoxic root in decalcified, acidified dune groundwater,In contrast, slacks to allow dominated flooding by high for productive more than 6 monthsand a already year. after 14-18 years, provided that local species, such as Phragmites australis, showed ten groundwater is steadily replenished by rainfall (thus Studiestimes higheron the accumulation accumulation rates rate (0.17- of 0.26 organic kg/m2/ matternot in exfiltrating).the soil support On Schiermonnikoog this idea this would take a thatyear) pioneer over a similar species time in periodgroundwater and comparable fed dune slackbit can longer, keep namely the soil 1,2/0,5x(14-18) organic = 34-43 years. matterannual (SOM) inundation very periods low for (176-240 a long days). time. So, Sharhudinonce et Acidificational. (2104) sampledwill rapidly follow after decalcification, successional stages in dune slacks of the Dutch Wadden islands and showed that late successional species have invaded the slack, the because the acid front is moving downwards about in most dune slack SOM accumulations increased linearly during the first 50-60 pioneer species cannot keep the nutrient availability 3-10 times faster than the decalcification front. The years and then levelled off (Berendse et al.1998, Figure 11). However, slacks low for competitors and disappear within a decade. arrival of the acid front is manifested in soil moisture with low productive species, such as Littorella uniflora, showed low accumulation The rate of SOM accumulation in wet dune slacks and groundwater by a sharp drop in pH, Ca and rates (0.02-0.08 kg/m2/year), and persisted even over a period of more than 90 is, therefore primarily controlled by plant productivity. HCO , and a sharp increase for among others Al and years (Sharuhdin et al. 2014). In contrast, slacks dominated3 by high productive species,Both above-ground such as Phragmites biomass and australisSOM accumulation, showed ten tiSiOmes2. We higher therefore accumulation conclude that decalcification and ratescan remain(0.17- very 0.26 low kg/m over a2 /year)long period over of timea similar when time periodnearly simultaneous and comparable acidification annual form an excellent inundationdune slacks periods are flooded (176 during-240 mostdays). of the So, year once and late successionalexplanation of speciesdeclining havepopulations of fen orchid invadedplants withthe adaptiveslack, the traits pioneer are able tospecies maintain cannot keepon the older nutrient dunes without availability (sufficient) low eolian inputs of forvegetation competitors succession and disappearin a pioneer stage.within a decade. calcareous sand and without exfiltration of calcareous groundwater.

Figure 11. SOM accumulation data from dune slacks in the Netherlands (open symbols) and also the dunes in Wales (black filled circles).

dominated by Salix repens, Older dune slack with high values of organic matter Betula pubescens or Phragmites accumulation are dominated australis by shrubs or trees (Salix

repens or Betula pubescence), while older dune slack with

low accumulation values are dominated by Littorella

uniflora ( from: Sharhudin et dominated by Littorella al. 2014).

uniflora

Figure 11. SOM accumulation data from dune slacks in the Netherlands (open symbols) 13 and also the dunes in Wales (black filled circles). Older dune slack with high values of organic matter accumulation are dominated by shrubs or trees (Salix repens or Betula pubescence), while older dune slack with low accumulation values are dominated by Littorella uniflora ( from: Sharhudin et al. 2014).

The rate of SOM accumulation in wet dune slacks is, therefore primarily controlled by plant productivity. Both above-ground biomass and SOM accumulation can remain very low over a long period of time when dune slacks

20

The situation in the United Kingdom demonstrates the Future impact of sea level rise importance of young dune slack habitats for the survival In natural dune systems with regular formation of of Liparis loeselii. About 90% of the UK population is dune slacks, management is not required to keep a now only found in the coastal dunes of South Wales metapopulation of L. loeselii viable over a long time and these population are rapidly declining). Dune span. However, metapopulation models have shown stabilization was one of the factors that caused this (Oostermeijer & Hartman 2014) that management to decline (Jones 1998). Recently principles of dynamic prolong the life span of individual populations, such as coastal management have been applied in Wales and mowing to maintain an open vegetation structure, can new dune slack are being formed, allowing the orchid increase the viability of a metapopulation to a modest populations to move more easily between old sites extent, because seed sources remain available for Futureand new impact sites. So, in existing of fixedsea dune level complexes, rise several decades (Grootjans et al. 2014). Future promotionimpact of secondary of sea dune slack level formation rise through Our hydrological research on Texel revealed that blow-outs is the most natural and cost-effective most dune slacks with Liparis populations were fed by In natural dune systems with regular formation of dune slacks, management is In naturalstrategy dune to maintain systems viable metapopulationswith regular of early formation calcareous of and dune anoxic groundwaterslacks, frommanagement rather small is not requiredto mid-successional to keep species. Buta metapopulationwhen no or very few hydrologicalof L. loeselii systems. viable over a long time span. not required to keep a metapopulation of L. loeselii viable over a long time span. However,new slacks m areetapopulation formed, overlap in time modelsmay be lacking have shownThese systems (Oostermeijer are vulnerable to sea level& Hartman rise, 2014) However,and m localetapopulation extinction might occur models as was described have by shownwhich (Oostermeijerleads to a narrower beach & plain, Hartman shrinking of 2014) that management to prolong the life span of individual populations, such as that managementJones (1998) for severalto prolong sites at the coast the of Wales.life spanUnder ofthe individual groundwater lenses populations, and reducing the discharge such as mowingsuch conditions, to maintain new introductions an open from population vegetation of groundwaterstructure, (Figure can 12). Sand increase nourishment the before viability of a mowing to maintain an open vegetation structure, can increase the viability of a metapopulationof other islands might to be ahelpful modest in sustaining extent, the becausethe water line seed not only sources prevents erosion remain of the beach available for metapopulationmetapopulation. to a modest extent, becauseand seeddune areas, sources but may even remain temporary increaseavailable the for several decades (Grootjans et al. 2014). several decades (Grootjans et al. 2014). discharge of groundwater to the newly-formed dune Figure 12. Hydrological system of a Wadden Sea island (left) slack, when the groundwater lens of the island is andOur simulation hydrological results of volume of freshwaterresearch lens (right) on as Texel expanding revealed in reaction that to sand mostnourishment. dune slacks with Ourinfluenced hydrological by sea level rise researchand sand nourishment on onTexel Texel, revealed that most dune slacks with Liparis populations were fed by calcareous and anoxic groundwater from rather Liparis populationsde Hors. Scenario A= were present situationfed by (dark calcareousblue), B= sea and anoxic groundwater from rather smalllevel hydrological rise of 1 meter without systems. sand nourishment (red). C = small hydrologicalsea level rise of 1 metersystems. with sand nourishment (light blue; increase of coast line by sand nourishment is 800m).

Figure14 12. Hydrological system of a Wadden Sea island (left) and simulation results of Figurevolume 12. Hydrological of freshwater system lens (right)of a Wadden as influenced Sea island by sea(left) level and rise simulation and sand results nourishment of on volumeTexel, of freshwater de Hors. Scenario lens (right) A= presentas influenced situation by sea(dark level blue), rise B= and sea sand level nourishment rise of 1 meter on Texel,without de Hors. sand Scenario nourishment A= present (red). situation C = sea (darklevel riseblue), of 1B= meter sea level with risesand of nourishment 1 meter without(light sand blue; nourishment increase of (red). coast C line = sea by sandlevel nourishmentrise of 1 meter is 800m).with sand nourishment (light blue; increase of coast line by sand nourishment is 800m).

These systems are vulnerable to sea level rise, which leads to a narrower beach Theseplain, systems shrinking are vulnerable of the groundwater to sea level lenses rise, andwhich reducing leads to the a narrowerdischarge beach of plain,groundwater shrinking of (Figure the groundwater 12). Sand lensesnourishment and reducing before the waterdischarge line ofnot only groundwaterprevents (Figureerosion 12) of the. Sand beach nourishment and dune areas,before btheut maywater even line temporarynot only increase preventsthe discharge erosion of of the groundwater beach and todune the areas,newly -bformedut may dune even slack,temporary when increasethe the dischargegroundwater of groundwater lens of the island to the is newly expanding-formed in dunereaction slack, to sandwhen nourishment. the groundwater lens of the island is expanding in reaction to sand nourishment.

22 22

Further information/further reading Lammerts EJ Grootjans AP (1998). Key environmental variables determining the occurrence and life span of basiphilous dune Adema, E.B., Van Gemerden, H. & Grootjans, A.P. (2003a). slack vegetation. Acta Botanica Neerlandica 47: 369-392. Is succession in wet calcareous dune slacks affected by free sulfide? Journal of Vegetation Science 14: 153-162 Kooijman AM, Bruin CJW, Van de Craats A, Grootjans AP, Oostermeijer JGB, Scholten R. & Shahrudin RB. (2016). Past Adema, E.B. & Grootjans, A.P. (2003b). Possible positive- and future of the EU-habitat directive species Liparis Loeselii feedback mechanisms: plants change abiotic soil parameters in in relation to landscape and habitat dynamics in SW-Texel, the wet calcareous dune slacks. Plant Ecology 167: 141-149 Netherlands. Science of the Total Environment 568: 107-117.

Adema EB, Van de Koppel J, Meijer HAJ, Grootjans AP (2005) Oost AP, Hoekstra, P., Wiersma, A, Flemming, B, Lammerts Enhanced nitrogen loss may explain alternative stable states in EJ, Pejrup, M, Hofstede, J, van der Valk, B. Kiden, P. dune slack succession. Oikos 109:374-386. Bartholdy, EJ, van der Berg MW, Vos, PC, de Vries, S. & Wang ZB. (2012). Barrier island management; lessons from Ballarini M, Wallinga J, Murray AS, van Heteren S, Oost AP, the past and directions for the future. Ocean and Coastal Bos AJJ, van Eijk CWE (2003) Optical dating of young coastal Management 68: 18-38. dunes on a decadal time scale Quaternary Science Reviews. 22:1011-1017. Oostermeijer JGB, Hartman Y (2014). Inferring population and metapopulation dynamics of Liparis loeselii from single-census Berendse F, Lammerts EJ, Olff H (1998) Soil organic matter and inventory data. Acta Oecologia 60:30-39 accumulation and its implications for nitrogen mineralization and plant species composition during succession in coastal Pawlikowski P (2008). Distribution and population size of dune slacks. Plant Ecol. 137:71-78 the threatened fen orchid L. loeselii loeselii (L.) Rich. in the Lithuanian Lake District (NE Poland). Bot. Stec.12:53-59 Davy AJ, Grootjans AP, Hiscock K, Peterson J. (2006) Development of eco-hydrological guidelines for dune Petersen J (2000). Die Dünentalvegetation der Wattenmeer-Inseln habitats – Phase 1. English Nature Research Reports, No 696, in der südlichen Nordsee. Eine pflanzensoziologische und Peterborough ökologische Vergleichsuntersuchung unter Berücksichtigung von Nutzung und Naturschutz. - 336 pp., Husum Grootjans, AP, P.J. Stuyfzand PJ H. Everts FH, De Vries NPJ. Kooijman AM, Oostermeijer JGB Nijssen M, Wouters Shahrudin, R (2014). Do we really need management to preserve B, Petersen, J & Shahrudin, R. (2014). Ontwikkeling van pioneer stages in wet dune slacks? Dissertation University of zoet-zoutgradiënten met en zonder dynamisch kustbeheer; Groningen, 104pp. een onderzoek naar de mogelijkheden voor meer natuurlijke ontwikkelingen in het kustgebied. VBNE-rapport 2014/ Shahrudin RB, Dullo, B, Woudwijk, W, de Hoop, P. & OBN193-DK. Grootjans, AP (2014). Accumulation rates of soil organic matter in wet dune slacks on the Dutch Wadden Sea Islands. Grootjans, AP, Shahrudin, RB, Van de Craats, A., Kooijman Plant and Soil, 380(1- 2): 181-191. AM, Oostermeijer, G, Petersen, J, Amatirsat, D, Bland, C. & Stuyfzand PJ (2017). Window of Opportunity of Liparis loeselii Stuyfzand PJ (1993). Hydrochemistry and hydrology of the populations during vegetation succession on the Wadden Sea coastal dune area of the Western Netherlands. PhD Thesis Free islands. Journal of Coastal Conservation 21: 631-641. University of Amsterdam, published by KIWA, ISBN 90-74741- 01-0, 366 p. Jones PS (1998) Aspects of the population biology of L. loeselii loeselii (L.) Rich. var. ovata Ridd. ex Godfrey (Orchidaceae) in the Wheeler BD, Lambley PW, Geeson J. (1998). Liparis loeselii dune slacks of South Wales, UK. Bot. J. Linn. Soc. 126:123-139. loeselii (L.) Rich, in eastern England: constraints on distribution and population development. Bot. J. Linn. Soc.126:141-158 Lamers LPM, Vile MA, Grootjans AP, Acreman, MC, van Diggelen R, Evans MG, Richardson CJ, Rochfort L., Wotavová K, Balounová Z, Kindlmann P (2004). Factors Kooijman A, Roelofs JGM & Smolders, AJP (2015). affecting persistence of terrestrial orchids in wet meadows and Ecological restoration of rich fens in Europe and North America: implications for their conservation in a changing agricultural from trial and error to an evidence-based approach. Biological landscape. Biol. Cons. 118: 271-279. Reviews 90: 182–203.

15 The Wyldlannen Friesland; an example of The Wyldlannen Friesland; an example of targets that targets that cannot be reached any more cannot be reached any more.

Ab Grootjans, AbHenk E Grootjans,verts. Henk Everts Thursday Aug 23rd 11.00-16.00 Thursday Aug 23rd 11.00-16.00

Human impact The last 50 years the surface water that is used to flood the summer polders mainly originates from the large

rivers Rhine and IJsel. This river water is polluted with

nutrients, chloride and sulfate (Van Duren et al. 1998). Due to long term drainage practises in agricultural

Polder “WyldPolderlannen” ‘Wyldlannen’ near Eernewoude near (Fr) The polder “Wyldlannen”Eernewoude in the (Fr) province of FrieslandFigure (5”54”E; 1. A well 53’13”N)Figure-developed 1. A well-developedconsists nutrient poor nutrient Cirsio- poorMolinietum Cirsio-Molinietum meadow with sedges, sparse meadow with sedges, sparse reed and many flowering of c. 60 ha of degenerated and species-poor Cirsio-reedMolinietum and many vegetation. flowering individuals In the of Cirsium dissectum. he polder ‘Wyldlannen’ in the province of Friesland individuals of Cirsium dissectum. past these mesotrophic(5"54"E; fen 53'13"N) meadows consists were of c. 60 very ha of rich degenerated in species . They harboured many Red-List speciesand species-poorthat are currently Cirsio-Molinietum endangered vegetation. in most of Europe (figure 2). Examples are:T Cirsium dissectum, Carex dioica, Carex hostiana, Carex In the past these mesotrophic fen meadows were very pulicaris, Dactylorhiza incarnata, and Parnassia palustris. The polder is flooded rich in species. They harboured many Red-List species during winter and spring with surface water during several months. This flooding that are currently endangered in most of Europe practises has been done for centuries. Before the agricultural peat meadows (figure 2). Examples are:Cirsium dissectum, Carex dioica, were intensively drained (before World War II) these ‘summer polders’ were the Carex hostiana, Carex pulicaris, Dactylorhiza incarnata, lowest spots in the landscape and the floodwater originated from groundwater and Parnassia palustris. The polder is flooded during from the Drenthian Plateau (to the east). This groundwater was rich in iron and calcium, but poorwinter in andnutrients. spring with surface water during several months. This flooding practises has been done for centuries. Before the agricultural peat meadows were intensively drained (before World War II) these ‘summer polders’ were the lowest spots in the landscape and

the floodwater originated from groundwater from the Figure 2. Impression of a lakes and ‘summer polders’, Drenthian Plateau (to the east). This groundwater was surrounded by agricultural peat meadows with very low rich in iron and calcium, but poor in nutrients. water levels.

Figure 2. Impression of a lakes and ‘summer polders’, surrounded by agricultural peat meadows with very low water levels. 16

So, the flood water was cleaned, but remained base-poor. However, after 10 years of monitoring we found that neither continuous mowing nor sod cutting resulted in the restoration of the mesotrophic fen meadow vegetation (Figure 4). peatlands surrounding the lakes and adjoining summer polders, the peat surface have dropped several meter (subsidence) and the summer polders are now the highest spots in the landscape (Figure 2). Consequently the water levels in the summer polders have dropped a lot (up to one meter). Consequently grass species like Agrostis canina and Phalaris arundinacea gained dominance due to eutrophication of the flood water and increased fluctuations of the water table (Figure 3).

Figure 4. Regular mowing ( site 3) nor sod cutting (site 4) original Figure 4. Regularresulted mowing in the ( site restoration 3) nor sod of cuttingthe original (site fen4) resultedmeadow in in the the restoration of the . Although some recovery was observed in the mown site during the fen meadow inWyldlannen. the Wyldlannen Although some recovery was observed in the first 6 years. However, none of the characteristic species of fen meadows returned. mown site during the first 6 years. However, none of the Was is perhapscharacteristic that the species species of fen could meadows not reachreturned. the area anymore, because the species were not available any more in the local soil seed bank, or the remaining populations foundin the that surroundings neither continuous were too mowing far away nor .sod Or cutting the environment was not suitable anymore;resulted iron, in the calcium restoration and of bicarbonate the mesotrophic water fen in the surface water was too low to increasemeadow thevegetation base-saturation (Figure 4). and the pH. Was is perhaps that the species could not reach the Analyses of the soil seed banks in the reference site, the sod cut site and the Figure 3. Position of the summer polder ‘Wyldlannen’(degraded (light (mown)area anymore, site showed because thatthe species the reference were not availablesite had viable seed of Cirsium green) surrounded by open water and terrestrializingdissectum fens , notany moreof Carex in the hostiana local soil or seed Pedicularis bank, or the palustris remaining, which were all present and reeds (yellow). To the left the low lying and heavily in the vegetation.populations In the in degradedthe surroundings sits (both were mown too far away.and sod cut , no viable seeds drained agricultural peat polders can be seen. 1 = reference of characteristic species were found. are with well developed Cirsio-Molinietum vegetation, Or the environment was not suitable anymore; iron, 2= degraded Cirsio-Molinietum stands near the openWe water, decided calciumto introduced and bicarbonate juveniles water of rare in the species surface to water the experimentalwas plots in 3 = very degraded Cirsio-Molinietum stands next to 1994a sod cut (Carex too hostiana, low to increase Cirsium the dissectum base-saturation and Pedicularisand the pH. palustris, and also site where the acidified and degraded vegetation has been added lime (CaCOAnalyses3) to of the the sod soil cutseed and banks mown in the variants. reference site, removed, 5 = helophyte filter to clean the polluted surface the sod cut site and the (degraded (mown) site showed water. Pedicularis palustris died the same year. Carex hostiana and Cirsium dissectum survived onethat year the and reference then sitedied. had Adding viable seedlime of to Cirsium Carex hostiana died have a significant positivedissectum, effect not of on Carex the hostiana growth or of Pedicularis the species palustris, in 1995,but only in the sod Attempts to restore species rich fencut plots. Howeverwhich were none all presentof the introduced in the vegetation. individual In the survived the following meadows failed winter. degraded sits (both mown and sod cut, no viable seeds In 1986 the Frisian nature protection agency (‘ItSo, Fryske something of characteristic was wrong specieswith the were pH, found. which was not increasing after sod Gea’ = The Frisian Landscape) initiated a projectcutting, in floodingWe withdecided clean to introduced surface water juveniles and of also rare not species after to liming the soil (only order to restore the species-rich Cirsio-Molinietumonce in 1994).the experimental plots in 1994 (Carex hostiana, Cirsium meadows. They raised the water levels in the ditches, dissectum and Pedicularis palustris, and also added lime which resulted in some rewetting, but not in an (CaCO3) to the sod cut and mown variants. increase of characteristic fen meadow species. In Pedicularis palustris died the same year. Carex 1991 a small irrigated wetland (helophyte filter) was hostiana and Cirsium dissectum survived one year and constructed to purify the polluted surface water before then died. Adding lime to Carex hostiana died have a it would flood the meadows. Water analyses showed significant positive effect on the growth of the species that the filter worked very well. The nutrient contents in 1995,but only in the sod cut plots. However none of of nitrogen and phosphorous dropped by 90% after the introduced individual survived the following winter. passing the helophyte filter. However the surface So, something was wrong with the pH, which was not water that eventually reached the meadows was low increasing after sod cutting, flooding with clean surface in calcium and bicarbonate, due to dilution with water and also not after liming the soil (only once in precipitation water (Van Duren et al. 1998). 1994). So, the flood water was cleaned, but remained We analyzed the base saturation of the soil 4 times base-poor. However, after 10 years of monitoring we within 10 years and found that in spring the base

17 So, the restoration measures to increase the base saturation did work during flooding, but that was apparently not sufficient to restore the environment for Cirsio-Molinietum vegetation.

Perhaps the soil had become too eutrophic due to the longtime flooding with polluted surface water. Too test this we carried out a full factorial fertilization experiment in both the sod cut area and the uncut mown area. The results are shown in figure 7 and they clearly indicate that phosphorus was limiting plant growth, in particular in the sod cut site. Well developed Cirsio-Molinietum fen meadows usually are phosphorus limited (Egloff 1983, Wassen et al. 2005). So, a shift from phosphorus limitation to nitrogen or potassium limitation had not occurred.

Figure 7. ResultsFigure of a full7. Results factorial of fertilization a full factorial experiment fertilization in a degraded experiment (mown) Cirsio-Molinietum meadow and inin a aneighboring degraded site(mown) where Cirsio-Molinietum the top soil has been meadow removed and (ca. in 10a cm; stripped site).The results indicate clear phosphorus limita neighboring site where the toption, soil inhas particular been removed in the sod (ca. cut site. 10 cm; stripped site).The results indicate clear phosphorus In 1999 we measured the water composition of the top layer during most of the vegetation season.limitation, We in compared particular in the the degraded sod cut site. (mown) site with a reference of a rather well developed Cirsio-Molinietum site (surface water fed) and also with a restored fen meadow in a groundwater fed site in a brook valley reserve (“Barten”). ThenPerhaps it became the soilclear had that become the degraded too eutrophic (and alsodue tothe reference site) in the Wyldlannen)the longtime showed flooding very low with pH pollutedvalues and surface high sulwater.fate valuates in autumn when the groundwater levels were very low. Too test this we carried out a full factorial fertilization experiment in both the sod cut area and the uncut mown area. The results are shown in figure 7 and they clearly indicate that phosphorus was limiting Figure 5. Dry weight and survival of three introduced species Figure 5. Dry weight and survival of three introduced species (Ca (Carex hostiana, Cirsium dissectum and Pedicularis palustris; plant growth, in particularrex hostiana, in the sodCirsium cut site. dissectum Well and Pedicularis palustris 200 individuals per species)) in; degraded200 individuals mown plots, per species)developed) in degraded Cirsio-Molinietum mown plots, fen meadows degraded usually sod are cut plots and in plots that degraded sod cut plots and receivedin plots that additional received additional lime. Thephosphorus results clearly limited show (Egloff that 1983, the Wassen environmentalet al. 2005). conditionslime. The results did notclearly meet show thethat therequirements environmental of the species, not even after liming So, a shift from phosphorus limitation. to nitrogen or conditions did not meet the requirements of the species, not Figure 5. Dry weight and survival of three introduced speciespotassium (Ca rlimitationex hostiana, had not Cirsium occurred. dissectum Weeven analyzed after liming. the base saturation of the soil 4 times within and Pedicularis palustris; 200 individuals per species)) in degradedIn 1999 we measuredmown plots, the10 water degradedyears composition and sod foundcut of the that in spring the bas plots andsaturation in plots increased that received significantly,e additional saturation in both thelime. sod increased Thecut resultstop significantly,layerclearly during show most that of in the the both vegetation environmental the season. sod Wecut area . conditionsandarea the and did the uncut,not uncut, meet mownmown the requirementsplots plots (Figure (Figure 6). of the 6). species, compared not even the after degraded liming (mown) site with a reference of So, the restoration measures to increase the base a rather well developed Cirsio-Molinietum site (surface We analyzed the base saturation of the soil 4 times within 10 years and found saturation did work during flooding, but that was water fed) and also with a restored fen meadow in a e saturation increased significantly, in both the sod cut area that inapparently spring not the sufficient bas to restore the environment for groundwater fed site in a brook valley reserve (‘Barten’). Base saturation sod cut and theCirsio-Molinietum uncut, mown vegetation. plots (Figure 6). Then it became clear that the degraded (and also the mown reference site) in the Wyldlannen) showed very low pH

values and high sulfate valuates in autumn when the

Base saturation groundwatersod cut levels were very low. Increased sulfate concentration in autumn, mown combined with a strong drop in pH points to oxidation of iron sulfate (FeS) in the upper layers of the soil after

oxidation during periods with low water levels. Low

water levels occur usually in autumn. The oxidation of iron sulfate is more pronounced at 25 cm below the

surface compared to the top 10 cm, where most of the FeS has already been oxidized.

But where do these high pyrite contents of the Figure 6. Increase in base saturation of the sod cut and not sod cut top soil in a degraded Cirsio soil come from. It has been suggested that the pyrite - Molinietum in the nature reserve Wyldlannen (sites 3has and a marine 4 in figureorigin and 3). was formed when the sea Figure 6. Increase in base saturation of the sod cut and not regularly flooded the area several centuries ago. Other sod cut top soil in a degraded Cirsio-Molinietum in the nature reserve Wyldlannen (sites 3 and 4 in figure 3). point to a more recent origin and relate it to increased Figure 6. Increase in base saturation of the sod cut and not sod cut top soil in a degraded Cirsio- Molinietum in the nature reserve Wyldlannen (sites 3 and 4 in figure 3). 18

pH in soil water 1999 Wyld Wyld Refer. Refer. Barten Barten depth 10 25 15 25 5 20 Jan 6,6 6,3 6,7 7,1 Febr 7,2 7 April 6,3 6,4 6,5 6,8 7 7,1 May 6,7 7 June 6,3 6,6 6,6 6,9 July 6,3 6,8 5,8 6,9 Sept 6,2 6,3 6,7 5,7 6,6 Oct 4,9 5,3 5,4 5,6 5,7 6,7

2– pH in soil water SO4 in soil water (mg/l) 1999 Wyld Wyld Refer. Refer. Barten Barten 1999 Wyld Wyld Refer. Refer. Barten Barten depth 10 25 15 25 5 20 depth 10 25 15 25 5 20 Jan 6,6 6,3 6,7 7,1 Jan 40,9 32,3 9,6 8,5 Febr 7,2 7 Febr 7,5 8,6 April 6,3 6,4 6,5 6,8 7 7,1 April 54,9 101,1 69,3 57,4 6,7 8,5 May 6,7 7 May 8,5 8 June 6,3 6,6 6,6 6,9 June 73,3 49,2 13,8 11,6 July 6,3 6,8 5,8 6,9 July 4,5 9,7 Sept 6,2 6,3 6,7 5,7 6,6 Sept 436 62,4 Oct 4,9 5,3 5,4 5,6 5,7 6,7 Oct 294 501 173,5 271,1 14,3 20,1

Table 1. Changes in pH in soil water of the top soil of a Table 2. Changes in sulfate concentrations during the SO 2– in soil water (mg/l) degraded4 Cirsio-Molinietum site, a well-developed reference season in the soil water of the top soil of a degraded Cirsio- 1999 Wyld Wyld Refer. Refer. Barten Barten site area (both in the fen meadow reserve ‘Wyldlannen’ Molinietum site, a well-developed reference site in the fen depth 10 25 15 25 5 20 influenced by surface water, compared to another fen meadow reserve ‘Wyldlannen’, compared to a fen meadow Jan 40,9 32,3 9,6 8,5 meadow reserve influenced by iron and calcium rich reserve influenced by iron and calcium rich groundwater Febr 7,5 8,6 groundwater (‘Barten’). (‘Barten’). April 54,9 101,1 69,3 57,4 6,7 8,5 May 8,5 8 atmosphericJune SO2 deposition73,3 several49,2 decades13,8 ago 11,6 anaerobic groundwater is present (Grootjans et al. 2002). (KemmersJuly et al. 2000). A very likely origin of4,5 pyrite 9,7is They have adopted new targets for the area and alsoSept the high loads of sulfate-rich436 groundwater62,4 during dropped the aim to restore species-rich meadows Oct 294 501 173,5 271,1 14,3 20,1 almost 50 years of pumping polluted river water into (Cirsio-Molinietum). Instead they try aim at the low lying polders of Friesland. The iron has been maintaining small sedge vegetation (Caricion nigrae), brought in the past by groundwater discharge from which is slightly wetter and more acid then the Cirsio- 2- the Drenthian Plateau. The sulfate (SO4 from the Molinietum meadows. By pumping more clean surface large rivers is reduced to sulfite (S2-) when the flood water into the reserve during the summer or to prolong water stagnates on the surface of the polders and then the flooding period in spring such a vegetation type can binds with iron. Although the most important source be maintained. of pyrite is not yet known, we may conclude that the water management of the past 50 years has made the problems for nature conservation worse. And when Further information/further reading the subsidence of the surface in the agricultural peat Egloff, Th. 1983. Der Phosphor als primär limitierender Nährstoff polders accelerated a situation of no return occurred. in Streuwiesen (Molinion). Berichte der Geobotanischen Institut ETH. Stiftung Rübel 50: 119-148. 9 So the low water levels in autumn is responsible for the acidification of the profile and this overrides all Grootjans, A.P., Bakker, J.P., Jansen, A.J.M., Kemmers, R.H. 2002. Restoration of brook valley meadows in the Netherlands. restoration measures during the wet season. And low Hydrobiologia 478, 149-170. water levels during the end of the summer cannot be prevented anymore, because the nature reserve is now Kemmers, R.H. Jansen, P.C., & Van Delft, S.P.J. 2000. De regulatie van de basentoestand in kwelafhankelijke the highest point in the landscape; the groundwater in schraallanden en laagvenen. Report Ministry of Africulture, the reserve is flowing towards the low-lying agricultural Nature and fisheries/ Alterra Wageningen. areas at all times. Pegtel. D.M. 1983. Ecological aspects of a nutrient-deficient wet grassland (Cirsio-Molinietum). Verhandlungen der Geselschaft Finding new Targets für Ökologie 10: 217-228. 9 This means that acidification of the top soil cannot be Smolders, A.J.P., Lamers, L.P.M., Lucassen, E.C.H.E.T., van der prevented unless base-rich groundwater is pumped up Velde, G., Roelofs, J.G.M. 2006. Internal eutrophication: How from deeper layers during both dry and wet periods. It it works and what to do about it - a review. Chemistry and Ecology (22) 2, 93-111. Fryske Gea, the owner of the reserve has rejected this option, because it too artificial and too expensive. Top Van Duren, I.C., Strykstra, R.J., Grootjans, A.P., Ter Heerdt, G.N.J. & Pegtel, D.M. (1998). Applied Vegetation Science 1: soil removal is not an option, because it makes things 115-130. worse. A new top soil is exposed with more iron sulfate, creating more problems during dry periods. Top soil Wassen MJ, Olde Venterink H, Lapshina ED, & Tanneberger F (2005). Endangered plants persist under phosphorous removal should only be applied when strong discharge of limitation. Nature 437: 547-551.

19

The Drentsche Aa National Landscape Park; after 50 years of restoration.The Drentsche Aa National Landscape Ab Grootjans, Christian Hoetz, Henk Everts, Piet Schipper, and Enno Bregman.Park; after 50 years of restoration

Friday Aug 24th 09.30-12.30 Ab Grootjans, Christian Hoetz, Henk Everts, Piet Schipper, and Enno Bregman Friday Aug 24th 09.30-12.30

The Drentsche Aa National Landscape ThePark Drentsche Aa National Landscape Park The Drentschehe Drentsche Aa AaReserve Reserve is is aa wellwell preserved preserved brook valley system in the north east ofbrook the Netherlands,valley system in named the north after east the of the small river that dominates the area. It is located in the province of Drenthe and a plan to protect most of the catchment Netherlands, named after the small river that areaT was launched in 1965 after continuous destruction of almost all brook valley dominates the area. It is located in the province of systems in the Netherlands due to agricultural improvement plans. At the end of theDrenthe 1960- andties a few plan brook to protect valley most systems of the catchment with a more or less intact geomorphology andarea hydrological was launched system in 1965 hadafter remained.continuous Figure 1. Relief of the Drentsche Aa catchment area with sites destruction of almost all brook valley systems in Figurewe will 1 .visit: Relief 2 = Oudeof the Molen, Drentsche 3, 4 = Taarlo,5, Aa catchment 6 = Rolderdiep. area with sites we will visit: 2 = Oude the Netherlands due to agricultural improvement Molen, 3, 4 = Taarlo,5, 6 = Rolderdiep. 33 plans. At the end of the 1960-ties few brook valley The natural mires in the Drentsche Aa area were

systems with a more or less intact geomorphology and formed in eroded melt water valleys since 10400 BP hydrological system had remained. Humanat the deepest impact parts. At 1800 BP the peat formation spread out over the valley shoulders. Peat formation Human impact Geomorphologicallystopped in early medieval and times.hydrologically In the beginning the Drentsche most Aa landscape is relative to other brook valley systems in the Netherlands little disturbed, but the vegetation Geomorphologically and hydrologically the Drentsche of the peat was formed by alder forests and eutrophic has changes very much, due to agricultural use during many centuries. These Aa landscape is relative to other brook valley systems changessedge marshes. in time In are a later illustrated stage, when in figure the peat 2. grew in the Netherlands little disturbed, but the vegetation thicker, mesotrophic small sedges with no or little tree has changes very much, due to agricultural use during growth were the dominant vegetation types. During many centuries. These changes in time are illustrated in the early middle ages almost all of the natural fens figure 2. changed into slightly drained fen meadows. Many 34

20 Figure 2. Changes in time in a brook valley landscape due to landscape use. On the right end side oligotrophic natural mires (fens = dark yellow, and isolated bogs = light yellow) were dominant, with eutrophic flood mires in the lower reaches. These mires changes in fen meadows due to shallow drainage and increased flooding. In a later stage modern Figureagriculture 2. left Changes a landscape inwith time deep drainagein a brook channels valley landscapemeander freely due and to have landscape maintained use their. naturalOn the right end side oligotrophicand heavy fertilized natural grasslands mires with a very(fens low =biodiversity dark yellow, course and in theisolated landscape bogs (Figure = 1).light The areayellow) also has were dominant, (from: Grootjans & Van Diggelen 1995). White arrows with eutrophic flood mires in the lower reaches.valuable These cultural qualitiesmires changessuch as a characteristic in fen meadows due to represent groundwater flows. Brown colors represent peat shallow drainage and increased flooding. In a later stage modern agriculture left a landscape with soils, while grey colors represent sand or clay deposits (left). combination of historical villages and fields. There deep drainage channels and heavy fertilizedare alsograsslands many archaeological with a sites, very such low as dolmens biodiversity (from: Grootjans & Van Diggelen 1995). White arrows represent groundwater flows. Brown colors small ditch systems were constructed, especially in and burial mounds. National Landscape Parks in represent peat soils, while grey colors represent sand or clay deposits (left). areas with strong groundwater discharge. The large the Netherlands are different from nature reserves, Thescale reclamationnatural and mires fertilization in ifthe fen meadowsDrent andsche Aawhere area the focus were lies on formedminimizing humanin eroded impacts. melt water heathlands started around 1920 when more intensive In a National Landscape Park also cultural aspects valleys since 10400 BP at the deepest parts. At 1800 BP the peat formation agriculture started to develop and artificial fertilizers are important (Spek et al. 2015). Within the landscape spreadbecame available. out Inover the 1960ties the andvalley 1970ties shoulders. intensive Park, Peat for instance, formation many small stopped villages and surroundingin early medieval times.agriculture andIn deepthe drainage beginning practices destroyed most of theagricultural peat fieldswas cover formed about 50% byof the alderPark area. forests and eutrophicmost of the existing sedge landscape marshes. and turned it intoIn a laterMore naturalstage, areas occupywhen only the 35 %. (Roelsmapeat et grewal., thicker, monotonous grasslands and croplands, with very 2004). mesotrophic small sedges with no or little tree growth were the dominant little biodiversity. In the brook valleys the large scale After more than 50 year of restoration management vegetationdrainage systems ledtypes. to increased During flooding the events, early middlein the Drentsche ages Aa catchmentalmost area, all restoration of the natural fens changedwhich triggered into the development slightly of moredrained infrastructure fen meadows.outcomes have beenMany very successfulsmall inditch most of thesystems wet were constructed,to regulate the surface especially water levels (Grootjans in areas & Van withmeadows. strong groundwater discharge. The large scaleDiggelen reclamation 1995, Van der Spek etand al. 2015). fertilization if fenFigure meadows 3. Shows the and vegetation heathlands changes after 20started around years of restoration management (mowing without 1920Restoring when the more species intensive rich fen agriculturefertilization started and to additional develop rewetting). and In 1982artificial most fertilizers becamemeadows available. (since 1965) In the 1960ties andof the 1970tiesarea consisted ofintensive intensely use agriculturalagriculture and deep drainageBetween 1965 andpractices 2006 the area destroyed that was bought most by ofgrasslands the existing with Lolium perennelandscape and Holcus and lanatus turned it into monotonousthe state and protected grasslands nature areas steadily and increased croplands, grasslands with with very small little nature reservesbiodiv withersity. Caltha In the brook form a few hundred hectares to almost 5,000 hectares. palustris meadows (Calthion palustris) and small areas valleysMost of this the area waslarge managed scale by the drainage State Forestry systemsof small led sedge to vegetation increased (Caricion flooding nigrae). In 2012 events, the which triggeredCommission, the the state developmentnature conservation agency.of more infrastructurevariation in vegetation to types regulate was much larger. the Calthion surface water levelsIn 2006, around(Grootjans 32,000 hectares & Van of the Diggelen Drentsche Aa 1995,palustris Van meadows der Spek and small et sedge al. vegetation2015). had brook valley were designated as a National Landscape expanded and new vegetations type of nutrient poor RestoringPark. The aim of this the appointment species was to maintain rich fen moistmeadows meadows (Junco-Molinion) (since had 1965) developed after the characteristic landscape and its natural and 20 years of mowing, without fertilizing. In these new recreational qualities. Nowadays the Drentsche Aa types Juncus acutiflorus had become the dominant Between 1965 and 2006 the area that was bought by the state and protected brook valley contains various valuable ecosystem species and many individuals of the orchid Dactylorhiza naturetypes, harbouring areas many steadily rare plant and increased animal species, form majalis a few appeared. hundred Rewetting washectares achieved by to closing almost 5,000 hectares.such as the Northern Most sedge of (Carexthis aquatilis) area (Grootjans was managed small agricultural by the ditches State (from: Forestry Bakker et al. C 2015).ommission, the state& Van Tooren nature 1984, Spek conservation et al. 2015). Most brooks agency. still In 2006, around 32,000 hectares of the Drentsche Aa brook valley were designated as a National Landscape Park. The aim of this appointment was to maintain the characteristic landscape21 and its natural and recreational qualities. Nowadays the Drentsche Aa brook valley

35

contains various valuable ecosystem types, harbouring many rare plant and animal species, such as the Northern sedge (Carex aquatilis) (Grootjans & Van Tooren 1984, Spek et al. 2015). Most brooks still meander freely and have maintained their natural course in the landscape (Figure 1). The area also has valuable cultural qualities such as a characteristic combination of historical villages and fields. There are also many archaeological sites, such as dolmens and burial mounds. National Landscape Parks in the Netherlands are different from nature reserves, where the focus lies on minimizing human impacts. In a National Landscape Park also cultural aspects are important (Spek et al. 2015). Within the landscape Park, for instance, many small villages and surrounding agricultural fields cover about 50% of the Park area. More natural areas occupy only 35 %. (Roelsma et al., 2004).

After more than 50 year of restoration management in the Drentsche Aa catchment area, restoration outcomes have been very successful in most of the wet meadows.

1982 1994 – 1996 2008 – 2012

Figure 3. Vegetation development during 20 years in an area Where the stratigraphy of the subsoil is intact, two Figureof 34.2 3ha. inVegetation the middle reaches development of the Drentsche Aa duringNational 20aquifers years can inbe distinguished,an area of divided 34.2 by twoha poorlyin the middle Landscape Park. reaches of the Drentsche Aa National Landscapepermeable Park. layers. The first water layer in the subsoil is called phreatic groundwater. The second water layer FigureHydrological 3. Shows systems the analyses vegetation changesis called the firstafter aquifer 20 and theyears third waterof layer restoration is The Drentsche Aa area is a basin where the water called the second or deep aquifer (Figure 3) (see also Calthionmanagement palustris meadows (mowing and small without sedge vegfertilizationetation had expandedand additional and new rewetting). In 1982 is drained towards the sea through the brook valley Magri & Bregman, 2011). Phreatic groundwater from the vegetationsmost oftype the of area nutrient consisted poor moist of intenselymeadows use(Junco agricultural-Molinion) had grasslands with Lolium system. Its hydrology can be characterized as a phreatic hydrological system has only travelled a short developedperenne after and20 years Holcus of mowing, lanatus without grasslands fertilizing. with In thesesmall new nature types reserves with Caltha Juncus acutiflorusPleistocene groundwaterhad become system the wheredominant the brook species andperiod many through individuals shallow of decalcified the soil formations palustris meadows (Calthion palustris) and small areas of small sedge vegetation orchid Dactylorhizavalleys are fed frommajalis a higher appeared. lying plateau Rewetting along the was achievedbefore it exfiltrates. by closing It is smalloften eutrophic and oxygen agricultural(Caricionsouthern ditches edges nigrae) (from: (Engelen Bakker. 1984,In Oude2012 et al Munninck. 2015).the 1985). variation rich, in but vegetationacidic and mineral types poor. Groundwater was much that larger. The hydrological system is quite complex, because originates from the first aquifer often comes from a Hydrologicalthere are manysystems impermeable analyses layers in the subsoil local hydrological system. Its chemical composition is throughout the area. These layers block the flow of 36intermediate between phreatic and deep groundwater. The Drentsche Aa area is a basin where the water is drained towards the sea groundwater. Rainwater infiltrates the soil there, Deep groundwater has been underground for long through the brook valley system. Its hydrology can be characterized as a which flows north-easterly to the lower lying area. The periods (Elshehawi et al. 2018). The water is alkaline, Pleistocene groundwater system where the brook valleys are fed from a higher maximum difference in height is 22 meters (Figure 4). oxygen poor, mineral rich, has a high pH and is poor lying plateau along the southern edges (Engelen 1984, Oude Munninck 1985).

Figure 4: Hydrogeological sketch of the groundwater flows from the Drenthian plateau to the Wadden Sea, including the aquifers. 1 shows exfiltration of phreatic groundwater; 2 is local exfiltration of shallow groundwater; 3 is (sub) regional exfiltration of deep groundwater (figure modified from: Everts & De Vries, 1991).

Figure 4: 22Hydrogeological sketch of the groundwater flows from the Drenthian plateau to the Wadden Sea, including the aquifers. 1 shows exfiltration of phreatic groundwater; 2 is local exfiltration of shallow groundwater; 3 is (sub)regional exfiltration of deep groundwater (figure modified from: Everts & De Vries, 1991).

The hydrological system is quite complex, because there are many impermeable layers in the subsoil throughout the area. These layers block the flow of groundwater. Rainwater infiltrates the soil there, which flows north-easterly to the lower lying area. The maximum difference in height is 22 meters (Figure 4).

Where the stratigraphy of the subsoil is intact, two aquifers can be distinguished, divided by two poorly permeable layers. The first water layer in the subsoil is called phreatic groundwater. The second water layer is called the first aquifer and the third water layer is called the second or deep aquifer (Figure 3) (see also Magri & Bregman, 2011). Phreatic groundwater from the phreatic hydrological system has only travelled a short period through shallow decalcified soil formations before it exfiltrates. It is often eutrophic and oxygen rich, but acidic and mineral poor. Groundwater that originates from the first aquifer often comes from a local hydrological system. Its chemical composition is intermediate between phreatic and deep groundwater. Deep groundwater has been underground for long periods (Elshehawi et al. 2018). The water is alkaline,

37

in nutrients. In the brook valleys this water exfiltrates it is necessary to have a basic understanding of the to the surface and enters the valleys. The exfiltration geology of the area (Figure 5). intensity is determined by the groundwater pressure The subsoil of the brook valley is stratified and the and the presence of impermeable soil layers. The whole characteristics of each layer influence the hydrological area of the Drentsche Aa used to be considered as one system. The area has a hydrological base which is regional groundwater system. But later suggested largely impermeable to water due to its high fraction of that it actually consists of at least five hydrological clay, but broken near pro- and postglacial reactivated or subsystems. (Schipper & Streefkerk 1993). The relative induced slip faults which separated deep and shallow shallow (up to 50 meter below the surface) and the hydrological systems and disrupted the seal capacity surface systems itself have been very well studied and of the hydrological base and overlaying clays. This modelled in the past and the results are, up to now, disrupted hydrological base lies at a depth of about oxygenthe mainpoor, mineral base for rich nature/, has a high and pH landscape and is poor inplanning. nutrients. In the brook50 to 150 meters (Magri & Bregman 2011). At locations valleys this water exfiltrates to the surface and enters the valleys. The exfiltrationHowever, intensity decreasing is determined funds forby naturethe groundwater conservation pressure and thewhere impermeable layers have been eroded by the presence of impermeable soil layers. The whole area of the Drentsche Aa used to be andconsidered increasing as one threatsregional groundwaterof agricultural system. pollution But later suggestedfrom that itsubglacial channels or weak fault zones, water can actuallyhigher consists grounds of at leastnow five require hydrological a better subsystems understanding. (Schipper & of Streefkerk flow from one aquifer to another or exfiltrate to the 1993). The relative shallow (up to 50 meter below the surface) and the surface systemsdeeper itself groundwater have been very flows well studiedand there and modelledinteraction in the withpast and thesurface. Large local differences occur within the basin results are, up to now, the main base for nature/ and landscape planning. However,the hydrological decreasing funds system for nature of the conservation Drentsche and Aa increasingas a whole. threats ofconcerning thickness, composition and even presence agriculturalThis would pollution enable from highera more grounds balanced now require long terma better nature understanding ofof the formations. At certain locations the formations deeper groundwater flows and there interaction with the hydrological system of theand Drentsche landscape Aa as policy.a whole. This would enable a more balanced long termhave been pushed to the surface and eroded because of nature and landscape policy. the movement of underlying salt domes or other glacio-

induced or older tectonic processes (Roelsma et al. 2004,

Bregman & Magri 2011, Bregman et al. 2015). Recent

detailed 3D studies, based on seismic and geological

and geomorphologic data showed that the present

Holocene river patterns and initial erosion of salt

domes, (re-)activated faults consequently influenced

deep groundwater flow.14 C dating indicate a start of

peat formation at 10400 BP in a lake with brine water

influence; downstream the valley base at 1.5 m peat

formation started at 1800 BP in the uplifted areas. The

presence of brine water in deeper layers is caused by

FigureFigure 5. Hypothetical 5. Hypothetical sketch of sketchgroundwater of groundwater flows based on geological flows data,based macro -ionicdischarge of deep groundwater in the surroundings of compositionon geological of groundwater data, macro-ionicand distribution composition patterns of plant of speciesgroundwater (After: Elshehawi et al. 2018). (Permian) salt domes. Salt has a higher temperature and distribution patterns of plant species (After: Elshehawi Modelset al. of 2018). the genesis of the North Netherlands landscapes were based onbecause of a better heat conductivity. This leads to geological data and hypothesis postulated in the 70ties- 90ties of the pastdouble diffuse convection (DDC): because of the warmer century excluded pro- and postglacial differential Saalian and for sure WeichselianModels reboundof the genesisand impact of of the subglacial North formed Netherlands deep eroded older Elsterianenvironment the groundwater has a higher density and channel systems. Growing awareness and interest of the Drentsche Aa area at thelandscapes South edge ofwere a tectonic based basin on geological with up-coning data (Permian) and salt domes hasflows to surface. The seepage of this (bicarbonate rich) initiatedhypothesis cooperation postulated with Dutch in -the , Danish70ties-- and90ties German of the univer pastsities (TUgroundwater from the regional systems results in a century excluded pro- and postglacial38 differential different vegetation in seepage areas compared to areas

Saalian and for sure Weichselian rebound and impact only influence by local groundwater systems. (figure 5). of subglacial formed deep eroded older Elsterian To summarize: the brook valleys can be fed by channel systems. Growing awareness and interest of exfiltrating phreatic groundwater, groundwater from the Drentsche Aa area at the South edge of a tectonic the first and from the second aquifer, or even by the basin with up-coning (Permian) salt domes has brine groundwater from below the clay layers (Magri & initiated cooperation with Dutch-, Danish- and German Bregman 2011) that was in former times considered to universities (TU Kopenhagen, FU Berlin) and has be the ‘hydrological basis’. In case the river or the old generated a better insight of the deep (thermohaline) glacial erosion channelshave cut through the various ground waterflow patterns. clay layers, wet meadows in the valleys can even be fed To comprehend the hydrological system of the area, by 2-4 different groundwater flows (Figure 6). Near area’s

23 Fig. 6. Different groundwater flows in the upper- middle and lower reaches of the Drentsche Aa catchment area. The shallow groundwater in particular have been polluted by agriculture. Also effects of groundwater abstraction influence the groundwater flows and the discharge of groundwater in the brook valleys.

Calcium poor groundwater

Calcium rich groundwater

Moderately polluted groundwater

Very polluted groundwater

lower course of the Drentse A. This nature reserve was with faults structures the deep groundwater pressure originally established in 1965 (first initiatives) to protect and salinity is higher, as well as the groundwater typical flood mire vegetation withCarex acuta, Carex temperature (0.8 – 1.5 degrees). aquatlis and Carex elata and the cultural heritage of the landscape elements. Groundwater extraction for drinking Repeated vegetation mapping of the area (over 37 water production years) showed that between 1975 and 1996 showed At present there are 6 groundwater extraction facilities that there were few changes in wetness and trophy within the border of the National Landscape Park indication of the vegetation between 1974 and 1996 or very close to it. One of the facilities (Loon) have (Figure 7). Eutrophic vegetation types were dominant been reduced from 6 to 3 million cubic meter in 2013 and relatively stable for almost 22 years. Mesotrophic already. Groundwater is usually extracted from the species showed a small increase at the expense of second aquifer. Only one (De Punt, in the lower course) oligotrophic types in 1982. This effect partly disappeared used mostly surface water from the river to produce again in 1994-96. As can be seen in figure 6, there was a drinking water, but will be closed in the near future. large drop in nutrient availability in 2011. Mesotrophic Groundwater extraction from the second aquifer vegetation types became dominant at the expense causes a drop in water pressure in this aquifer, which of eutrophic types. Figure 7 also showed that the leads to less exfiltration of nutrient-poor and base- flood meadows acidified between 1996 and 2011; the rich groundwater in the brook valleys. This leads to intermediate alkaline types showed a large decline in lower water levels in the wet meadows and causes the cover from 74% to 35 %. They are mostly replaced by replacement of groundwater by acidic rainwater in the neural types, but also by intermediate acidic and acidic topsoil (Grootjans et al. 1988, Van Diggelen et al., 1998). vegetation types. The vegetation types that declined One of the areas that are most affected by groundwater were the four communities of Carex aquatilis, Carex acuta abstraction is the Kappersbult (c. 25 ha.), situated in the and Glyceria maxima. They were mainly replaced by four

24 Kappersbult: Vegetation cover of plant communities grouped by wetness indication Diggelen et al.., 1990). The possible improvements to the 80 hydrological system because of the reduction of GWE by the pumping station of De Punt are not noticeable 60 dry from these results. This development is in line with 40 moist the prediction of both van Diggelen et al. (1990) and wet Schipper & Streefkerk (1993) that the hydrological 20

Percentage cover system of the Kappersbult is strongly affected by several very wet factors, therefore solving only one problem will not lead 0 other 1974-75 1982 1994-96 2011 to hydrological improvements. Year Since the Kappersbult used to be a eutrophic and alkaline marsh, the current shift to neutral and Kappersbult: mesotrophic conditions is detrimental to the native Vegetation cover of plant communities vegetation. Although the environmental value of the grouped by trophy indication Kappersbult is still very high, this is likely to decrease as 90 the degradation of the nature reserve progresses. 70 External and internal drainage of wet 50 oligotroph Externalmeadows and internal drainage of wet meadows. ic 30 mesotrop In 1995In 1995 about about half half of of the the vegetation vegetation on peatpeat soilssoils were were considered to be affected hic eutropic by considereddesiccation to (figure be affected 8a). Groundwaterby desiccation extraction (figure 8a). for drinking water production

Percentage cover 10 is partly responsible for this environmental problem, especially in the areas near Groundwater extraction for drinking water production the pumping stations (Hoetz, 2013). The intensive drainage that is required for -10 1974-75 1982 1994-96 2011 agricultureis partly isresponsible the other formain this cause environmental. problem, especially in the areas near the pumping stations Year But also internal drainage had a negative impact on restoration. During the last

century(Hoetz, mowing 2013). The of intensivemeadows drainageon peat that soils is occurredrequired forwith small tractors, which wouldagriculture get stuck is the in otherthe peat main when cause. the drainage ditches were not functioning. Kappersbult: Since But1996, also most internal of the drainage mowing had is acarried negative out impact using large mowing machines with Vegetation cover of plant communities veryon broad restoration. tracks Duringthat exerts the last little century pressure mowing per cm2of of soil. Since the use of this equipment, more than 600 ha of wet meadows have been rewetted by removing grouped by pH indication meadows on peat soils occurred with small tractors, 100 all the former ditches (Figure 8b). In 2008 the area with desiccated meadows on peatwhich soils wouldwas reduced get stuck to in nearly the peat 10 when% of thethe peatlanddrainage areas. 80 alkaline

60 intermediate alkaline 40 neutral intermediate

Percentage cover 20 acidic acidic 0 1974-75 1982 1994-96 2011 Year Figure 7: Vegetation cover of plant communities grouped by Figure 7: Vegetation cover of plant communities grouped by their indication of wetness, their indication of wetness, nutrient availability and pH in thenutrient Kappersbult availability (Drentsche and Aa). pH in the Kappersbult (Drentsche Aa).

communities of Carex nigra, Carex aquatilis, Equisetum fluviatile, and partly Pedicularis palustris. The causes of this decrease in nutrient availability and acidification is not only due to groundwater 42 Figure 8. The left figure (8a) shows areas that from a extraction that triggered increased infiltration of rain Figurevegetation 8. The leftcomposition figure (8 a)point shows of few areas are that influence from a negativelyvegetation composition point of few are byinfluence internal negatively of eternal bydrainage internal (survey of eternal carried drainage out in 1995?). (survey carried out in 1995?). The water, but also a decrease in flooding frequency by right figure (8b) shows the areas (in blue) where since 1996 large scale rewetting The right figure (8b) shows the areas (in blue) where since the river. Therefore, the meadows would only receive occurred by removing all (former) agricultural ditches. 1996 large scale rewetting occurred by removing all (former) nutrient-poor rainwater and became oligotrophic (Van Avoidingagricultural g reenhouseditches. gas emissions by rewetting

A secondary benefit of rewetting areas with peat soils is that this reduces peat 25 mineralization. A high water table prevents oxygen to penetrate the soil and thus reduces aerobic decomposition of the organic components of the peat. Peat

43

mineralization causes eutrophication and land subsidence (De Vries et al., 2008). The peat also releases all of its carbon and part of its nitrogen as greenhouse gas (GHG) emissions to the atmosphere. In contrast, methane emissions increase with a higher water table, as less oxygen is available for decomposition. Therefore the decomposition becomes anaerobic. This partially offsets the

emission reduction of CO2 and N2O after the rewetting (Couwenberg, 2009). Nonetheless, there is usually a net reduction of the total GHG emissions (Augustin et al., 2011).

Figure 9. Some impressions of the rewetten hay meadows in the Drentsche Aa National ditchesLandscape were not functioning.Park, with Since Caltha 1996, mostpalustris of the (top)Figure and 9. ironSome impressionsdeposition of the along rewetten former hay meadows ditches. mowing is carried out using large mowing machines in the Drentsche Aa National Landscape Park, with Caltha palustris (top) and iron deposition along former ditches. with very broad tracks that exerts little pressure per cm2 of soil. Since the use of this equipment, more than 600 water table prevents oxygen to penetrate the soil and ha of wet meadows have been rewetted by removing all thus reduces aerobic decomposition of the organic the former ditches (Figure 8b). In 2008 the area with components44 of the peat. Peat mineralization causes desiccated meadows on peat soils was reduced to nearly eutrophication and land subsidence (De Vries et al., 10 % of the peatland areas. 2008). The peat also releases all of its carbon and part of its nitrogen as greenhouse gas (GHG) emissions to the Avoiding greenhouse gas emissions by atmosphere. In contrast, methane emissions increase rewetting with a higher water table, as less oxygen is available for A secondary benefit of rewetting areas with peat decomposition. Therefore the decomposition becomes soils is that this reduces peat mineralization. A high anaerobic. This partially offsets the emission reduction

26 of CO2 and N2O after the rewetting (Couwenberg, 2009). quaternary geology. Proefschrift. Rodopi, Amsterdam. Nonetheless, there is usually a net reduction of the total Elshehawi, S., Bregman, E.P.H. Schot, P. & Grootjans A.P. GHG emissions (Augustin et al. 2011). (submitted). Natural isotopes identify origin of groundwater The rewetting measures of SBB (Figure 9) have flows influencing wetland vegetation in the Drentsche Aa brook valley, the Netherlands. therefore inadvertently led to a secondary benefit. Not only did they increase biodiversity, but they have Engelen, G.B. 1984. Hydrological systems analysis. TNO-DGV, also reduced GHG emissions in the rewetted areas. report no 84-20. Rewetting measures are costly. To counter desiccation Grootjans, A.P., & Van Tooren, B.F. 1984. Ecological notes on within the National Landscape Park Drentsche Aa, over Carex aquatilis communities. Vegetatio 57: 79-89.

5 million euros have been allocated for the period of Grootjans, A.P., Van Diggelen, R., Everts, F.H., Schipper, P.C., 2007 to 2013 (Thije & Folkertsma, 2009). Hoetz (2013) Streefkerk, J., De Vries, N.P.J., & Wierda, A. (1993). Linking has tried to assess how much GHG emission had ecological patterns to hydrological conditions on various spatial scales: a case study of small stream valleys. In: Vos, C.C., & been avoided by rewetting 634 ha of peat soils in the Opdam, P. (eds.): Landscape Ecology of a Stressed Environment Drentsche Aa and how much could have been sold on (pp. 60–78). Chapman and Hall, London. the voluntary carbon market, when the rewetting had Grootjans, A.P., Bakker, J.P., Jansen, A.J.M., Kemmers, R.H. been carefully monitored. Hoetz (2013) even expected 2002. Restoration of brook valley meadows in the Netherlands. that this could even become a new business model to Hydrobiologia 478, 149-170. fund restoration management in peat areas (see also: Joosten, H. 2010. The global peatland CO2 picture - Peatland Joosten, 2010, 2011). Hoetz (2013) calculated that the status and drainage related emissions in all countries of the world. Wetlands International, Ede. increased groundwater levels in 635 hectares of peat soil has led to a reduction of ca. 2,000 tonnes of CO2 Joosten, H. 2011. Selling peatland rewetting on the compliance emissions per year, due to reduced peat mineralization. carbon market, in: Tanneberger, F., Wichtmann, W. (Eds.), Carbon credits from peatland rewetting – Climate – biodiversity Methane emissions were estimated to increase with – land use. Schweizerbart Science Publishers, Stuttgart, 99-105. ca. 1600 tonnes of CO2-equivalent per year, due to Magri, F., & Bregman, E. (2011). Regional-scale numerical increased anaerobic decomposition. The rewetting model of coupled fluid flow and mass transport along a deep measures thus led to a net emissions reduction of ca. geological profile in the Drenthe area, The Netherlands. Final Report of the Demo Pilot Project. Geowissenschaften Institut für 400 tonnes of CO2-equivalent per year. Selling that on Geologische Wissenschaften, Berlin. the voluntary carbon market would yield a sum of ca 150,000 for the coming 50 years at the current price of Rappol, M., 1984. Till in southeast Drenthe and the origin of 7€ /ton. Higher prices are sometimes paid, up to 35€ / the Hondsrug complex, the Netherlands. Eiszeitalter und Gegenwart 34:7-27. ton (Joosten 2011?), which would amount to 700,000 € / ton over a 50 years period, or 140,000 a year. That would Roelsma, J., Wanningen, H., van der Bolt, F.J.E. 2004. Systeemverkenning Drentsche Aa. Alterra-report 967. Reeks not cover the cost of the rewetting measures (more than monitoring stroomgebieden 2-I. Alterra, Wageningen. 700,000 per year (period 2007-2013), but could be a helpful in carrying out the mowing efforts). Schipper, P.C., Streefkerk, J.G. 1993. Van stroomdal naar droomdal; Integratie van hydrologisch en ecologisch onderzoek ten behoeve van het beheer in de Drentse A. Rapport 1993-2, Further information/further reading Staatsbosbeheer, Afdeling Terreinbeheer, Driebergen.

Bakker, J.P., Everts, H., Grootjans, A.P., De Vries N.P.J., De Vries Van Diggelen R., Grootjans, A.P., & Burkunk, R. (1994). Y. (2015). Het grote experiment – Vijftig jaar natuurbeheer. Assessing restoration perspectives of disturbed brook valleys: In: Spek, T., Elerie H., Bakker, J.P., Noordhoff, I. (eds.), the Gorecht area, The Netherlands. Restoration Ecology 2, Landschapsbiografie van de Drentsche Aa (pp. 418-461). 87-96. Koninklijke Van Gorcum BV, Assen. Van Diggelen, R., Beukema, H., & Noorman, K. J. (1995). Bregman. E.P.H., Maas, G., Makaske, B., & Meyles, E, (2015). Ranunculus hederaceus L. as indicator of land use changes in Vormgegeven door ijs, water en wind. In: Spek T, Elerie H, the Netherlands. Acta Botanica Neerlandica 44, 161–175. Bakker JP, Noordhoff I (eds): Landschapsbiografie van de Drentsche Aa, pp. 18-53. Van Gorcum, Assen. https://www.youtube.com/watch?v=QE8gDgXBzzA

Couwenberg, J. 2009. Methane emissions from peat soils (organic https://www.youtube.com/watch?v=OkXvyQZUXoM soils, histosoils) Facts, MRV-ability, emission factors. Wetlands International, Ede. https://youtu.be/eDC_H-sdB48

De Gans, W., 1981. The Drentsche Aa Valley System. A study in https://www.youtube.com/watch?v=hb1QFvsdia0

27 The Dwingelderveld; the largest wet heathland complex in Western Europe The Dwingelderveld; the largest wet heathland complex in Western Europe. Ab Grootjans, Gert Jan Baaijens Ab Grootjans, Gert Jan Baaijens Friday Aug 24th 13.30-17.00 Friday Aug 24th 13.30-17.00

Small heathland bogs in the National Park Dwingelderveld

he Dwingelderveld is a nature reserve in the south-west of the province of Drenthe. It is the Tlargest uninterrupted wet heathland in Western Europe. The area contains extensive wet and dry heath, small heathland bogs, remnants of small raised bogs, Figure 1: drift sands and juniper shrubs. Both the nature values and the location of the area within a virtually intact Map of ca. 1900 of the Dwingelderveld. Most of the area consisted of Heathland, with in the North farming-village landscape are unique. Therefore, the and North-East areas of sand blown dunes and Dwingelderveld was proclaimed as a National Park in afforestation with pins to stop the sand blowing. Blue lines represent old erosion gullies. 1991.

The area of ca. 3800 hectares (Figure 1) is not only designated as a protected nature reserve as part of

the national ecological network and is additionally both a Birds Directive and a Habitats Directive area. It

harbors 126 bird species (90 breeding birds), including

Small heathland bogs in the National Park Dwingelderveld the Crane, Woodlark, Black woodpecker, Whinchat,

The Dwingelderveld is a nature reserve in the south-west of the province of Drenthe. It is the largest Figure 1: uninterrupted wet heathland in Western Europe. The area contains extensive wet and dry heath, Map of 1995 showing that the pine plantations small heathland bogs, remnants of small raised bogs, drift sands and juniper shrubs. Both the nature have expanded a lot at the cost of the heath. The Map of ca. 1900 of the Dwingelderveld. Most of values and the location of the area within a virtually intact farming-village landscape are unique. agricultural enclaves in the Dwingelderveld are Therefore, the Dwingelderveld was proclaimed as a National Park in 1991. the area consisted of Heathland, with in the North indicated in yellow and originate from the and North-East areas of sand blown dunes and 1930ties. afforestation with pins to stop the sand blowing. Blue lines represent old erosion gullies.

Figure 1: Map of ca. 1900 of the Dwingelderveld. Most of the Map of 1995 showing that the pine plantations Map of 1995 showing that the pine plantations have expanded have expanded a lot at the cost of the heath. The area consisted of Heathland, with in the North and North-East Thea lot Province at the cost of theis heath.ensuring The agricultural that the enclaves objectives in the for the area are achieved. areas of sand blown dunes and afforestation with pins to stop agricultural enclaves in the Dwingelderveld are Dwingelderveld are indicated in yellow and originate from the Staatsbosbeheer (Dutch Forestry Service) and Natuurmonumenten (Dutch the sand blowing. Blue lines represent old erosion gullies. indicated in yellow1930ties. and originate from the 1930ties. Society for the Preservation of Nature) own and manage the majority of the Dwingelderveld.

28 With 822 mm annual precipitation the Dwingelderveld is one of the wettest parts of the Netherlands and the rain water cannot easily infiltrate into the soil. During the last glacial period a thick layer of boulder clay was deposited, which was later partly or totally eroded by old river systems (Figure 2).

48

The Province is ensuring that the objectives for the area are achieved. Staatsbosbeheer (Dutch Forestry Service) and Natuurmonumenten (Dutch Society for the Preservation of Nature) own and manage the majority of the Dwingelderveld.

With 822 mm annual precipitation the Dwingelderveld is one of the wettest parts of the Netherlands and the rain water cannot easily infiltrate into the soil. During the last glacial period a thick layer of boulder clay was deposited, which was later partly or totally eroded by old river systems (Figure 2).

48

Stonechat, Shoveler, and Common teal. The area has pines was carried out to stop the sand blowing. In also been designated for the rare Northern crested newt. 1936 about 50 ha of agricultural land was created in The Province of Drenthe is the competent authority and the center of the heathland area, which expanded has the ultimate responsibility for setting up the Natura to circa 200 ha in 1950; this is the Noordenveld. 2000 Dwingelderveld management plan. Deep drainage ditches were dug, lowering the water The Province is ensuring that the objectives for the table till 70cm below the surface. The drainage area are achieved. Staatsbosbeheer (Dutch Forestry water from the agricultural flooded the some of the Service) and Natuurmonumenten (Dutch Society for the remaining oligotrophic heath pools, resulting in severe Preservation of Nature) own and manage the majority eutrophication. Another problem was the lowering of the Dwingelderveld. of the water tables of the brook valleys situated on With 822 mm annual precipitation the both sides of the Dwingelderveld. The combination Dwingelderveld is one of the wettest parts of the of agricultural drainage, pine afforestation with Netherlands and the rain water cannot easily infiltrate higher evapotranspiration and lowering water tables into the soil. During the last glacial period a thick layer in the brook valleys led to severe desiccation of the of boulder clay was deposited, which was later partly or Dwingelderveld. totally eroded by old river systems (Figure 2).

Figure 2. Simplified geology and geomorphology of the Atmospheric Nitrogen deposition FigureDwingelderveld 2. Simplified area (from Grootjansgeology et al.and 1998) geomorphology (top) and of the Dwingelderveld area (from As in most areas in the Netherlands atmospheric Grootjansheath pool typeset al. in the1998) Dwingelderveld (top) and bottom). heath The pool small types in the Dwingelderveld bottom). The smallbogs alongbogs the along valley theslopes valley and in theslopes pingo andremnants in the pingonitrogen remnants deposition were is aonce serious influenced problem for bynutrient groundwater,were once influenced but are by groundwater,now fed bybut precipitationare now fed by water.poor ecosystems,The pools leadingabove to the grass boulder encroachment clay and have always been fed by precipitation water, and may have been formed relatively precipitation water. The pools above the boulder clay have soil acidification. In the 1960-1970 acidification of recentlyalways been (after fed byBakker precipitation et al., water, 1986; and mayBaaijens have been et al., 2018). precipitation water (due to SO2-emissions) had severely formed relatively recently (after Bakker et al., 1986; Baaijens Thiset al., widespread2018). occurrence of boulder clay indamaged the upper many heathland soil layers areas andhas bogs. led In to the 1990 much variation in hydrological conditions and tiesthus the a emission varied of heathland SO2 had been landscape reduced considerably, with Erica tetralix dominating the wet parts, andbut the occurrence atmospheric nitrogenof numerous deposition small still increased. bogs.This widespreadIn the dryer occurrence pa rtsof boulder Calluna clay invulgaris the upper isThe co effects-dominant. of increased In nitrogenthe past deposition (some has been hundredsoil layers hasyears led to ago),much variation drift insands hydrological were extensiverecorded wellin inthe the Dwingelderveld.dry parts Inof 1967/1968 the Dwingelderveldconditions and thus due a varied to heathland extensive landscape sheep with grazing.Baaijens Since measured the the beginningwater composition of theof ca 30 1920’sErica tetralix afforestation dominating thewith wet pines parts, and was occurrence carried outbogs to stopin the theDwingelderveld sand blowing. and estimated In 1936 met that the aboutof numerous 50 ha small of agricultural bogs. In the dryer land parts was Calluna created inannual the centernitrogen ofdeposition the heathland was about 22 area, kg N per ha/ whichvulgaris expanded is co-dominant. to circaIn the past200 (some ha inhundred 1950; this isyear the (Baaijens, Noordenveld. 1982). In 2002 Deep this research drainage was repeated ditchesyears ago), were drift sandsdug, were lowering extensive thein the water dry parts table (Verschoortill 70cm et al.below 2003) andthe it wassurfa foundce. that The the pH, the drainageof the Dwingelderveld water from due to extensivethe agricultural sheep grazing. floodedcalcium-, the ammonium- some of and the nitrate remaining concentration had all oligotrophicSince the beginning heath of pools,the 1920’s resulting afforestation in severewith eutrophication.increased: so acidification Another had problem dropped, waswhile nutrients the lowering of the water tables of the brook valleys situated on both sides of the Dwingelderveld. The combination of agricultural drainage, pine afforestation with 29 higher evapotranspiration and lowering water tables in the brook valleys led to severe desiccation of the Dwingelderveld.

Atmospheric Nitrogen deposition

As in most areas in the Netherlands atmospheric nitrogen deposition is a serious problem for nutrient poor ecosystems, leading to grass encroachment and soil acidification. In the 1960-1970 acidification of precipitation water (due to SO2- emissions) had severely damaged many heathland areas and bogs. In the 1990

49 had increased. The increased concentrations of calcium A cross-section through the erosion gulley (from north were explained by the rewetting measures since 1987 to south; figure 4) shows that during the winter period supplying the bogs with more groundwater. all heath bogs are in contact with the groundwater A detailed overview of changes in environmental (light blue), but still the water level of the bog itself is conditions in small bogs in the area during almost higher than the local groundwater level. In autumn the 50 years was presented by Van Dam et al. (2013). local groundwater table is lower (dark blue) and several They studied water chemistry, water levels, and the bogs have lost contact with the local groundwater, composition of vegetation, algae, and macrofauna especially in the right side of the transect. Groundwater in the bogs. Data from 2010/2011 and compared the has been lost in areas where boulder clay has been data with similar research in 1990, 1994, 2003, and eroded away or where the clay layer is very thin. If also older data. The found that during the last 10-20 the local groundwater drops below the bottom of the years the quality of most parameters had improved, bog the infiltration of bog water increases (Dekker mostly due to a steep decrease in SO2 (acidification) et al., 1986); the bog losses water to the groundwater. combined with large scale restoration measures. The And the local groundwater level is drained by the pH, for instance rose from 4.0 in 1990 to 5.5 in 2010. low levels in the central agricultural area and the But the macrofauna did not recover. The explanation of low water level in the small valleys on both sides of the authors was that mineralization of organic matter the Dwingelderveld. During most of the year almost in the bogs could be responsible. The last decades the no water flow occurs between the small bogs in the temperature in summer had increased by 1.6 degrees, erosion gully. In dry periods the bog loose water due while the sulphate concentration in the water had to evaporation and they lose (small amounts of water decreased. This could point to increased sulphate due to leakage through the organic layers. But in wet reduction in the peat, which together with higher pH periods the situation is different. After prolonged rains conditions, could have increased mineralization of and decreasing evapotranspiration the bogs are filled organic matter and associated availability of nutrients with precipitation water and start to overflow, loosing (see also Smolders et et al. 2006). So, the legacy of acidification and increased sulphate concentrations in the bog system, still had a negative impact on the macrofauna, even after the input atmospheric input of

SO2 had stopped decades ago.

Interactions between different

sand hydrological systems Figure 3 shows a conceptual model of possible groundwater flow in and around the heathland bogs in groundwater the Dwingelderveld. The bogs itself are mainly found in old gullies, which are remnants of past (erosive) surface Boulder clay water systems. These gullies have been filled with much larger sand bogs in the past. Most of these bogs have disappeared, partly by exploitation by man, but possible Figurealso due 3. PositionFigure of 3.small Position heathland of small heathland bogs (orange) bogs (orange) in former in former gullies (yellow), which are to desiccation of the whole area as a result dueunderlain to by imperviousgullies (yellow), podzolic which layers are underlain (B-horizons; by impervious black) podzolic and organic deposits (gyttja). The arrows indicate groundwater flows. The top segment shows the groundwater flows layers (B-horizons; black) and organic deposits (gyttja). The hydrological changes in wider surrounding withinplateaus the shallow sand layer (up to 5 meters). The bottom segment (5-10 meter below arrows indicate groundwater flows. The top segment shows and river valleys. But these old bogs left a marksurface) in the indicates groundwater flow above the boulder clay deposit, which have a high the groundwater flows within the shallow sand layer (up to form of impervious organic layers in the subsoilresistance (gyttja to downward water flow if the layer is more than 2 meters thick. When the boulder clay layer5 meters). is less The than bottom 50 segmentcm thick (5-10 groundwater meter below surface)infiltrates to lower sandy layers and podzolic layers). and flows to indicatesthe small groundwater (drained) flow aboveriver thevalleys boulder situatedclay deposit, on both sides of the The present bogs are in fact regeneratingDwingelderveld. Sphagnum which have a high resistance to downward water flow if mats that are floating on the water after farmers had the layer is more than 2 meters thick. When the boulder clay layer is less than 50 cm thick groundwater infiltrates to lower taken away most of the peat during the past centuries. sandy layers and flows to the small (drained) river valleys situated on both sides of the Dwingelderveld.

30

Figure 4: Cross-section through the length of an erosion gully showing the position of 8 small bogs. The groundwater level of winter 2002 and autumn 2002 is indicated in blue colors. Water levels drop to the right due to leakage through a thin layer of boulder clay in the Groot Koelevaartsveen, which in turn loses groundwater towards drainage ditches in a large agricultural enclave (changed after Verschoor et al. 2003). A= Wolfsklauw Veen, B = Zandveen, C = Barkman’s veentje, D = Groote Veen, E = Lange Veen, F = Groot Koelevaartsveen.

51

sand

groundwater

Boulder clay

sand

Figure 3. Position of small heathland bogs (orange) in former gullies (yellow), which are underlain by impervious podzolic layers (B-horizons; black) and organic deposits (gyttja). The arrows indicate groundwater flows. The top segment shows the groundwater flows within the shallow sand layer (up to 5 meters). The bottom segment (5-10 meter below surface) indicates groundwater flow above the boulder clay deposit, which have a high resistance to downward water flow if the layer is more than 2 meters thick. When the boulder clay layer is less than 50 cm thick groundwater infiltrates to lower sandy layers and flows to the small (drained) river valleys situated on both sides of the Dwingelderveld.

A cross-section through the erosion gulley (from north to south; figure 4) shows that during the winter period all heath bogs are in contact with the groundwater (light blue), but still the water level of the bog itself is higher than the local groundwater level. In autumn the local groundwater table is lower (dark blue) and several bogs have lost contact with the local groundwater, especially in the right side of the transect. Groundwater has been lost in areas where boulder clay has been eroded away or where the clay layer is very thin. If the local groundwater drops below the bottom of the bog the infiltration of bog water increases (Dekker et al., 1986); the bog losses water to the groundwater. And Figurethe local4: Cross-section groundwater through the length leve ofl anis erosion drained gully byGroundwater the low levelsfrom neighboring in the hillscentral was still agricultural relatively showingFigure the4: positionCross- ofsection 8 small bogs.through The groundwater the length level of an erosioncold, while gully shallow showing surface thewater position in the bog of itself 8 small was bogs. ofarea winter 2002and and the autumn low 2002 iswater indicated inlevel blue colors. in the small valleys on both sides of the The groundwater level of winter 2002 and autumnwarmed 2002 up. isOne indicated bog (called in ‘Poort blue 2’) colors.shows inflowing Water levels WaterdropDwingelderveld. levelsto the drop right to the due right During to due leakage to leakage most through through of athe thina thin year layer almost of bou lderno claywater in the flow Groot occurs Koelevaartsveen, between layer of boulder clay in the Groot Koelevaartsveen, which in cold groundwater from the neighboring sand hill at the which in turn loses groundwater towards drainage ditches in a large agricultural enclave (changed turnthe loses small groundwater bogs towards in the drainage erosion ditches ingully. a large In right`.dry periods the bog loose water due to after Verschoor et al. 2003). A= Wolfsklauw Veen, B = Zandveen, C = Barkman’s veentje, D = agriculturalevaporation enclave (changedand they after Verschoor lose (small et al. 2003). amounts A= In ofthe bogwater itself thedue water to isleakage heated up and through infiltrates the WolfsklauwGroote Veen, Veen, BE == Zandveen, Lange CVeen = Barkman’s, F = Grootveentje, Koelevaartsveen. D = organic layers. But in wet periods the situationagain to the is opposite different. side. In anotherAfter prolongedbog (called rains Groote Veen, E = Lange Veen, F = Groot Koelevaartsveen. and decreasing evapotranspiration the 51peatbogbogs nrare 4076) filled no incoming with groundwater precipitation can be water water and tostart the groundwater to overflow, in the gullies.loosing Some water bogs also to thedetected. groundwater Temperature profilesin the aregullies. stratified.; Some warm inbogs receivealso localreceive groundwater local from groundwater surrounding dunes from surroundingthe upper layers dunes(very recent and warming) the andBog in theAsphodel lower and the Bog Asphodel (Narthecium ossifragum) is often layers as well, (warming from last summer). (Narthecium ossifragum) is often abundant in such bogs (Van Dijk et al. 2009; abundant in such bogs (Van Dijk et al. 2009; see also The importance of additional inflow of ground- or see also Figure 5) Figure 5) surface water to a growing bog has been researched Rooke (2002) studied the relationships between the by Patberg et al., 2013). Between 2007 and 2009 they Rooke (2002) studied the relationships between the occurrence of Narthecium occurrence of Narthecium ossifragum and possible flow ossifragum and possible flow of local groundwater. As a tracer for local of local groundwater. As a tracer for local groundwater Figure 5. Overview of two small bogs with active growing Sphagnum carpets (left: ‘Barkmansveen’ with flowering flowgroundwater he used temperature flow heprofile used in the temperature bogs (Fig. 6). profile in the bogs (Fig. 6). Rhynchospera alba, and right ‘Poort 2’ with an abundance of The measurements were carried out in spring Bog Asphodel (Narthecium ossifragum) (fruit setting). Photos: 2001 during a period of rathe high temperatures. André Jansen.

Figure 5. Overview of two small bogs with active growing Sphagnum carpets (left:31 “Barkmansveen” with flowering Rhynchospera alba, and right “Poort 2” with an abundance of Bog Asphodel (Narthecium ossifragum) (fruit setting). Photos: André Jansen.

The measurements were carried out in spring 2001 during a period of rathe high temperatures. Groundwater from neighboring hills was still relatively cold, while shallow surface water in the bog itself was warmed up. One bog (called “Poort

52

2”) shows inflowing cold groundwater from the neighboring sand hill at the right`.

Figure 6: Temperature Poort 2 profiles in two small bogs in > 8 the Dwingelderveld. In the upper bog with Narthecium

7,5 - 8 ossifragum, cold groundwater is flowing into the bog from the right. In the small bog without 7- 7,5 Narthecium ossifragum (‘4076’) 6 m the temperature profile shows 0,5 m Veentje 4076 only horizontal stratification 6,5 - 7 indicating that no water movement is (after: Rooke 2002). < 6,5

sand

analyzed the composition of bog water during the development were based on the growth of Sphagna that Figureseason 6: Temperatureto investigate whatprofiles could in be two limiting small factors bogs in the couldDwingelderveld. be seen on aerial In the photographs upper bog of 1982 with and 2006. Nartheciumfor Sphagnum ossifragum, growth in cold the field.groundwater Patberg iswanted flowing to into theFrom bog this from analyses the right.Patberg In concluded the small that bog Sphagnum without Narthecium ossifragum (“4076”) the temperature profile shows only horizontal know that successful restoration of Sphagnum growth growth was mainly determined by differences in CO , stratification indicating that no water movement is (after: Rooke 2002). 2 was indeed depended on the position of the bog in the iron and silica concentrations of the surface water In thelandscape bog itselfas was suggestedthe water by Verschooris heated et al., up (2003). and infiltrates(table 1). again to the opposite side. In Inanother his research bog Patberg (called distinguished peatbog two categoriesnr 4076) no incomingDifferences ingroundwater nutrient concentrations can be were not detected.of bogs; wellTemperature developing bog profilesvegetation areafter rewettingstratified.; significant,warm in indicatingthe upper input layers of groundwater (very with recent(‘good’) warmin and notg) well and developing in the bog lower vegetation layers as well, high(warming concentrations from oflast CO2 ,summer). iron and silica were after rewetting (‘bad’). ‘Good’ and ‘bad’ vegetation more important for abundant Sphagnum growth than The importance of additional inflow of ground- ornutrient surface concentrations water to ina thegrowing water. bog has been researched by Patberg et al., 2013). BetweenThe research 2007of Patberg and et al. 2009 (2013) confirmedthey Bog surface water analyzed the composition of bog water during experimentalthe season research to investigate on limiting factors what of Sphagnum good bad significant could be limiting factors for Sphagnum growthspecies in the carried field. out byPatberg Smolders wanted et al. (2003). to But where – HCO3 15 ± 18 25 ± 51 n.s. know that successful restoration of Sphagnum doesgrowth this COwas2 and indeed iron rich dependedwater come from? on There are CO2 1215 ± 730 743 ± 626 * the position of the bog in the landscape as wastwo suggested main sources by of CO Verschoor2 : (i) from local et groundwater al., pH 4 ± 0.3 5 ± 1 * (2003). In his research Patberg distinguishedthat two picked categories it up from the of root bogs; zone in surroundingwell developing bog vegetation after rewetting (“good”)infiltration and notareas, well and (ii) developing from peat decomposition bog in NH4 77 ± 70 72 ± 80 n.s. vegetation after rewetting (“bad”). “Good” andremnant “bad” peat vegetation in the bog itself. development Farmers left a shallow NO3 8 ± 14 9 ± 15 n.s. were based on the growth of Sphagna that couldlayer be of seenpeat in onthe pastaerial and photographsthat appeared to be an K 26 ± 19 34 ± 28 n.s. of 1982 and 2006. From this analyses Patberg concludedimportant source that of SphagnumCO2- for the growing growth Sphagnum P 5 ± 8 4 ± 7 n.s. was mainly determined by differences in CO2, speciesiron and at the silica surface concentrations (Smolders et al., 2001, of Tomassen et al. 2011). the surfaceAl water10 ± (table8 1).8 ± 9 n.s. Ca 31 ± 35 34 ± 32 n.s. Differences in nutrient concentrations were notFlow significant, of water indicating in the actively input growingof Cl 256 ± 80 290 ± 152 n.s. groundwater with high concentrations of COSphagnum2, iron and layer silica (acrotelm); were more Buoyency Fe 42 ± 107 12 ± 14 * important for abundant Sphagnum growth thandriven nutrient water concentrations flow in the Mg 26 ± 14 30 ± 13 n.s. water. Na 192 ± 48 227 ± 95 * Water flow on the microscale can occur in the S 22 ± 16 28 ± 22 n.s. living Shpagnum layer of small bogs (acrotelm). The mat of growing Sphagnum plant is floating on the Si 36 ± 22 19 ± 22 * 53 water. During the day the temperature in the top

Table 1 Chemical composition of surface water in well layer can increase considerably in the top 10-15 cm, developing and badly developing bogs after restoration while during the night the top layer is cooled down. measures in the Dwingelderveld (in µmol/l).The last column indicates if difference between the two bog types are Especially in spring and autumn relatively large significantly different (*) or not (n.s). differences in temperatures can occur between day

32 and night. Around midnight a cold layer of surface Restoration measures to restart water is covering a warm layer of water that was Sphagnum growth warmed op the previous day. Cold water is denser than After the establishment of the National Park in 1991 warm water and tends to drop to lower layers, while measures were taken to reduce the effects of drainage, warm water tends to move upwards. The shallow peat both inside the park and also outside. Inside the layers is preventing such water flow if the differences Park drainage ditches that had been dug in the wet between day and night temperatures are small. But erosion gullies to improve the productivity of the pine when the differences are larger than 10 degrees and plantations, were eliminated and pines were cut in the the resistance to water flow in the upper peat layer is immediate surroundings of the bogs. In some areas low, then water flow occurs in the top layer during the with larger bog remnants pine forest was cut. Outside night. Afterthe reserve the agriculturalestablishment drainage of theditches National at the borders Park in 1991 measures were taken to reduce the effects of drainage, both inside the park and also outside. Inside the This phenomenon is called buoyency- driven flow of the Park were also removed and weirs were placed Park drainage ditches that had been dug in the wet erosion gullies to improve the of water. This mechanism in which water from deeper in the agricultural channels in order to keep more Afterproductivity the establishment of the pine plantations, of the National were eliminated Park in and 1991 pines measures were cut werein the taken to parts of the acrotelm mixes with precipitation water water in the area and to have a more stable fluctuation reduceimmediate the surroundingseffects of drainage, of the bogs. both In insidesome areasthe park with andlarger also bog outside. remnants Inside the in the top has been described in bogs by Rappoldt et al. pattern in the ditches. This also reduce the groundwater Parkpine drainageforest was ditches cut. Outside that had the beenreserve dug agricultural in the wet drainage erosion ditchesgullies toat improvethe the (2003). This mechanism has been described in various bordersfluctuation of inthe the Park heathlands were alsoof the removed Park. and weirs were placed in the agricultural productivity of the pine plantations, were eliminated and pines were cut in the types of soils and in lakes, but was not known in bogs. channelsIn 2010/2011 in order the agricultural to keep enclavesmore water within in the the area and to have a more stable immediate surroundings of the bogs. In some areas with larger bog remnants But in the 1970-ties G.J. Baaijens already mentioned fluctuationPark were finally pattern bought in theby the ditches. government This andalso large reduce the groundwater fluctuation in pine forest was cut. Outside the reserve agricultural drainage ditches at the the possible importance of ‘warm water pumps’ for the thescale heathlands restoration measuresof the Park. were carried out. The borders of the Park were also removed and weirs were placed in the agricultural availability of nutrients in bogs. He formulated this largest enclave was the ‘Noordenveld’ that had been channels in order to keep more water in the area and to have a more stable hypothesis based on day and night measurements of a big obstacle to create a more natural hydrological the pH in the topsoil of bogs. fluctuationsystem in the patternDwingelderveld in the (Figures ditches. 8 and This9). The also reduce the groundwater fluctuation in Figure 7 illustrates the occurrence of buoyency driven theagricultural heathlands ditches of were the removed Park. and in a large area coveringflow of watera warm in an experimentallayer of waterSphagnum that recurvum was warmed op the previous day. Cold water is denser than warm water and tends to drop to lower layers, while warm bog in the laboratory (Rappoldt et al., 2003; Adema et Figure 8. Former water tends to move upwards. The shallow peat layers is preventing such water al., 2006). The authors point to the possible importance drainage flowof buoyencyif the differences driven flow ofbetween water for theday availability and night of temperatures are small. But when ditch in the the differences are larger than 10 degrees and the resistance to water flow in Noordenveld. nutrients and CO2 for Sphagnum growth in the top layer. the upper peat layer is low, then water flow occurs in the top layer during the This was later confirmed experimentally by Patberg Figure 8. Former drainage ditch in the Noordenveld night. (2012).

0 14.75 - 15.00 ºC 14.50 - 14.75 ºC 14.25 - 14.50 ºC 5 14.00 - 14.25 ºC Figure 8. Former drainage ditch in the Noordenveld 13.75 - 14.00 ºC 13.50 - 13.75 ºC 10 13.25 - 13.50 ºC X 13.00 - 13.25 ºC 12.75 - 13.00 ºC 15 12.50 - 12.75 ºC 12.25 - 12.50 ºC 12.00 - 12.25 ºC 20 11.75 - 12.00 ºC 11.50 - 11.75 ºC

0 5 10 15 Y Figure 7: Spatial differentiation in temperature in an Figure 7: Spatial differentiation in temperature in an experimental desighn where difference in day experimental desighn where difference in day and night Figure 9. Aerial photograph of the experimental site in and night temperatures were imposed on a floating mat of Sphagnum recurvum. The differentiation temperatures were imposed on a floating mat of Sphagnum February 2012. (dry site in purple and wet site in blue). in temperature at a scale of 20 by 20 cm occurred at midnight when warm water from deeper layer recurvum. The differentiation in temperature at a scale of Clearly visible are the at that time still unexcavated was raising to the surface, while parts the cold water was still present (from: Baaijens et al. 20 by 20 cm occurred at midnight when warm water from Figureagricultural 9. Aerial soils photographsurrounding ofthe the experimental site in February 2012. (dry site in purple and wet 2018). site in blue). Clearly visible are the at that time still unexcavated agricultural soils surrounding the deeper layer was raising to the surface, while parts the cold Experimental plots and the mounds of removed topsoil next Experimental plots and the mounds of removed topsoil next to the dry site. water was still present (from: Baaijens et al. 2018). to the dry site. This phenomenon is called buoyency- driven flow of water. This mechanism in which water from deeper parts of the acrotelm mixes with precipitation water in the top has been described in bogs by Rappoldt et al. (2003). This mechanism 33 has been described in various types of soils and in lakes, but was not known in 56 bogs. But in the 1970-ties G.J. Baaijens already mentioned the possible importance of ‘warm water pumps’ for the availability of nutrients in bogs. He Figure 9. Aerial photograph of the experimental site in February 2012. (dry site in purple and wet formulated this hypothesis based on day and night measurements of the pH in site in blue). Clearly visible are the at that time still unexcavated agricultural soils surrounding the the topsoil of bogs. Experimental plots and the mounds of removed topsoil next to the dry site.

Figure 7 illustrates the occurrence of buoyency driven flow of water in an experimental Sphagnum recurvum bog in the laboratory (Rappoldt et al., 2003; Adema et al., 2006). The authors point to the possible importance of buoyency 56 driven flow of water for the availability of nutrients and CO2 for Sphagnum growth in the top layer. This was later confirmed experimentally by Patberg (2012).

Restoration measures to restart Sphagnum growth

55

In 2010/2011 the agricultural enclaves within the Park were finally bought by the government and large scale restoration measures were carried out. The largest enclave was the “Noordenveld” that had been a big obstacle to create a more natural hydrological system in the Dwingelderveld (Figures 8 and 9). The agricultural ditches were removed and in a large area also the fertile top soil was removed. By doing this also the eutrophication of surrounding wet heathlands by polluted surface water came to an end.

Evaluation of restoration measures

Ten years after the start of (hydrological) restoration measures, the success of 35 rewetted bogs was assessed (Everts et al., 2002; Baaijens et al., 2005). The presence of a well-developed and characteristic bog vegetation was the main criterium for restoration success. Three categories of success were distinguished (Fig. 10): (i) successful, (ii) less successful, but promising and (iii) not yet successful. An interesting pattern was found on the landscape scale; bogs within the former erosion gullies were more successfully restored compared to bogs situated outside the gully or at the beginning of a gully. Apparently the position within the gullies was important for successful restoration, because water losses were small within the gullies

Further information/further reading N 1 km Adema EB, Grootjans AP, van Belle J Smolders F & Baaijens GJ (2006). Buoyency driven flow in peat bogs; evidence for nutrient cycling in the field. Journal of Hydrology 327: 226-234.

Local watershed Baaijens GJ (1982). Natuur en landschap in Dwingelo. In: Smit, R. (ed.). Fragmenten uit de geschiedenis van Dwingelo, pp 211- 245. Stichting het Drentse boek. Ruinen.

Barkman JJ & Baaijens GJ (1992). Plant communities and synecology of bogs and heath pools in The Netherlands. In: Verhoeven, JTA. (ed.), Fens and Bogs in The Nether¬lands, pp. 173-235. Geobota¬ny 18. Kluwer, Dordrecht. Erosion gully Casparie WA (1972). Bog development in southeastern Drenthe Restoration successful (the Netherlands). Vegetatio 25: 1-271. Less successful De Gans, W., 1981. The Drentsche Aa Valley System. A study in Not yet successful quaternary geology. Proefschrift. Rodopi, Amsterdam. Figure 10. Evaluation of restoration success of 35 small bogs Grootjans AP, Adema EB, Baaijens GJ, Rappoldt C. & Verschoor ten years after implementation of hydrological measures Figure 10. Evaluation of restoration success of 35 small bogs ten years A.after (2003). implementation Mechanisms behindof restoration of small bog to rewet the bogs (after: Verschoor et al. 2003). The arrow hydrological measures to rewet the bogs (after: Verschoor et al. 2003). Theecosystems arrow indicate in a cover the sandflow landscape. Archiv für Naturschutz directionindicate of surface the flow water direction within theof surface former watererosion within gullies. the und Landschaftsforchung 2003: 43-48. former erosion gullies. Lanting AH (2009). Heath land bogs and pingo remnants in Lheebroekerzand and Gasselterveld. Are they sensitive to changes in regional water tables? Master thesis, IVEM/EES, also the fertile top soil was removed. By doing57 this also University of Groningen. the eutrophication of surrounding wet heathlands by polluted surface water came to an end. Patberg, W., Baaijens, GJ, Elzinga, T., & Grootjans, A.P. 2013. The importance of carbon dioxide in the development of Evaluation of restoration measures Sphagnum bogs: a field study. Preslia 85: 389–403. Ten years after the start of (hydrological) restoration Rappol M (1984). Till in southeast Drenthe and the origin of the Hondsrug complex, the Netherlands. Eiszeitalter und measures, the success of 35 rewetted bogs was assessed Gegenwart 34:7-27. (Everts et al., 2002; Baaijens et al., 2005). The presence Rappoldt C, Adema EB, Baaijens GJ, van Duijn CJ, Grootjans of a well-developed and characteristic bog vegetation AP &Pieters GJJM (2003). Buoyancy-driven flow in a peat was the main criterium for restoration success. Three bogs as a mechanism for nutrient cycling. Proceedings of the categories of success were distinguished (Fig. 10): (i) National Academy of Science of America 100: 14937-14942.

successful, (ii) less successful, but promising and (iii) Smolders AJP, Tomassen HBM, Pijnappel HW, Lamers LPM & not yet successful. An interesting pattern was found Roelofs JGM (2001). Substrate derived CO2 is important in the development of Sphagnum spp. New Phytologist (2001) 152: on the landscape scale; bogs within the former erosion 325-332. gullies were more successfully restored compared to bogs situated outside the gully or at the beginning of https://www.youtube.com/watch?v=QE8gDgXBzzA

a gully. Apparently the position within the gullies was https://www.youtube.com/watch?v=OkXvyQZUXoM important for successful restoration, because water https://youtu.be/eDC_H-sdB48 losses were small within the gullies. After 40 years Natuurmonumenten and https://www.youtube.com/watch?v=hb1QFvsdia0 Staatsbosbeheer have succeeded in procuring the last agricultural enclaves within the Dwingelderveld, which has greatly increased restoration prospects of the National Park. However the focus is now on reducing the atmospheric nitrogen deposition reaching the area from intensively used agricultural areas in the surrounding. Also the deep drainage of such areas remains worrying.

34 The National Park Weerribben-Wieden; The Nationaldeveloping Park Weerribben fens in-Wieden an infiltration; area Verhoeven & Bobbink, 2001). These conditions were quite common in The developing fens in an infiltration area. Netherlands until the 1950s and occurred in groundwater discharge areas and floodplain areas of streams and small rivers. Complexes of turf ponds with these Annemieke Kooijman, AnnemiekeCasper Cusell & Iwan Mettrop. Kooijman, Casper Cusellterrestrializing & Iwan Mettrop fen systems can be found in The Netherlands where the Holocenic low fen areas border the higher sandy areas of Pleistocenic origin, e.g. De th Saterday Aug 25th Saterday09.00-16.00 Aug 25 09.00-16.00 Wieden, De Weerribben and the Vechtplassen area.

Figure 1. Excursion sites in National Park Weerribben- Wieden (from: The National Park Weerribben- Cusell 2014). Wieden The National Parkhe NationalWeerribben Park Weerribben-Wieden-Wieden is one of the largest RAMSAR and Natura 2000 Figure 1. wereExcursion quite sites common in National in The ParkNetherlands Weerribben until-Wieden the (from: Cusell 2014). The National Park Weerribben-Wieden is one of the largest RAMSAR and Natura wetland areas in NW-Europe, and located in 1950s and occurred in groundwater discharge areas 2000 wetland Tareas in NW-Europe, and located in NWThe- Overijssel.National park The is areaa hotspot for wetland diversity, including characteristic originally consistedNW-Overijssel. of peat bog, The area which originally was consisteddrained ofand peat brownmossesconverted and floodplainto of agriculture mineral areas-rich of fstreamsens, and and the small only rivers. place in the Netherlands where in the 15th-16th bog,Century. which was Besides drained lakes, and converted the area to agriculture has severalthis in habitat complexesComplexes type (H4170A) of of turf turf ponds is stillwith relatively these terrestrializing common. Thefen area is also home to th th ponds, which havethe 15 terrestrialized-16 Century. Besides in the lakes, past the 100 area years. has severalhabitat Turf pond directivesystems complexes can plants be found such in inas The Liparis Netherlands loeselii whereand Hamatocaulis the vernicosus, and The Netherlandscomplexes have mostly of turf ponds,been whichexcavated have terrestrialized below themany water birds,Holocenic table butterflies in lowthe fen 18and areasth dragonflies. border the higherThis excursion sandy areas takes us to four different rich-fen complexes, in which several management problems and solutions have and early 19th incentury. the past 100They years. are Turf now pond characterized complexes in The by veryof species Pleistocenic-rich origin, fen e.g. De Wieden, De Weerribben vegetation. TheNetherlands plant communities have mostly inbeen the excavated terrestrializing below the fens inand turf the Vechtplassenponds are area. undergoing successionwater table infrom the 18thaquatic and early to19th century.semi-terrestrial They are vegetationThe National under park is a hotspot for60 wetland diversity, mesotrophic andnow slightlycharacterized eutrophic by very species-rich conditions fen vegetation.(Van Wirdum including et al. characteristic 1992, brownmosses of mineral-rich The plant communities in the terrestrializing fens in fens, and the only place in the Netherlands where this turf ponds are undergoing succession59 from aquatic habitat type (H4170A) is still relatively common. The to semi-terrestrial vegetation under mesotrophic area is also home to habitat directive plants such as and slightly eutrophic conditions (Van Wirdum et al. Liparis loeselii and Hamatocaulis vernicosus, and many 1992, Verhoeven & Bobbink, 2001). These conditions birds, butterflies and dragonflies. This excursion takes

35 been studied (Cusell 2014, Mettrop 2015):Stobbenribben, Veldweg, Kiersche Wiede and Meppelerdieplanden (Figure 1).

Human impact on hydrology and water composition

Before humans impacted the area, this part of NW-Overijssel was covered with mires, first fens and later also large bogs. The bog dried out by natural causes. From the 15th century onwards, peat was cut on an ever increasing scale. Most of the drained peat lands were converted into agricultural meadows. In the 1940- ties part of the former Zuiderzee was diked and drained. The new low water levels in this polder, called the Noordoostpolder (North-East Polder), causes a severe drop in water levels in the surrounding part of NW-Overijssel. Together with intensification of agricultural and associated deep drainage of the peatlands severe soil subsidence in the agricultural areas was initiated. This led to the situation that the remaining wetlands and fens were the highest areas in the landscape and were surrounded by low lying polders; seepage areas had become infiltration areas and this was irreversible (Figures 2 and 3).

a b

Figure 2. (a) NW-Overijssel as a pristine landscape with us to four different rich-fen complexes, in which several Figure 2. (a) NW-Overijssel as a pristine landscapegrowing bogs surroundedwith growing by fens. (b)bogs After thesurrounded mires had by management problems and solutions have been studied dried out and were later reclaimed for peat digging, open fens(Cusell. (b) 2014, After Mettrop the 2015): mires Stobbenribben, had dried Veldweg, out and waterwere bodies later were reclaimed left in which wherefor peatterrestrializing digging, open waterKiersche bodies Wiede and were Meppelerdieplanden left in which (Figure where1). terrestrializingmires developed. Intensive mires agriculture developed. initiated severe Intensive peat agriculture initiated severe peat subsidence, subsidence,which turned which turned a thea the former exfiltration exfiltration area area Human impact on hydrology and into a infiltration area. These hydrological changes were into a infiltration area. These hydrological changesaccelerated were by the reclamationaccelerated of the Noord-Oostby the Polderreclamation in of waterthe Noord composition-Oost Polder in part of the formerpart Zuider of the former Zee. Zuider This Zee. cause This cause also also aa severe severe drop drop in Beforegroundwater humans impacted levels the in area, NW this-Overijssel part of (right infigure groundwater after levels Van in NW-OverijsselWirdum et (right al. figure 1992). after Van Wirdum et al. 1992). NW-Overijssel was covered with mires, first fens and Nowadays,later also large thebogs. Theremaining bog dried out wetland by natural of aboutareas 9500 was initiated. ha has This anled toaverage the situation surface that the level ofcauses. 0.3 –From 0.6 the m 15 thbelow century onwards,mean peatsea was level cut (BMSL).remaining wetlands and fens were the highest areas on an ever increasing scale. Most of the drained peat in the landscape and were surrounded by low lying lands were converted into agricultural meadows. In polders; seepage areas had become infiltration areas the 1940-ties part of the former Zuiderzee was diked and this was irreversible (Figures 2 and 3). and drained. The new low water levels in this polder, Nowadays, the remaining wetland of about 9500 ha called the Noordoostpolder (North-East Polder), has an average surface level of 0.3 – 0.6 m below mean causes a severe drop in water levels in the surrounding sea level (BMSL). part of NW-Overijssel. Together with intensification The most species-rich succession stage is the of agricultural and associated deep drainage of the floating rich fen, which belongs to the base-rich peatlands severe soil subsidence in the agricultural transition fens (H4170A) in the EU-habitat directive. NP 61 Weerribben-Wieden is the only area in the Netherlands

in which this habitat type is still relatively common. Unfortunately, these base-rich fens are also the most threatened. One problem is that formation of new ones has not yet been observed, not even in the Weerribben- Wieden, where the terrestrialization process is more advanced than in other Dutch wetland areas. The other problem is that succession towards Sphagnum peatlands is going unnecessarily fast, due to high atmospheric N-deposition, P-eutrophication of surface water and insufficient buffer capacity. However, in the Weerribben-Wieden area, we can see that rich fens can be preserved much longer than expected, if sufficient Figure 3. Impression of the hydrological system of the FigureWeerribben 3. Impression in NW-Overijssel of the with hydrological low lying polder system areas. of thebase-rich Weerribben and nutrient-poor in NW-Overijssel (surface) water with is supplied. low lyingSurface polderwater is pumpedareas. fromSurface these agriculturalwater is areas pumped from Surfacethese wateragricultural levels are allowedareas intoto fluctuate the into the lakes and terrestrializing fens. In most of the area lakes and terrestrializing fens. In most of the area thebetween precipitation 0.73 – 0.83 water m BMSL is throughoutforming rai then year. the precipitation water is forming rain water lenses in the The surrounding polders have one to two meter lower waterfloating lenses mats in thewhich floating rapidly causes mats acidification which rapidly of the topcauses acidification of the top layers (after Van layersDiggelen (after etVan al. Diggelen 1996). et al. 1996). surface levels, and are drained by about 30 pumping

The most species-rich succession stage is the floating rich fen, which belongs to 36 the base-rich transition fens (H4170A) in the EU-habitat directive. NP Weerribben-Wieden is the only area in the Netherlands in which this habitat type is still relatively common. Unfortunately, these base-rich fens are also the most threatened. One problem is that formation of new ones has not yet been observed, not even in the Weerribben-Wieden, where the terrestrialization process is more advanced than in other Dutch wetland areas. The other problem is that succession towards Sphagnum peatlands is going unnecessarily fast, due to high atmospheric N-deposition, P-eutrophication of surface water and insufficient buffer capacity. However, in the Weerribben-Wieden area, we can see that rich fens can be preserved much longer than expected, if sufficient base-rich and nutrient-poor (surface) water is supplied.

Surface water levels are allowed to fluctuate between 0.73 – 0.83 m BMSL throughout the year. The surrounding polders have one to two meter lower surface levels, and are drained by about 30 pumping stations which pump excess water into the higher lying wetland. As a result, there is almost no direct discharge of groundwater anymore in National Park Weerribben-Wieden, and the remaining wetland is losing water to the surrounding polders by groundwater recharge. In the wetland reserve itself, surface water levels are regulated by one main pumping station at the western border of the nature reserve, which removes water during wet periods, and sporadically pumps water in during pronounced dry periods. However, water is coming in from the surrounding agricultural polders in wet periods.

Water balance studies for National Park Weerribben-Wieden (Cusell 2014) show that the water input of the present wetland consists of rainwater (about 35%), drainage water from the adjacent upland (about 20%), and water pumped in

62 stations which pump excess water into the higher lying fens (H7140B) and 17 kg N ha-1 year-1 for Scorpidium- wetland. As a result, there is almost no direct discharge dominated rich fens (H7140A). Almost all P comes from of groundwater anymore in National Park Weerribben- the low-lying agricultural polders (Figure 4), which may Wieden, and the remaining wetland is losing water lead to eutrophication of ponds and fens. An additional to the surrounding polders by groundwater recharge. problem is that almost all Ca comes from these polders In the wetland reserve itself, surface water levels are as well. As groundwater seepage is non-existent in the regulated by one main pumping station at the western National Park, supply of base-rich surface water with border of the nature reserve, which removes water during high concentrations of calcium and bicarbonate is the wet periods, and sporadically pumps water in during only way to keep the buffer capacity in base-rich fens, pronounced dry periods. However, water is coming in the most diverse but also most vulnerable habitat type, from the surrounding agricultural polders in wet periods. high enough. Water balance studies for National Park Weerribben- Wieden (Cusell 2014) show that the water input of More natural water levels: discussion the present wetland consists of rainwater (about 35%), and experiments drainage water from the adjacent upland (about 20%), In the Netherlands, hydrology of most wetlands is and water pumped in from lower lying agricultural highly artificial. Most areas have fixed water levels, polders (about 45%). Mean annual precipitation which are artificially kept at a particular height, is about 800 mm. The discharge of polder water required by law. Water is pumped in from other is about 50% smaller in summer than in winter, sources in dry periods, and pumped out in wet periods. due to the precipitation surplus in winter and the This water is often polluted, and many wetland areas evapotranspiration surplus during large parts of the have problems with surface water quality. One of the summer. potential solutions may be more natural water levels, or The water composition in the present wetland is larger fluctuations in the actual level. The idea is that thus largely determined by season and especially the when water levels are allowed to fluctuate, it is possible land use of the surrounding polders. During the second to keep surplus water inside the system in winter, so half of the 20th century, when new polders were created that there is less need to pump in polluted water from and manure was being used excessively, which led to elsewhere in summer. More fluctuating water levels severelyfrom lowerincreased lying N-and P-inputsagricultural into the Nationalpolders (aboutmay have 45%). a number Mean of positive annual effects, precipitationsuch as (i) lower is Park.about For N,800 atmospheric mm. Thedeposition discharge was an important of polder input water of polluted is about water in50% general, smaller (ii) lower waterin summer additionalthan in input. winter, due to the precipitationlevels in summer surplus which could in promote winter germination and the evapotranspirationAlthough the present estimated surplus deposition durin ofg about large partsof shoreline of the species, summer. (iii) higher water levels in winter 19 kg N ha-1 year-1 is lower than in many other Dutch which could improve breeding grounds for fishes and areas,The itwater is still above composition the critical N-deposition in the of present (iv)wetland flooding ofis the thus rich fens largely in winter, determinedwhich could by aboutseason 10 kg and ha-1 year-1 especially for Sphagnum the-dominated land use poor of theincrease surrounding buffer and reducepolders. acidification During potential. the second For some isolated wetland areas this may work to some Figurehalf 4.of Mean the annual 20th input century, of N, P and Ca when in NP Weerribben- new polders were created and manure was being degree. However, for Weerribben-Wieden in particular, Wieden.used Aboutexcessively, half of N comes fromwhich atmospheric led deposition,to severely increased N-and P-inputs into the but almost all P from the surrounding low-lying agricultural there may also be a number of negative effects, such National Park. For N, atmospheric deposition was an important additional input. polders. However, almost all Ca, needed to maintain high as (i) larger accumulation of rain water inside the area buffer capacity in base-rich fens, comes from the agricultural which leads to lower buffer capacity and increased polders as well. Data are from Cusell (2014).

Figure 4. Mean annual input of N, P and Ca in NP Weerribben-Wieden. About half of N 37 comes from atmospheric deposition, but almost all P from the surrounding low-lying agricultural polders. However, almost all Ca, needed to maintain high buffer capacity in base-rich fens, comes from the agricultural polders as well. Data are from Cusell (2014).

Although the present estimated deposition of about 19 kg N ha-1 year-1 is lower than in many other Dutch areas, it is still above the critical N-deposition of about 10 kg ha-1 year-1 for Sphagnum-dominated poor fens (H7140B) and 17 kg N ha-1 year-1 for Scorpidium-dominated rich fens (H7140A). Almost all P comes from the low-lying agricultural polders (Figure 4), which may lead to eutrophication of ponds and fens. An additional problem is that almost all Ca comes from these polders as well. As groundwater seepage is non-existent in the National Park, supply of base-rich surface water with high concentrations of calcium and bicarbonate is the only way to keep the buffer capacity in base-rich fens, the most diverse but also most vulnerable habitat type, high enough.

More natural water levels: discussion and experiments

In the Netherlands, hydrology of most wetlands is highly artificial. Most areas have fixed water levels, which are artificially kept at a particular height, required by law. Water is pumped in from other sources in dry periods, and pumped out in wet periods. This water is often polluted, and many wetland areas have problems with surface water quality. One of the potential solutions may be more natural water levels, or larger fluctuations in the actual level. The idea is that when water levels are allowed to fluctuate, it is possible to keep surplus water inside the system in winter, so that there is less need to pump in polluted water from

63 acidification, (ii) severe drought in summer in the base- Year Phanerogam N:P ratio Number of rich fens which may lead to acidification and increased biomass (g m-2) (g g-1) replicates mineralization of C, N and P, and (iii) input of N and P 1984 1123 (241) 16.0 (1.8) n = 3 (with from the agricultural surroundings especially in winter, 10 subplots) which means that flooding of the rich fens may occur 1990 512 (73) 19.1 (3.1) n = 5 with highly polluted water. We were able to study some 2005 212 (136) 22.4 (2.2) n = 5 of these questions (Cusell 2014, Mettrop 2015), and will 2010 187 (110) . n = 5 discuss some of the results in the four field sites. 2011 229 (96) 23.7 (1.5) n = 3 2012 287 (74) 22.1 (1.1) n = 5 De Stobbenribben: reference site with occasional ditch flooding Table 1. Changes in aboveground biomass and foliar N:P ratio of vascular plants in the Stobbenribben fens between 1984 The Drentsche Aa The Stobbenribben is located in the and 2012 (Kooijman et al. 2016). Weerribben and consists of four adjacent rectangular ponds of approximately 20-40 m width and 200 m hence increased uptake of nutrients along the way. length; depth of the ponds to the Pleistocene sandy Also, the local ditch at the northeastern side of the fen underground is approximately 2.5-3 m (Van Wirdum complex was cleaned and enlarged. 1991). The complex is one of the best example of floating In 1988, the Stobbenribben belonged to the best rich fens, with many characteristic rich-fen species, such fens of the Netherlands in 1988 (Kooijman & Paulissen as S. scorpioides and the EU Habitat directive species 2006; Kooijman et al. 2016). Brownmosses such as S. Liparis loeselii. The fens are annually mown to prevent scorpioides were present almost everywhere, except establishment of shrubs and trees. In the southeast, for the most isolated end dominated by Sphagnum the complex borders an agricultural polder, which has and the area adjacent to the ditch, dominated by been drained in the 1950s and is situated several meters eutrophic mosses. In the past decades, the fens both lower than the fens themselves. As a result, the fens are deteriorated and improved. In 2013, large parts of the subject to a downward water movement (infiltration) of fens had become dominated by Sphagnum spp., first byS. approximately 1-2 m year-1. The water loss is far greater subnitens, then S. fallax and now S. palustre, probably due than the precipitation surplus, and compensated to increased isolation from base-rich ditch water and by lateral inflow of surface water from a ditch at high atmospheric N-deposition (Kooijman et al. 2016). the northeastern side of the fens. This water flows However, the zone with more eutrophic mosses underneath and through the floating root mat. Towards adjacent to the ditch had become much smaller through the southwestern end, the fens are hydrologically time. Also, aboveground biomass in the rich-fen zone isolated by peat baulks, and mineral-rich ditch water showed a fourfold decrease (Table 1), probably due to becomes increasingly mixed with rain water. As a result, improved surface water quality. The fens were already the fens generally show a vegetation gradient from P-limited in the past (Kooijman 1993), but P-limitation brownmoss communities with S. scorpioides closer to has become stronger (Cusell et al. 2014; Kooijman et al. the ditch to Sphagnum-dominated vegetation in the 2016). As a result, characteristic brownmosses and other more isolated parts. rich-fen species not only persisted, but also expanded Although water quality was better than in many in the area up to 100 m distance from the ditch. This is other parts of the Netherlands, various measures have due to occasional flooding of the fen with ditch water been applied to improve water quality since the 1970s, during wet periods, when a lot of water is pumped in including major changes in the regional water flows. the wetland area from the agricultural surroundings. The inlet has been shifted from relatively eutrophic Inundation from time to time leads to increased rivers to the much cleaner lake Vollenhoven. Also, water buffer capacity, and prevents the establishment of and purification plants have been constructed, and P-input acidification bySphagnum spp. from urban areas has strongly decreased (Cusell 2014). Local measures to improve water quality and increase De Veldweg: reference site with supply of base-rich ditch water were taken in 1992. The occasional flooding from small ditches flow of water to the ditch in the Stobbenribben fen This fen complex is located in the Wieden, and complex was redirected, leading to a longer pathway, consists of approximately one meter of peat on top

38 of Pleistocene cover sand. The fens partly consist of rich fen, dominated by the brownmosses S. scorpioides and especially S. cossoni. Base-rich water is supplied to interior parts of the fen by small ditches. This particular fen was used as control site in the water level manipulation experiment, which was conducted from summer 2008 to summer 2014 (Cusell et al. 2015; Mettrop et al. 2015a). In this particular fen area, water levels were not experimentally manipulated, in contrast to the real test area, because it served as control site. Different vegetation types were included in the study, such as mesotrophic base-rich fens with Scorpidium spp. (H4170A); slightly eutrophic base-rich fens with the brownmoss Calliergonella cuspidata (H4170A) and Sphagnum-dominated transition fens (H4170B). In summer 2009, which was supposed to be a dry period in which the effect of lower water levels should be tested in the experimental area, rainfall suddenly became rather large. This led to increased water levels in the control area, and the lower lying rich fens even flooded with water from the small ditches. During the Figure 5. Experimental flooding in winter in de Kiersche measurement period, buffer capacity in the two rich fen Wiede.Figure 5. Experimental flooding in winter in de Kiersche Wiede. plots did not remain the same, but instead increased due to the unexpected flooding. In theSphagnum - periods,Different large vegetation parts of the area types were actuallywere includedflooded in the study, such as mesotrophic base- dominated fens, however, buffer capacity remained the (Figurerich fens 5). with Scorpidium spp. (H4170A); slightly eutrophic base-rich fens with same. Throughout the Wieden, small ditches have been theHowever, brownmoss flooding inCalliergonella winter did not lead cuspidatato increased (H4170A) and Sphagnum-dominated created to supply base-rich surface water to the interior buffertransition capacity fens at all, as(H4170B). was expected fromIn thismesocosm particular fen area, water levels were of the fens, and they almost always have a positive experimentsexperimentally (Cusell etmanipulated al. 2013, Mettrop with et al. 2015b).large In pumping stations. effect on the rich fen habitats. winter, the water levels inside the fen was already very In winter, water levels were 17 cm higher than normal for a period of ten days in high, and evapotranspiration very low, so the water winter. During these periods, large parts of the area were actually flooded De Kiersche Wiede: experimental site could flow over the fen, but did not infiltrate. From (Figure 5). with manipulation of water level the control experiment, we learned that (unexpected) This fen complex is also located in the Wieden, and summerHowever, flooding flooding can lead in to increasewinter indid buffer not capacity, lead to increased buffer capacity at all, as also consists of approximately one meter of peat on sowas the floodingexpected experiment from mesocosmwas also applied experiments in summer (Cusell et al. 2013, Mettrop et al. top of Pleistocene cover sand. The fens partly consist 2013 and 2014 (Mettrop 2015a). This indeed led to 2015b). In winter,et al.the water levels inside the fen was already very high, and of rich fen, dominated by the brownmoss Hamatocaulis increase in buffer capacity in especially the two rich evapotranspiration very low, so the water could flow over the fen, but did not vernicosus. This particular fen was used as experimental fens, which further supports that flooding with base- infiltrate. From the control experiment, we learned that (unexpected) summer site in the water level manipulation experiment, which rich, clean surface water can be a serious management flooding can lead to increase in buffer capacity, so the flooding experiment was was conducted from summer 2008 to summer 2014 option, if the water contains enough calcium and also applied in summer 2013 and 2014 (Mettrop et al. 2015a). This indeed led to (Cusell et al. 2015; Mettrop et al. 2015a). bicarbonate. This is further studied in ongoing OBN- increase in buffer capacity in especially the two rich fens, which further supports Different vegetation types were included in the study, experiments in other wetland areas in the Netherlands, that flooding with base-rich, clean surface water can be a serious management such as mesotrophic base-rich fens with Scorpidium such as Wieden, Naardermeer, Nieuwkoop and Rottige option, if the water contains enough calcium and bicarbonate. This is further spp. (H4170A); slightly eutrophic base-rich fens with Meente. the brownmoss Calliergonella cuspidata (H4170A) and studiedIn de Kiersche in ongoing Wiede, we OBN also -triedexperiments to simulate in other wetland areas in the Netherlands, Sphagnum-dominated transition fens (H4170B). In this summersuch as drought. Wieden, In summer, Naardermeer, water levels inNieuwkoop the ditches and Rottige Meente. particular fen area, water levels were experimentally of the entire area were 8 cm lower than normal for a manipulated with large pumping stations. period of ten days. However, during the measurement In winter, water levels were 17 cm higher than periods, rainfall was usually so high that water levels 67 normal for a period of ten days in winter. During these inside the fens were not lowered at all. This made it

39 impossible to test the effect of summer drought in the Further information/further reading field. However, drought manipulation measurements in Cusell C, Lamers LPM, van Wirdum G & Kooijman AM (2013). the laboratory learned that drought can have very rapid Impacts of water level fluctuation on mesotrophic rich fens: effects through the immediate access of oxygen when acidification versus eutrophication. Journal of Applied Ecology, 50:, 998-1009. water tables drop (Mettrop et al. 2014). Mineralization of C and N strongly increase under oxygen-rich conditions, Cusell C (2014). Preventing acidification and eutrophication in rich fens: water level management as a solution? PhD-thesis especially in base-rich fens. With more fluctuating University of Amsterdam. water levels, lower water tables in summer may thus reduce the intake of polluted water in the wetland area Cusell C, Kooijman AM & Lamers, LPM (2014). Nitrogen or phosphorus limitation in rich fens? Edaphic differences explain from elsewhere, but for rich fens, lower water levels contrasting results in vegetation development after fertilization. are not at all a good idea. Moreover, in the Weerribben- Plant and Soil 384: 153-168.

Wieden area, the water which would be supplied in Cusell C, Mettrop IS, van Loon EE, Lamers LPM, Vorenhout dry periods is relatively clean. The contribution of lake M & Kooijman AM (2015). Impacts of short-term droughts Vollenhove to the annual P-input is not more than and inundations in species-rich fens during summer and winter: Large-scale field manipulation experiments. Ecological 1% (Figure 3. This is much less than the amount of P Engineering 77: 127–138. coming in from the agricultural polders, and not at all Kooijman AM (1993). Causes of the replacement of Scorpidium worth the risk of increased mineralization of C and N scorpioides by Calliergonella cuspidata in eutrophicated rich from large peatland areas. This also means that the fens 1. Field studies. Lindbergia 18: 78-84. discussion about more fluctuating water levels is more Kooijman AM & Paulissen MPCP (2006). Acidification rates in complicated than generally thought. wetlands with different types of nutrient limitation. Applied Vegetation Science: 205-212. Meppelerdieplanden: flooding with Kooijman AM, Cusell C, Mettrop IS & Lamers LPM (2016). Fe-rich water Recovery of rich-fen bryophytes in floating rich fens over the The Meppelerdieplanden are clayey peatlands located past 25 years by improvement of nutrient status and inundation with base-rich surface water. Applied Vegetation Science 19: along the Meppelerdiep river, with rich-fens dominated 53-65. by the brownmoss Hamatocaulis vernicosus. The fens are regularly flooded with Fe-rich water from a nearby Mettrop, IS, Cusell C, Kooijman AM, Lamers LPM (2014). Nutrient and carbon dynamics in peat from rich fens and ditch with local seepage water, which keeps the buffer Sphagnum-fens during different gradations of drought, Soil capacity and pH high enough to prevent acidification. Biology & Biochemistry 68: 317-328.

However, the fens are also relatively rich in P (Cusell Mettrop, IS 2015. Water level fluctuations in rich fens; an et al. 2014, Mettrop 2015). There is still (or again) a lot assessment of ecological benefits drawbacks. PhD-thesis of discussion about the relation between Fe and P, and University of Amsterdam. whether high Fe levels are reducing P-availability to Mettrop IS, Cusell, C, Kooijman, AM and Lamers LPM (2015a). the vegetation, or not. Very likely, high Fe levels can Short-Term Summer Inundation as a Measure to Counteract Acidification in Rich Fens. PLoS ONE 10: e0144006. catch P from the water, which leads to low phosphate concentrations and relatively clear water. However, Mettrop IS, Rutte MD, Kooijman AM, & Lamers LPM (2015b). in peatlands, a lot of P is probably weakly sorbed The ecological effects of water level fluctuation and phosphate enrichment in mesotrophic peatlands are strongly mediated by to complexes of Fe and organic matter, rather than soil chemistry. Ecological Engineering 85: 226-236. precipitated as iron phosphates, and plant roots Mettrop, I.S., Neijmeijer, T., Cusell, C., Lamers, L.P.M., probably take this up rather easily (Mettrop et al. 2018). Hedenäs, L and Kooijman, A.M. 2018. Calcium and iron However, high P-levels in the plant of Fe-rich soils as important drivers of brown moss composition through may not necessarily lead to high biomass production, differential effects on P-availability. Journal of Bryology, in press. but may also act as protection against Fe-toxicity. This Van Diggelen R, Molenaar WJ & Kooijman AM (1996). needs to be further studied. Vegetation succession in a floating mire in relation to management and hydrology. Journal of Vegetation Science 7, 809-20.

Van Wirdum G, den Held AJ & Schmitz M (1992). Terrestrializing fen vegetation in former turbaries in the Netherlands. In: Fens and Bogs in the Netherlands: Vegetation, History, Nutrient Dynamics and Conservation (ed. J.T.A. Verhoeven), pp. 323-60. Kluwer Academic Publishers, Dordrecht.

40 Restoration of bog relics in the Restoration of bog relics in the Netherlands; RestorationNetherlands; at what Restoration costs? at what costs?

Ab Grootjans, André Jansen & Gert Jan van Duinen. Ab Grootjans, André Jansen & Gert Jan van Duinen th th SundaySunday 24 26 th- Monday- Monday 27 27 thAugust August

important driving force for reclamation of the marshes, fens and bogs during the Middle Ages was the church, especially the monasteries.

Since the shift of the economic centre from Flanders to Holland around 1600, peat was extracted by dredging on a large scale for fuel, both for industry

and households. This large-scale peat dredging started in the western part of the Netherlands and in the peatlands surrounding the former Zuiderzee. When

these peat stocks had been depleted and an increasing area of land disappeared under water, the excavation of the vast bogs in the northern part of the Netherlands

started.

Around 1850 the peat had already been excavated in large parts of the northern Netherlands, especially

in Friesland, Groningen and the northern part of th Drenthe (Figure 1). During the 19 century and the beginning of the 20th century the excavation continued

at an enormous scale, not only in the northern part of the country, but also in the south-east (Figure TheThe former former distribution of BogsBogs in in the the Netherlands2), until coal became the most important fuel for Netherlands industry and households. In 1939, peat supplied only ince the end of the last glacial period huge areas of 3% of the national energy demand (Gerding 1995). Since fensthe and end bogs of have the developed last glacial in the Netherlands.period huge areasAfterwards, of fens almost and all of bogs the remaining have bogs were developedThe peat in reachedthe Netherlands. its largest distribution The peat around reached minedits largest for agricultural distribution and horticultural around use, i.e. for soil 100S AD. Then, almost two-thirds of the Netherlands was covered with peat. By 100 AD. Then, almost two-thirds of the Netherlands improvement and horticultural crop production, and sea level rise, peatlands disappeared due to drowning and erosion, or became was covered with peat. By sea level rise, peatlands to produce active coal for water purification purposes covered by marine or riverine clays (Figure 1, 800 AD). These deterioration disappeared due to drowning and erosion, or became (Joosten 2009). processes continued for centuries. Therefore, at the end of the Middle Ages covered by marine or riverine clays (Figure 1, 800 AD). In addition to the extraction of peat in a large-scale (Figure 1, 1500 AD), the acreage had shrunk further. Since the Middle Ages also humanThese deterioration activities, processes mainly continued salt extraction for centuries. (“moernering” and industrial in manner,Dutch) peat and was alsoturf cut in small peat extractionTherefore, at in the the end southwestern of the Middle Ages bogs, (Figure resulted 1, 1500 in apits further by residents decrease of villages of theand hamlets.vast This happened peatlandAD), the acreage area. had Starting shrunk further.from Sincethe eighththe Middle century, not the only extensive at the edges ofmires the very of largethe bog complexes, Ages also human activities, mainly salt extraction but also in the smaller bogs along the German-Dutch (‘moernering’ in Dutch) and turf extraction in the70 border, as we will see in the Wierdense Veld. In the southwestern bogs, resulted in a further decrease of the eastern part of the country several remnants of these vast peatland area. Starting from the eighth century, the smaller peat bogs have not been reclaimed. They are extensive mires of the western and northern part of the almost all designated as nature reserves and Natura Netherlands (Holland and Friesland) were colonised 2000 sites. The Aamsveen is a characteristic example of and used as meadow, pasture and arable land. An this. In the south, almost all of these smaller bogs have

41 western and northern part of the Netherlands (Holland and Friesland) were colonised and used as meadow, pasture and arable land. An important driving force for reclamation of the marshes, fens and bogs during the Middle Ages was the church, especially the monasteries.

Fens &bogs

Fens/ bogs

Fens &bogs

Figure 1: Former distribution of fens and bogs in the Nowadays, only little remains of the former vast Netherlands.Figure 1: TheFormer map ofdistribution AD 2000 depicts of thefens distribution and bogs of in the Netherlands. The map of AD 2000 depicts the area of marshes, fens and bogs (Figure 1). Almost no peatdistribution soils. Nature of reserves peat soils.with fens Nature and bogs reserves are being with (a fens and bogs are being (a minor) part of it. After: minor)Bazelmans part of it.et After: al. (2011). Bazelmans et al. (2011). traces of the former mires can be found anymore in the Netherlands (Borger 1992). Not one bog or fen has been been completely excavated and subsequently cultivated left untouched. In 2005 ca. 290,000 ha. of peat soils into agricultural land. The large bogs in the north and had remained. About 40,000 ha of degenerated mire south were also mined to agricultural land, where a very relics and lakes remained, of which 30,000 ha is ‘fen’ distinctive allotment was created (Figure 3). The former (Vermeer & Joosten 1992) and about 12,000 ‘bog’ largely bogs in the north have long been used for large-scale 71covered with dominance of Molinia caerulea and Betula cultivation of grain and potatoes, now livestock farming pubescens. is gaining ground, which in the south have developed The area of actively growing bog vegetation into the stronghold of intensive livestock farming. dominated by Sphagnum (hummock) species did not

42 Figure 2 Distribution of raised bogs in the Netherland (c. 1500 AD). The largest bog Before their reclamation the bogs themselves were also agriculturally used. complexes could be found Grazing by sheep was wide-spread, as is indicated by the large amount of pollen in the northern provinces. Calluna vulgaris in peat bodies since the Middle Ages, as was found in the And in the south-eastern Aamsveen. Later, small farms (“boetes”)part. In the north were the built on the bog, where shepherds lived during the summer. Next hugeto the transboundary “boetes” they reclaimed grasslands in long and narrow lots, so-called “bovenveengraslanden”bog complex of the which were superficially drained. Subsequently, the vegetationBourtanger was Moor comple (B) wastely removed, after which the lots were excavated up to 90situated, cm deep.of which Then the manure was applied. These grasslands were grazed or mown.present A naturefew of reserve these grasslands have remained in the Bargerveen nature reserve,Bargerveen where isthey one ofhave the developed into species-rich grasslands with also a rich fauna.last remnants, whereas in the south the Peel In the reclaimed bogs, the techniquebogs(P) were of situated.buckwheat fire cultivation was wide- th spread, especially from the 18The century Wierdense up Veld until (W) the 1930s. For this cultivation, part of the bog was superficiallyand drained, the Aamsveen after (A)which are the surface was set on fire. After the fire was (finally) extinguished,examples of smaller the cropbogs was sown on ash-rich peat surface. In this way, mainly buckwheatin the eastern was part grown. of the Occasionally rape, potatoes, rye, and oats were grown (Vancountry. der After: Linden Casparie 2018). & This burning of the peat caused an enormous air pollution,Streefkerk that (1992). even reached Southern France and Hungary (Joosten 2009; Figure 4). At the end of the 19th century this burning practise was forbidden by law.

Figure 3: Characteristic allotment after peat excavation in the Figure 4: Map of the distribution of moor haze stemming northern part of the Netherlands. 1 = canal; 2 = side channel Figurefrom 4: buckwheat Map of the fi distribution re cultivation of inmoor north-western haze stemming Germany from buckwheat fi re cultivation in north- western Germany in the years 1848, 1857 and 1863. (After Prestel 1903, in: Joosten 2009). (‘wijk’) and 3 = farm. After: Visscher (1975). in the years 1848, 1857 and 1863. (After Prestel 1903, in: Joosten 2009). exceed a few hectares before 1994 (Joosten 1994), but has (‘boetes’) were built on the bog, where shepherds now increased to 7.6 ha (Jansen et al. 2013). The area of lived during the summer. Next to the ‘boetes’ they well-developed rich fin covers also only a few hectares reclaimed grasslands in long and narrow lots,74 so-

(Cusell 2014). called ‘bovenveengraslanden’ which were superficially Before their reclamation the bogs themselves were drained. Subsequently, the vegetation was completely also agriculturally used. Grazing by sheep was wide- removed, after which the lots were excavated up to 90 spread, as is indicated by the large amount of pollen cm deep. Then manure was applied. These grasslands Calluna vulgaris in peat bodies since the Middle Ages, were grazed or mown. A few of these grasslands have as was found in the Aamsveen. Later, small farms remained in the Bargerveen nature reserve, where they

43 have developed into species-rich grasslands with also a subsoil. Although, progressive desiccation had been rich fauna. delayed with these measures, it almost did not lead to In the reclaimed bogs, the technique of buckwheat the development of actively growing bog vegetation fire cultivation was wide-spread, especially from the with dominance of hummock species. However, swards 18th century up until the 1930s. For this cultivation, of hollows and lawns did develop over large sections in part of the bog was superficially drained, after which compartments separated by dams. The development the surface was set on fire. After the fire was (finally) towards hummock species stagnated; Sphagnum fallax extinguished, the crop was sown on ash-rich peat formed monotonous and species-poor vegetation, with surface. In this way, mainly buckwheat was grown. still abundant Molinia caerulea. At the same time these Occasionally rape, potatoes, rye, and oats were grown bog remnants became very important for many water (Van der Linden 2018). This burning of the peat caused birds and waders, and for birds of semi-open landscapes an enormous air pollution, that even reached Southern and heathlands. After millions of euros were spent on France and Hungary (Joosten 2009; Figure 4). At the end restoration measures, in the first decade of the 21st of the 19th century this burning practise was forbidden century research was done in the framework of the by law. OBN, the Dutch nature restoration programme, on the determining processes in bogs, on landscape as well Restoring remnants of oligotrophic as on site level, to learn the factors that the recovery bogs of bogs could favour. The research showed that the In the Netherlands the Korenburgerveen was development of the lawns stagnated due to excessive the first bog that was purchased as a nature nitrogen deposition and lack of CO2-and methane rich reserve. That happened in 1918 by the ‘Vereniging groundwater. The Irish-Dutch exchange program of Natuurmonumenten’, a Dutch NGO, which currently the State Foresty Commission (Staatsbosbeheer) also owns and manages almost 110,000 hectares of nature generated a lot of knowledge, especially about the reserves. In 1938, Natuurmonumenten purchased part hydrological functioning of pristine Atlantic bogs. This of the Fochteloërveen. It was only after the Second new knowledge resulted in an improved hydrological World War that the remnants of the other Dutch bogs designs of many reserves, including the construction of were purchased or designated as nature reserves, most buffer zones. of them in the 1950s and 1960s, the Wierdense Veld last in 1975. In almost all of these remnants the peat Bargerveen had been almost entirely excavated; in no more than four reserves, including the Bargerveen, small areas ithin the Netherlands, the Bargerveen Reserve remained unexcavated. The management of the bog Wis one of the largest bog remnants and also reserves consisted primarily of the removal of young the one where restoration is the most promising. The trees. Large parts, however, were not managed and reserve is managed by Dutch State Forestry service. became forested. In some areas ditches were closed The Bargerveen Reserve is a very small remnant to rewet the peat.Bargerveen In the 1980s, the first dams were Figure 5. Left: The location of the former Boertanger Moor bog constructed to retain water and to slow down its Within the Netherlands, the Bargerveencomplex (yellow) Reserve and the presentis one Bargerveen of the (green) largest on the bog superficial drainage, after which large-scale dams were border of Netherlands and Germany. Centre: The former extent laid out in theremnants 1990s or plastic and foils also were the placed one in thewhere ofrestoration the Boertanger is Moor the with most the Bargerveen promising. (red). Right:The thereserve is managed by Dutch State Forestrypresent service. Bargerveen Reserve with the Meerstalblok (red).

Figure 5. Left: The location of the former Boertanger Moor bog complex (yellow) and the present 44 Bargerveen (green) on the border of Netherlands and Germany. Centre: The former extent of the Boertanger Moor with the Bargerveen (red). Right: the present Bargerveen Reserve with the Meerstalblok (red).

The Bargerveen Reserve is a very small remnant (23 km2) of the former Boertanger Moor (3,000 km2), which once formed the border between Germany and the Netherlands (Fig. 5). The development of the bog complex was studied in detail by Casparie (1972).

Figure 6A shows an impression of the southern part of the former Boertanger Moor, where several raised bog domes had merged to form one large bog complex. A small river (Runde) was exporting excess surface water towards the river Hunze. Between the domes a large dystrophic lake (Zwarte Meer = ‘Black Lake’) was present. Because the bogs were dome-shaped, precipitation water flew from the top of the bog through the acrotelm (=living top layer of Sphagnum) to lower areas. Infiltration into the mineral underground was almost totally prevented by the presence of boulder clay layers in the subsoil and the almost impervious lake sediments and other organic layers with a high resistance to water flow. The bog complex was not only fed by precipitation water: locally groundwater from surrounding sandy ridges could enter the complex through so- called ‘hydrological windows’ where the boulder clay layers had eroded away (Casparie & Streefkerk 1992).

Only a very small part of the Dutch side of the Boertanger Moor has not been subject to peat extraction. This area, part of the Meerstalblok, was the only place where typical bog vegetation and a small bog lake (in Dutch ‘meerstal’) had remained before the area was declared a nature reserve in 1968 (Figure 6B). Although drainage and burning of the top layer for growing Buckwheat was practiced here, typical plant and invertebrate species persist in the area (Van

76

(23 km2) of the former Boertanger Moor (3,000 km2), A which once formed the border between Germany and the Netherlands (Fig. 5). The development of the bog complex was studied in detail by Casparie (1972). Figure 6A shows an impression of the southern part of the former Boertanger Moor, where several raised bog domes had merged to form one large bog complex. A small river (Runde) was exporting excess surface water towards the river Hunze. Between the domes a large dystrophic lake (Zwarte Meer = ‘Black Lake’) B was present. Because the bogs were dome-shaped, precipitation water flew from the top of the bog through the acrotelm (=living top layer of Sphagnum) to lower areas. Infiltration into the mineral underground was Meerstal almost totally prevented by the presence of boulder clay layers in the subsoil and the almost impervious lake sediments and other organic layers with a high resistance to water flow. The bog complex was not only fed by precipitation water: locally groundwater from C surrounding sandy ridges could enter the complex through so-called ‘hydrological windows’ where the boulder clay layers had eroded away (Casparie & Bog vegetation Streefkerk 1992). Molinia dominated heath Open water Only a very small part of the Dutch side of the Agricultural fields Boertanger Moor has not been subject to peat Peat layers Mineral soils extraction. This area, part of the Meerstalblok, was the only place where typical bog vegetation and a small Figure 6. A: Impression of the southern part of the former Boertanger Moor consisting of various largeFigure bogs 6.that A: hadImpression merged to of one the large southern bog complex. part of The the yellow square represents the present bog lake (in Dutch ‘meerstal’) had remained before the extensionformer of Boertanger the bog relict Moor Bargerve consistingen. B: Impression of various of the large cutover bogs and largely reclaimed peatland with in the lower corner the small bog remnant Meerstalblok before the start of restoration. C: area was declared a nature reserve in 1968 (Figure 6B). Impressionthat had of merged the present to one bog largereserve bog Bargerveen complex. showing The yellow restored bog vegetation around the Although drainage and burning of the top layer for coresquare reserve represents (Meerstalblok the in presentthe foreground) extension and large of the flooded bog bufferrelict zones on former agricultural land (Figure drawn by Ab Grootjans). growing Buckwheat was practiced here, typical plant Bargerveen. B: Impression of the cutover and largely reclaimed peatland with in the lower corner the small and invertebrate species persist in the area (Van Duinen bog remnant Meerstalblok before the start of restoration. 78 2013). The bog remnant is situated 4 meter above the C: Impression of the present bog reserve Bargerveen surrounding area, where large scale peat extraction in showing restored bog vegetation around the core reserve the 19th and 20th century has left some small strongly (Meerstalblok in the foreground) and large flooded buffer zones on former agricultural land (Figure drawn by Ab desiccated bog remnants amidst areas with shallow Grootjans). remaining peat layers that were transformed into pastures and arable fields. Currently, several hectares are now classified as active raised bog (H7110A). In case of open water Sphagnum Management and management starts floating when sufficient methane and carbon challenges dioxide is produced by the underlying substrate Since 1971 a network of smaller peat dams and more (Lamers et al. 2002), which may consist of peat or litter recently large dikes has been built in order to raise the formed by grasses or trees. Sphagnum mats keep floating water levels in the reserve (Figure 6C). Because of the for a long time in case little decomposed Sphagnum very large height differences between the areas with peat is present (Tomassen et al. 2004) or when slightly thick remaining peat layers and the cut-over areas, calcareous groundwater enters the peat from below and nearly all low lying areas became almost permanently stimulates decomposition (Smolders et al. 2003). In the flooded. In the central part of the reserve the regrowth low-lying areas with shallow peat layers and permanent of Sphagnum and Eriophorum species was good. flooding,Sphagnum growth is limited by lack of CO2.

45 The grass species Molinia caerulea expanded massively after rewetting. Experimental research in the field and in the laboratory revealed that the vigorous growth of Molinia was caused by the high nitrogen deposition from the air (Limpens et al. 2003, Tomassen et al. 2003). In the Bargerveen area the atmospheric N-deposition is 20-40 kg N ha-1yr-1. Critical loads for bogs are 5–10 kg N ha-1yr-1, above which the Sphagnum species are unable to absorb the nitrogen (Lamers et al. 2000) and nitrogen becomes available for vascular plants such as Molinia caerulea and Betula pubescens. Above 18 kg of N ha-1yr-1 the vascular plants start to shade the Sphagnum plants to the extent that their growth is reduced considerably (Limpens et al. 2003). In the Bargerveen Reserve the rapid growth of grasses and shrubs has been suppressed by introducing sheep and cattle grazing in the area. This has stimulated the growth of Sphagnum to the extent that in the best rewetted areas in the Meerstalblok Sphagnum has become dominant and grazing is no longer necessary. The final touch to rewetting of the central parts of the reserve was given by the establishment of large buffer zones on agricultural land around the reserve, where very high water levels were installed. The farmland was bought, mostly by the government.

Figure 7. The central part of the bog reserve Bargerveen Figure 7. The central part of the bog reserve Bargerveen showing the restored bog vegetation showing the restored bog vegetation growing at almost the Figure 7 shows that as a result the high peat dikes growing at almost the same height as the peat dikes that have been built to raise the water level same height as the peat dikes that have been built to raise in the central parts of the bog remnant are almost withthe water several level withmeters. several (Photo: meters. (Photo:André André Jansen, Jansen, December 2014). December 2014). overgrown by bog vegetation. Figure 7 shows that as a result the high peatThe dikes total budget in the scheduled central for partsthe period of 2013- the bog remnantOnly diffusion are of almostCO2 from the overgrown atmosphere is by not bog vegetation.2026 for further measures to restore the Bargerveen sufficient to supportSphagnum growth under water bog amounts to about 34 million Euro, including 30 The(Smolders total et al. budget 2001, Patberg scheduled et al. 2012). for the periodmillion 2013 for -restoration2026 for measures further and hydrologicalmeasures to restoreThe grass the species Bargerveen Molinia caerulea bogexpanded amounts tobuffer about zones 34 and millic. 4 millionon Euro,for continued including management 30 millionmassively forafter restoration rewetting. Experimental measures research and in thehydrological of the reserve. buffer These zonesfunds come and from c. the 4 Provincialmillion for continuedfield and in the management laboratory revealed ofthat the the vigorousreserve. Thesegovernment, funds the national come government, from the and Provincial the growth of Molinia was caused by the high nitrogen European Union. government, the national government, and the European Union. deposition from the air (Limpens et al. 2003, Tomassen The hydrological buffer zones surrounding the bog Theet al. 2003).hydrologica In the Bargerveenl buffer area thezones atmospheric surrounding relic the are 500-800mbog relic wide are on the500 Dutch-800m side (Figure wide 8). on 1 1 theN-deposition Dutch is side20-40 kg(Figure N ha- yr- 8).. Critical The loads German for authoritiesThe German authorities are planning are planning a a hydrologicalhydrological bogs are 5–10 kg N ha-1yr-1, above which the Sphagnum buffer zone of 300m wide at the eastern border of the buffer zone of 300m wide at the eastern border of the reserve. species are unable to absorb the nitrogen (Lamers et reserve. al. 2000) and nitrogen becomes available for vascular The hydrological buffer zones are largely established plants such as Molinia caerulea and Betula pubescens. and planned on cut-over peatlands with shallow layers Above 18 kg of N ha-1yr-1 the vascular plants start to of (black) peat that have been intensively fertilized and shade the Sphagnum plants to the extent that their 79 are extremely rich in nutrients, in particular phosphate. growth is reduced considerably (Limpens et al. 2003). After rewetting these former agricultural lands are In the Bargerveen Reserve the rapid growth of grasses highly productive and species like Reed (Phragmites and shrubs has been suppressed by introducing sheep australis) and Common Cattail (Typha latifolia), or and cattle grazing in the area. This has stimulated Willow (Salix) dominate the vegetation within a few the growth of Sphagnum to the extent that in the best years. Harvesting these crops is possible with machines rewetted areas in the Meerstalblok Sphagnum has that are adapted to driving in marshes and shallow become dominant and grazing is no longer necessary. water. Currently, a pilot project is starting to explore the The final touch to rewetting of the central parts of the productive use of rewetted agricultural fields on peat reserve was given by the establishment of large buffer (Paludiculture.com; Wichtmann et al. 2016). zones on agricultural land around the reserve, where Figure 8. Left: deeply drained agricultural fields; right: very high water levels were installed. The farmland was peripheral hydrological buffer zone to keep the water levels bought, mostly by the government. in the Bargerveen bog reserve high. (Photo: André Jansen).

Figure 8. Left:46 deeply drained agricultural fields; right: peripheral hydrological buffer zone to keep the water levels in the Bargerveen bog reserve high. (Photo: André Jansen).

The hydrological buffer zones are largely established and planned on cut-over peatlands with shallow layers of (black) peat that have been intensively fertilized and are extremely rich in nutrients, in particular phosphate. After rewetting these former agricultural lands are highly productive and species like Reed (Phragmites australis) and Common Cattail (Typha latifolia), or Willow (Salix) dominate the vegetation within a few years. Harvesting these crops is possible with machines that are adapted to driving in marshes and shallow water. Currently, a pilot project is starting to explore the productive use of rewetted agricultural fields on peat (Paludiculture.com; Wichtmann et al. 2016).

Day programme (Sunday 26 augustus: 9.30-15.30).

The excursion starts at the office of State Forestry Services (Staatsbosbeheer; Figure 9).

Figure 9. Excursion route in the Bog relic Bargerveen.

80

We will walk to the remaining small bog lake ‘Meerstal’ and pass the buffer area We will walk to the remaining small bog lake ‘Meerstal’ and pass the buffer area and water retention basins. We will also see the differences between the older and water retention basins. We will also see the differences between the older peat dams and the more recent large dikes constructed to raise the water table peat dams and the more recent large dikes constructed to raise the water table in the bog remnant (Figure 10). The vegetation succession following rewetting of in the bog remnant (Figure 10). The vegetation succession following rewetting of poorly humified peat and the effects of sheep grazing can be observed during our poorly humified peat and the effects of sheep grazing can be observed during our walk in the Meerstalblok. A last stop will be on one of the remaining pieces of walk in the Meerstalblok. A last stop will be on one of the remaining pieces of ‘bovenveengraslanden’, which consists of grasslands that have been used for ‘bovenveengraslanden’, which consists of grasslands that have been used for agriculture in the past (Figure 11). These grasslands can be rich in plant and agriculture in the past (Figure 11). These grasslands can be rich in plant and invertebrate species, but their preservation is difficult with the raised bog being restoredinvertebrate around species, them. but their preservation is difficult with the raised bog being restored around them. Excursion programme Sunday 26th august 9.30-15.30 The excursion starts at the office of State Forestry Services (Staatsbosbeheer; Figure 9). Figure 8. Left: deeply drained agricultural fields; right: peripheral hydrological buffer zone to keep Wethe will water walk tolevels the remaining in the Bargerveen small bog lake bog‘Meerstal’ reserve high. (Photo: André Jansen). and pass the buffer area and water retention basins. WeThe will hydrologicalalso see the differences buffer between zones the older are peat largely established and planned on cut-over damspeatlands and the more with recent shallow large dikes layers constructed of (black) peat that have been intensively fertilized to raise the water table in the bog remnant (Figure Figure 10. Large dam along the German-Dutch border under and are extremely rich in nutrients,Figure in 10particular. Large dam along thephosphate. German-Dutch border underAfter construction rewetting 10).these The vegetation former succession agricultural following rewetting lands of are construction.Figurehighly 10. Large productive dam along the German and-Dutch borderspecies under construction like Reed poorly humified peat and the effects of sheep grazing (Phragmites australis) and Common Cattail (Typha latifolia), or Willow (Salix) can be observed during our walk in the Meerstalblok. Adominate last stop will bethe on one vegetation of the remaining within pieces of a few years. Harvesting these crops is possible ‘bovenveengraslanden’,with machines which that consists are of grasslandsadapted to driving in marshes and shallow water. thatCurrently, have been used a forpilot agriculture project in the pastis starting to explore the productive use of rewetted (Figureagricultural 11). These grasslands fields oncan bepeat rich in (Paludiculture.com; plant and Wichtmann et al. 2016). invertebrate species, but their preservation is difficult with the raised bog being restored around them.

Day programme (Sunday 26 augustus: 9.30-15.30).

Figure 11.11. TraditionalTraditional grassland grassland on ondecomposed decomposed peat peatsoil, whichsoil, are now very rich in flora and fauna.Figure 11. Traditional grassland on decomposed peat soil, which are now very rich in flora and The excursion starts at the office of whichfauna.State are nowForestry very rich in Servicesflora and fauna. (Staatsbosbeheer;

FigureFigure 9. Excursion 9). route in the Bog relic Bargerveen. 81 81

Figure 9. Excursion route in the Bog relic Bargerveen. 47 80

Day programme (Monday 27 augustus: 9.30-12.00); Excursion programme We leave by bus from the sheep cage. A group crosses Wierdense Veldth . the Huurnerveld (in the northeast) where the Monday 27 august 9.30-12.00 vegetation and fauna of the peat pits will be examined Wierdense Veld and explanations are given on the restoration measures. The other group goes by bus to the Prinsendijk in the south, where attention will be payed to the vegetation development since the installation of a foil screen in the late 1980s. Wierdense Veld

he Wierdense Veld is located Tbetween Wierden and Nijverdal in the province of Overijssel (Figure 1?) and is an approximately 450 ha large remnant of an extensive bog landscape (Figure 2?). The area is owned by the Dutch state and since 1975 in leasehold and management by Landschap Overijssel. It is predominantly vegetated by dry and northeast) and Notterveen (in the west). The reserve wethas heathlands, been designated mostly dominated as Natura 2000 area. by Purple moorgrass (Molinia caerulea), in alternation with an We leave by bus from the sheep cage. A group crosses the Huurnerveld (in the often poorly-developed peat moss northeast) where the vegetation and fauna of the peat pits will be examined and vegetation of hummocks, hollows explanations are given on the restoration measures. The other group goes by bus and pools. The Wierdense Veld is to the Prinsendijk in the south, where attention will be payed to the vegetation a collective name for Westerveen development since the installation of a foil screen in the late 1980s. (in the southeast), Huurnerveld (in the northeast) and Notterveen (in the west). The reserve has been Wierdense Veld designated as Natura 2000 area. Originally, the bog complex of The Wierdense Veld is located between Wierden and Nijverdal in the province of the Wierdense Veld was limited to Overijssel (Figure 1) and is an approximately 450 ha large remnant of an the east by a moraine and in the extensive bog landscape (Figure 2). The area is owned by the Dutch state and west by the Regge, a brook (Figure since 1975 in leasehold and management by Landschap Overijssel. It is 2?). It covered a vast cover sand predominantly vegetated by dry and wet heathlands, mostly dominated by Purple moorgrass (Molinia caerulea), in alternation with an often poorlyplain-developed bordered by peatridges in the moss vegetation of hummocks, hollows and pools. The Wierdensewest and south,Veld which is a retarded collective name for Westerveen (in the southeast), Huurnerveldthe superficial (in flow the of ground- and surface water. In the north the plain had just a narrow outlet, which also 82 hampered drainage. Nowadays,

the bog (remnant) is limited to a small part of this sand plain. The first permeable layer is shallow Topographic map of (1:25.000) of the Wierdense Veld. due to the occurrence of a glacial

Topographic map of (1:25.000) of the Wierdense Veld. 48 Originally, the bog complex of the Wierdense Veld was limited to the east by a moraine and in the west by the Regge, a brook (Figure 2). It covered a vast cover sand plain bordered by ridges in the west and south, which retarded the superficial flow of ground- and surface water. In the north the plain had just a narrow outlet, which also hampered drainage. Nowadays, the bog (remnant) is limited to a small part of this sand plain. The first permeable layer is shallow due to the occurrence of a glacial loam deposit at a depth of approximately 6 meters. In the west this layer is absent. Peat development started presumably in the groundwater seepage areas at the western edge of the moraine Wierden-Hooge Hexel (Figure 3). The groundwater discharged with its highest intensity at the base of this moraine, where the slope of the ground level changes fairly abruptly from fairly steep to flat. Moreover, the phreatic aquifer is thin (15 meters or less), which further promotes upward discharge. Therefore, peat development in the western part of the plain where the first aquifer is thicker, and the poorly permeable loam layer lacks, started later.

83 loam deposit at a depth of approximately 6 meters. In the west this layer is absent. Peat development started presumably in the groundwater seepage areas at the western edge of the moraine Wierden- Hooge Hexel (Figure 3). The groundwater discharged with its highest intensity at the base of this moraine, where the slope of the ground level changes fairly abruptly from fairly steep to flat. Moreover, the phreatic aquifer is thin (15 meters or less), which further promotes upward discharge. Therefore, peat development in the western part of the plain where the first aquifer is thicker, and the poorly permeable loam layer lacks, started later. th FFigureigure 12: 2: 17th 17 century century map ofmap the Wierdenseof the Wierdense Veld. Detail Veld.of the mapDetail of of the map of Ten Have (end 17th century (source:Figure 2: Atlas 17th centuryvan der map Hagen, of the Royal Wierdense Library, Veld. The DetailHague. of Thethe mapposition of Ten of Havethe Wierdense (end 17th Veldcentury is Mire development in this part was a Ten Have (end 17th century (source: Atlas van der Hagen, Royal Library, The indicatedHague.(source: The positionwithAtlas a vanofblack the der Wierdense circle. Hagen, Veld Royal is indicated Library, with a Theblack circle.Hague. The position of the Wierdense Veld is consequence of the poor superficial lateral indicated with a black circle. drainage, promoted by bog development in the east and in the north. Due to increasing wetness of the plain, in which infiltration was the predominant hydrological process, poorly permeable layers, especially podzol B-horizons, could develop, which caused even wetter conditions, which eventually allowed bog development. Around 1850 a vast bog complex does still exists, although it is already strongly influenced by people: the area is intersected by roads and quays, there are here and there watercourses and vast fields with peat pits. In the second half of the 1930s, considerable parts of the Westerveld and Notterveen have been excavated. Later, after the Second World War, large-scale peat excavation, started in the south-east and west of the current Wierdense Veld. After the second half of the 1960s the reclamation activities stopped and the Wierdense Veld reached its current size. FigureFigure 13: 3: Height Height map map of the of Wierdense the Wierdense Veld and surroundings. Veld and surroundings. Desiccation is the main bottleneck Figure 3: Height map of the Wierdense Veld and surroundings. for bog recovery in the Wierdense Veld, especiallyMire the developmentcaerulea, indicating in this a deep part drop was of the a groundwater consequence of the poor superficial lateral enormous amount of infiltration. In an undisturbed table as a consequence of too low hydraulic heads in drainage,Mire development promoted in by this bog part development was a consequence in the east of theand poorin the superficial north. Due lateral to bog circa 40 mm water infiltrates to the mineral subsoil the sand subsoil. The impact of these too low heads increasingdrainage, promotedwetness ofby thebog plain,development in which in theinfiltration east and was in thethe north.predo minantDue to during a year. Hydrological modelling suggested a much is being amplified by the locally intersected poorly increasing wetness of the plain, in which infiltration was the predominant higher infiltration rate, varying from circa 125hydrological mm to permeable process, peat base. poorly The main permeable causes of desiccation layers, especially podzol B-horizons, hydrological process, poorly permeable layers, especially podzol B-horizons, circa 200 mm per year. The desiccation is reflectedcould in develop,are nevertheless which outsidecaused the reserve.even Inwetter the area, theconditions, which eventually allowed the wide-spread monotonous vegetation of Moliniabogcould development. develop,water levels which have beencaused reduced even on a large wetter scale by conditions, which eventually allowed bog development. 84 84 49

excavating the peat and the subsequent construction Fauna of an extensive drainage system with many deep water In the Wierdense Veld various old peat cutting pits are courses. Two drinking water extractions have also lead still present. Research in other Dutch bog remnants to a lowering of the hydraulic head Drinking water showed that (1) such peat cutting pits harbour many extractions and drainage system contribute to a similar rare and characteristic species of aquatic invertebrates, extent to the lowering of the groundwater tables in the and (2) rewetting had been beneficial to only a limited Wierdense Veld. part of the species spectrum of raised bogs. The aquatic Due to the severe desiccation well-developed plant microinvertebrate fauna seems to recover quickly after communities of hummocks are rare, despite the rewetting of drained and cut-over bogs, but this is not internal hydrological measures which have been taken the case for macroinvertebrates, like aquatic beetles in the late 1980s and the first decade of the 21th century. and caddisflies (Van Duinenet al. 2003, 2006). To Nevertheless, since then the vegetation develops in a define a proper restoration strategy for the Wierdense positive way. In the beginning of the 1990s hardly any Veld in the years 2003-2005 the situation (chemistry, Sphagnum magellanicum and S. papillosum were found. vegetation, aquatic invertebrates; Tomassen et al. 2005) aquaticA recent vegetation beetles survey and shows caddisflies a remarkable increase(Van Duinenat that timeet wasal. investigated 2003, 2006).and compared To to defineother a properof these species,restoration especially atstrategy and nearby forthe sitesthe Wierdensebog areas. Veld In total, in 163 the aquatic years macroinvertebrates 2003-2005 the situationwhere rewetting (chemistry, measures have vegetation, been taken. Despite aquatic invertebrates;were found, including Tomassen 39 species characteristic et al. 2005) of bogs. at thatthese recenttime positive was developments,investigated the development and compared Although to relativelyother manybog characteristicareas. In invertebrate total, 163 aquaticof a functioning macroinvertebrates bog system on landscape werelevel requires found, speciesincluding were found 39 in thespecies area, the characteristicnumber of species of bogs.many more Although rewetting measures,relatively especially many outside characteristic and individuals invertebrate per sampling species site was low.were This found might in the reserve. In the southeast and west of the area such indicate small population sizes. A large number the area, the number of species and individuals per sampling site was low. This measures are foreseen in the framework of Natura of characteristic species showed a preference for might2000. indicate small population sizes. A temporarylarge number water bodies of (Figure characteristic 14). The presence speciesof showed a preference for temporary water bodies (Figure 7). The presence of bothFigure 14.temporary Preference of the and 21 characteristic permanent aquatic beetle water decoratusbodies (22); offers 5. Acilius canaliculatusopportunities (12); 6. Hydroporus for many species found in the Wierdense Veld for temporary (hatched umbrosus (29); 7. Bidessus spec. (10); 8. Enochrus affinis speciespart of the bars)to orsurvive permanent in water the bodies area, (black parte.g. of in case(18); 9. Enochrusof extreme ochropterus drought, (11); 10. Berosus like luridus in (4); 2018. Especiallythe bars). The preferencethe small was calculated populations as the frequency of ofcharacteristic 11. Hydroporus speciesobscurus (8); might12. Hydroporus be melanariussensitive to occurrence in permanent and temporary sites, with the sum (4); 13. Hydroporus tristis (18); 14. Agabus melanocornis (5); suddenof these two changes frequencies convertedin water to 100%. table The fluctuations.numbers 15. HydroporusIn order pubescens reduc (15);e the16. Hydroporus risk of gyllenhalii extinction ofof samplingcharacteristic sites in which thespecies, species were founda step are given-wise (23);approach 17. Agabus labiatuswas (12);recommended 18. Rhantus suturellus for(7); 19. the between brackets. 1. Dytiscus lapponicus (4); 2. Hydroporus Hydroporus neglectus (3); 20. Agabus congener (3); 21. Ilybius restorationscalesianus (7); 3. measures Graphoderus zonatus in the (11); Wierdense4. Hygrotus Veld.aenescens (1). (From: Van Duinen et al. 2004).

100%

50% preference

0% 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 species

Figure50 7. Preference of the 21 characteristic aquatic beetle species found in the Wierdense Veld for temporary (hatched part of the bars) or permanent water bodies (black part of the bars). The preference was calculated as the frequency of occurrence in permanent and temporary sites, with the sum of these two frequencies converted to 100%. The numbers of sampling sites in which the species were found are given between brackets. 1. Dytiscus lapponicus (4); 2. Hydroporus scalesianus (7); 3. Graphoderus zonatus (11); 4. Hygrotus decoratus (22); 5. Acilius canaliculatus (12); 6. Hydroporus umbrosus (29); 7. Bidessus spec. (10); 8. Enochrus affinis (18); 9. Enochrus ochropterus (11); 10. Berosus luridus (4); 11. Hydroporus obscurus (8); 12. Hydroporus melanarius (4); 13. Hydroporus tristis (18); 14. Agabus melanocornis (5); 15. Hydroporus pubescens (15); 16. Hydroporus gyllenhalii (23); 17. Agabus labiatus (12); 18. Rhantus suturellus (7); 19. Hydroporus neglectus (3); 20. Agabus congener (3); 21. Ilybius aenescens (1). (From: Van Duinen et al. 2004).

86

RestorationRestoration measures measures

EndEnd of ofthe the 1980’s 1980’s a aplastic plastic sheet sheet was was placed placed in inthe the southern southern part part of ofthe the road road calledcalled Prinsendijk. Prinsendijk. After After the the year year 2000 2000 plans plans were were made made for for a partiala partial removal removal of of Restorationthethe drinking drinking water measureswater extraction extraction Wierden. Wierden. This This would would probably probably result result in insuitable suitable conditionsconditions for for restoration restoration of ofraised raised bog bog in inparts parts of ofthe the Wierdense Wierdense Veld. Veld. In In2011 2011 End of the 1980’s a plastic sheet was placed in the southern part of the road severalseveral restoration restoration measures measures were were carried carried out out (Figure (Figure 5) 5). . called Prinsendijk. After the year 2000 plans were made for a partial removal of the drinking water extraction Wierden. This would probably result in suitable conditions for restoration of raised bog in parts of the Wierdense Veld. In 2011 several restoration measures were carried out (Figure 5).

Figure 15. Bund with plastic sheet inside and weir shortly FigureafterFigure their 5 . construction 5Bund. Bund with inwith 2011plastic plastic at the sheet eastern sheet inside edge inside of and the and weir weir Excursionshortly shortly after after their their programmeconstruction construction in in2011 2011 at atthe the easternWierdenseeastern edge Veld edge reserve.of theof the Wierdense Wierdense Veld Veld reserve. reserve. Monday 27th august 9.30-17.30

both temporary and permanent water bodies offers Aamsveen (13.00-16.30) Figureopportunities 5. Bund for manywith speciesplastic to sheet survive inside in the area,and weir shortly after their construction in 2011 at the easterne.g. in case edge of extreme of the drought,Wierdense like inVeld 2018. reserve. Especially After leaving the bus, we will walk into two groups, in the small populations of characteristic species might be opposite direction. One group starts in the north, the

sensitive to sudden changes in water table fluctuations. other in the south. During our walk the vegetation In order reduce the risk of extinction of characteristic and fauna of the lagg and some compartments will species, a step-wise approach was recommended for the be examined and the hydro-ecological functioning restoration measures in the Wierdense Veld. will be explained. Moreover, some soil profiles will be examined in order to discuss the historical development of the bog. Aamsveen

he Aamsveen (Figure 16) is the north-western TDutch remnant of an originally circa 2000 ha large cross-border bog complex of which the major part (circa 1800 ha) lay on German territory (Fig. 17). The remaining German part consists of the Hündfelder Moor, Alstätter Venn, the Amtsvenn and Graeser Venn. The Aamsveen, with an area of about 140 hectares, is located about three kilometres south-east of Enschede, and consist for about 40 hectares of the habitat type Regenerating bogs (H7120), whereas as the remainder PyrgusPyrgus malvae malvae (Aardbeivlinder). (Aardbeivlinder). probably belonged largely to the lagg, which today is Pyrgus malvae (Aardbeivlinder). 87 87covered by grasslands and forests. The reserve was acquired by the Dutch state at the end of the years 1940. Restoration measures Nowadays he NGO ‘Landscape Overijssel’ is owner Pyrgus malvae (Aardbeivlinder). End of the 1980’s a plastic sheet was placed in the and manager of the reserve and aims at restoration of southern part of the road called Prinsendijk. After the 87 a complete, transboundary bog landscape, including a year 2000 plans were made for a partial removal of well-developed lagg. the drinking water extraction Wierden. This would The entire bog complex is situated in a glacial basin probably result in suitable conditions for restoration of which the subsoil consists of Tertiary clays with a of raised bog in parts of the Wierdense Veld. In 2011 total depth of several tens of meters on which a thin several restoration measures were carried out (Figure package boulder clay is deposited. 15). In the West the bog is bounded by a moraine of

51

Day programme (Monday 27 augustus: 9.30-17.30). Aamsveen (13.00- 16.30)

After leaving the bus, we will walk into two groups, in opposite direction. One groupice-pushed starts Tertiary in clays. the This north, hydrological the base other is in theHere, south. peat formation During started our as awalk consequence the vegetation of covered with a thin layer of cover sands that nowhere is rising groundwater tables. Therefore, the origin of the and fauna of the lagg and some compartments will be examined and the hydro- thicker than about 5 meters. This sand layer wedges out Aamsveen bog is a water rise mire, which constituted ecologicalagainst the Tertiary functioning clays of the moraine. will Inbe this explained. shallow the peatMoreover, base of the latersome Aamsveen soil bog.profiles will be examinedand thin sand layerin order several depressionsto discuss occur. the The historical Startingdevelopment from the Middle-Atlanticum of the bog. (circa 6000 deepest is located just across the border in Germany. BC). the vegetation became gradually more base- and nutrient-poor. The proportion of Black alder (Alnus Figure 16: Topographical map (1:25.000) of the Aamsveen and Aamsveenthe neighbouring part of the German Hündfdelder Moor. glutinosa) then decreased and that of birch (Betula spec.) increased (Van der Linden, 2018). Gradually, birch carrs

The Aamsveen (Figure 1) is the north-western Dutch remnant of an originally circa 2000 ha large cross-border bog complex of which the major part (circa 1800 ha) lay on German territory (Fig. 2). The remaining German part consists of the Hündfelder Moor, Alstätter Venn, the Amtsvenn and Graeser Venn. The Aamsveen, with an area of about 140 hectares, is located about three kilometres south-east of Enschede, and consist for about 40 hectares of the habitat type Regenerating bogs (H7120), whereas as the remainder probably belonged largely

88

Figure 17: The transboundary bog complex ‘Aemsveen’ with a central pool on a 17th century map of N. te Have. Private collection.

52 The number of tree pollen, also that of Black Alder, decreased, pointing to deforestation. Man cultivated the surroundings of the bog and an open landscape did arise. In the peat pollen are present of field weeds, rye and buckwheat. Rye is grown on the fields in the surroundings, buckwheat in the peat itself. Further, the amount of pollen of heather (Calluna vulgaris) had increased, pointing to grazing of the bog itself. Peat was cut in numerous pits (Figure 18). After the Second World War a small part has been cut off in an industrial way. Desiccation is the main problem, indicated by the abundant occurrence of Molina caerulea. Since the beginning of the 20th century in the lagg all characteristic species of alkaline fens have disappeared by desiccation. The same appears for species of fen meadows, but this happened later. From the beginning of the 21th century the last remaining species of weakly buffered conditions are seriously threatened. They still occur in low numbers in humid (Nardus stricta) swards, the plant community that has developed in Figure 18: The Huegenin-map of 1819-1829 of the Aamsveen a degradation series from alkaline fens. In alder carrs clearly shows peat cutting by pits. In pink bog and heathlands, in blue fens and fen meadows. In the middle in the lagg locally species of base-rich conditions still the Glanerbeek has two courses, pointing to a watermill occur (Figure 19). (‘Roodmolen’, Red mill). Along the excavated, southern In the 1990s the first measures have been taken course of the Glanerbeek forests (yellow) are depicted, which to reduce the impact of the desiccation and improve cover sand ridges, intersected by the brook. In white arable fields and in green meadows. the hydrology. Small dams were built, resulting into a chain of small compartments. Within 20 years these developed, in which beside peat mosses (Sphagnum measures have locally resulted in the abundant growth spec.) ground water fed species occurred. In the late- of peat-mosses again. Especially, in the transition Atlanticum (ca. 5000 BC) these carrs were fully acidified zone from the former bog to the former lagg. where and characterised by acidiphilous species as Eriophorum weakly buffered and carbon dioxide-rich groundwater vaginatum. From around 4500 BC hummock forming discharges, the development of a well-developed peat mosses from the section Acutifolia, presumably hummock-hollow pattern shows the possibilities for Sphagnum rubellum came to dominance. The mire further restoration. In this vegetation Reed (Phragmites became completely rainwater fed and eventually a thick australis) accompanies peat mosses as Sphagnum player of old peat moss peat was formed. Due to the magellanicum (which was extincted in the reserve), S. growth of the bog, the water level in the area increased, papillosum, S. palustre, S. fimbriatum, S. fallax, Oxycoccus which caused the emergence of fens in other, formerly palustis and Andromeda polifolia. Further, several ditches higher elevated depressions. Later, these fens also and a deep water course on the Dutch-German border developed into bogs, which eventually fused into one have been filled in. large dome. During the Subatlanticum (ca. 400 BC) it Besides the excavation of the larger part of the bog, became even wetter, which gave rise to the formation the presence of many ditches and trenches and the of young peat moss peat. The hummocks now consisted deep watercourse of the Glanerbeek are responsible mainly of Sphagnum imbricatum (S. austini). Also, the first for desiccation. Moreover, large peat excavations in signs of human influence became apparent by pollen of Germany have an negative influence on the water tables species, such as Plantago lanceolata and cereals (Secales). in the bog and the mineral subsoil. The decline in water Although, active bog development still occurred in tables has resulted in a wide-spread encroachment the Middle Ages, the human impact became larger. and gradual afforestation, which in turn, provides for a

53 Jansen AJM, Van Duinen GA, Tomassen HBM & N.A.C. Smits NAC (2012). Herstelstrategie H7120: Herstellende hoogvenen.

Joosten, H. (2009a) Human impacts: farming, fire, forestry and fuel. In: The Wetlands Handbook (eds. E. Maltby and T. Barker), pp. 689-718. Blackwell Publishing, Oxford.

Lamers LPM, Bobbink R & Roelofs JGM (2000). Natural nitrogen filter fails in raised bogs. Global Change Biology 6, 583-6.

Limpens J, Berendse F & Klees H (2003). N-deposition affects N availability in interstitial water, growth of Sphagnum and invasion of vascular plants in bog vegetation. New Phytologist 157, 339-47.

Patberg W, Baaijens GJ, Elzinga T & Grootjans AP (2013). The importance of carbon dioxide in the development of Sphagnum bogs: a field study. Preslia 85: 389–403. Figure 19: Locally in the lagg species of base-rich conditions still occur, such as Hottonia palustris. Smolders, A. J. P., Tomassen, H. B. M., Pijnappel, H. W., Lamers, L. P. M. & Roelofs, J. G. M. (2001). Substrate-derived further decline in groundwater tables due to increased CO2 is important in the development of Sphagnum spp. New Phytologist 152, 325–332. evaporation. Hydro-ecological research revealed that high water tables in the bog itself are the driving force Smolders AJP, Tomassen HBM, Van Mullekom M, Lamers LPM & Roelofs JGM (2003). Mechanisms involved in the re- for upward discharge of groundwater form the shallow establishment of Sphagnum-dominated vegetation in rewetted aquifer into the lagg. This discharging groundwater bog remnants. Wetland Ecology and Management 11, 403– is very base-rich due to the lime-rich character of the 418.

Tertiary clays. Therefore, measures to raise the water Tomassen HBM, Smolders AJP, Van Herk JM, Lamers LPM & table are required in the bog as well as in the lagg. The Roelofs JGM (2003). Restoration of cut-over bogs by floating raft formation: an experimental feasibility study. Applied prospects for re-activation of abundant - Sphagnum Vegetation Science 6, 141-52. growth in the bog remnant and restoration of a lagg with wide-spread and well-developed basiphilous Tomassen HBM, Smolders AJP, Lamers LPM & Roelofs JGM (2004). Development of floating rafts after the rewetting of vegetation are promising due to the favourable cut-over bogs: the importance of peat quality. Biogeochemistry hydrological conditions, and because measures in and 71, 69-87. along the edges of the reserve will be sufficient. The Tomassen H, Van Duinen GJ Smolders AJP, Brouwer E, van latter will contribute to a larger social and political der Schaaf S, van Wirdum G, Esselink H & J. Roelofs JGM support. (2005). Vooronderzoek Wierdense Veld. Eindrapportage. Onderzoekcentrum B-ware, Stichting Bargerveen, Wageningen Universiteit, NITG-TNO & Radboud Universiteit Nijmegen. Further information/further reading Van Duinen GA, Brock AMT, J.T. Kuper JT, Leuven RSEW, Bazelmans JH, Weerts H & Van der Meulen M (2011). Atlas van Peeters TMJ, Roelofs JGM, Van der Velde G, Verberk WCEP Nederland in het Holoceen, landschap en bewoning vanaf de & Esselink H. (2003). Do restoration measures rehabilitate laatste ijstijd tot nu. Uitgeverij Bert Bakker. fauna diversity in raised bogs? A comparative study on aquatic macroinvertebrates. Wetlands Ecology and Management 11: Borger GJ (1992). Draining-digging-dredging; the creation of a 447-459. new landscape in the peat areas of the low countries. Pages 131-172 in J. T. A. Verhoeven, editor. Fens and bogs in The Van Duinen GA, Zhuge Y, Verberk WCEP, rock AMT, Van Kleef Netherlands: Vegetation, history, nutrient dynamics and HH, Leuven RSEW, Van der Velde G & Esselink H (2006). conservation. Kluwer Academic Publishers, Dordrecht. Effects of rewetting measures in Dutch raised bog remnants on assemblages of aquatic Rotifera and microcrustaceans. Casparie WA (1972). Bog development in south-eastern Drenthe Hydrobiologia 565: 187-200. (the Netherlands). Vegetatio 25: 1-271. Van Duinen GA (2013). Rehabilitation of aquatic invertebrate Casparie WA & Streefkerk JG (1992). Climatological, stratigraphic communities in raised bog landscapes. PhD-thesis Radboud and palaeo-ecological aspects of mire development. In: Fens University Nijmegen. and Bogs in the Netherlands: Vegetation, History, Nutrient Dynamics and Conservation (ed. J.T.H. Verhoeven), pp. 81-129. Wichtmann, W, Schröder C & Joosten H (2016). Paludiculture Kluwer Academic Publishers, Dordrecht. – productive use of wet peatlands. Schweizerbart Science Publishers, Stuttgart. Cusell C (2014). Preventing acidification and eutrophication in rich fens: water level management as a solution? PhD-thesis University of Amsterdam.

54 The Polder Westbroek The Polder Westbroek Jos Verhoeven & Boudewijn Beltman.Jos Verhoeven & Boudewijn Beltman Wednesday Aug 29th 10.00 – 16.30 Wednesday Aug 29th 10.00-16.30

Historical setting and succession in turf ponds

urf pond complexes in The Netherlands have

mostly been excavated below the water table in the 18th and early 19th century. They are now T characterized by very species-rich fen vegetation. As many as 35 different vegetation types have been

described for such habitats in The Netherlands (Van

Wirdum et al. 1992). There are three different succession series, which each characterize a specific water chemistry in the open-water stage, i.e. brackish, fresh mesotrophic and fresh eutrophic. The succession goes

through 7 stages (aquatic, floating raft, helophyte, brown

moss, transitional fen and carr) and has been described by space-for-time substitution of an extensive database of vegetation relevées, see Figure 1, (Van Wirdum 1992, Verhoeven & Bobbink 2001).

Figure 1. Succession seres in terrestrializing ponds in The Netherlands. Names of plant communities according to the Zürich-Montpellier school of vegetation science (after Verhoeven & Bobbink (2001) and Van Wirdum et al. (1992)). Historical setting and succession in turf ponds

Turf pond complexes in The Netherlands have mostly been excavated below the water table in the 18th and early 19th century. They are now characterized by very species-rich fen vegetation. As many as 35 different vegetation types have been described for such habitats in The Netherlands (Van Wirdum et al. 1992). There are three different succession series, which each characterize a specific water chemistry in the open-water stage, i.e. brackish, fresh mesotrophic and fresh eutrophic. The succession goes through 7 stages (aquatic, floating raft, helophyte, brown moss, transitional fen and carr) and has been described by space-for-time substitution of an extensive database of vegetation relevées, see Figure 1, (Van Wirdum 1992, Verhoeven & Bobbink 2001).

95

Figure 1. Succession seres in terrestrializing ponds in The Netherlands. Names of plant 55 communities according to the Zürich-Montpellier school of vegetation science (after Verhoeven & Bobbink (2001) and Van Wirdum et al. (1992)).

The Polder Westbroek/Tienhoven

The polder Westbroek/Tienhoven is located 10 km North of Utrecht. Originally this area has been a peat bog, which was drained and converted to agriculture in the 15th-16th century. The area has several complexes of turf ponds, which have terrestrialized in the past 100 years. A number of late-succession ponds covered with Alder forest have been re-excavated in the 1990s and some extra turf ponds have been created at the time.

This excursion takes us on a 5-km walk across the Bert Bos-path and shows the various succession stages and the current status of the ponds that had been restored/created 25 years ago.

The eastern Vechtplassen area is situated north of the city of Utrecht and contains many turbaries in the polder Westbroek/Tienhoven (Figure 2). This is an example of at least 6 major complexes of long, rectilinear ponds located in a zone between the upper sandy areas and the low coastal plain in The Netherlands (Beltman et al. 2011a).

96

The Polder Westbroek/Tienhoven Fig. 2. Location of the polder Westbroek/Tienhoven.

The polder Westbroek/Tienhoven is located 10 km rectilinear ponds located in a zone between the NorthFig. 2. of Utrecht.Location Originally of the thispolder area Westbroekhas been a peat/Tienhoven upper. sandy areas and the low coastal plain in The bog, which was drained and converted to agriculture in Netherlands (Beltman et al. 2011a). theThe 15th ponds-16th century. have The area been has several created complexes by ofpeat dredgingThe ponds have in beenthe created 18th by- 19thpeat dredging centuries, and turfhave ponds, filled which uphave againterrestrialized with in vegetationthe past 100 andin the peat 18th-19 thsince centuries, then. and have They filled occur up again in a peat- years.meadow A number landscape of late-succession that ponds had covered been with reclaimedwith vegetation in the and peat14th since-16th then. Theycenturies occur in a (Borger Alder1992). forest Thehave been turf re-excavated ponds inare the 1990sstill andembedded peat-meadow in a landscape landscape that hadof beenpeat reclaimed meadows in used th th somefor cattleextra turf grazing. ponds have beenThe created vegetation at the time. in suchthe 14turf-16 pond centuries complexes (Borger 1992). Theis turfquite ponds varied and Thisspecies excursion-rich. takes Theus on aspecies 5-km walk- acrossrich theplant Bert communitiesare still embedded in inthe a landscape terrestrializing of peat meadows fens in turf Bos-path and shows the various succession stages and used for cattle grazing. The vegetation in such turf ponds are undergoing succession and can be attributed to the Phragmites and the current status of the ponds that had been restored/ pond complexes is quite varied and species-rich. The the brownmoss succession series, which are typically associated with created 25 years ago. species-rich plant communities in the terrestrializing mesotrophicThe eastern Vechtplassen and areaslightly is situated eutrophicnorth fensconditions in turf ponds are(Van undergoing Wirdum succession et and canal. 1992, ofVerhoeven the city of Utrecht & Bobbinkand contains 2001). many turbaries be attributed to the Phragmites and the brownmoss in the polder Westbroek/Tienhoven (Figure 2). This succession series, which are typically associated with is an example of at least 6 major complexes of long, mesotrophic97 and slightly eutrophic conditions (Van

56 These conditions were quite common in The Netherlands until the 1950s and occurred in groundwater discharge areas and floodplain areas of streams and small rivers. Complexes of turf ponds with these terrestrializing fen systems can be found in The Netherlands where the Holocenic low fen areas border the higher sandy areas of Pleistocenic origin, e.g. Rottige Meente, De Wieden, De Weerribben and the Vechtplassen area.

Vegetation map 1937 Vegetation map 1983

Fig. 3. Vegetation structure in the turf ponds of Westbroek in and acidification of its surroundings (Kooijman and 1937 and 1983, based on digitized aerial photographs (Bakker Fig.et al. 3. 1994). Vegetation Water: open structure water with submergedin the turf vegetation; ponds of PaulissenWestbroek 2006). in At1937 the end and of the1983, 1990s, based brownmoss on digitized aerialReed: helophyte-dominatedphotographs (Bakker vegetation; et Fen: al. brown 1994). moss; Water: fens open had waterbecome withovergrown submerged by Sphagnum vegetation; within a Reed: helophyteForest: carr. -dominated vegetation; Fen: brown moss;few Forest:years’ time. carr Also,. the characteristic brownmoss Scorpidium scorpioides, which was probably very InWirdum a study et al. 1992, of Verhoevenhistoric & Bobbinkaerial 2001).photographscommon of the in the52 16-hath century study but areahad already in thebecome eastern VechtplassenThese conditions werearea quite commonspanning in The a 50-yearrare in thetime 1980s (Faberperiod et al. 2017),(1937 has now-1987), disappeared four successionalNetherlands until stagesthe 1950s andwere occurred identified, in i.e. openaltogether. water, The State tall Forest reed, Service, short who owned fen andmeadow andgroundwater carr forestdischarge (Bakkerareas and floodplain et al. areas1994). of Manymanaged ponds this fen area,in the decided Westbroek to re-excavate/Tienhoven a number streams and small rivers. Complexes of turf ponds of ponds to restore the early-succession vegetation polder were still open water in 1937, but terrestrialized rather readily in a period with these terrestrializing fen systems can be found types, so that the complete successional series would oinf The20 -Netherlands40 years, where with the Holocenicthe share low fen of areas open waterbe safeguarded decreasing on the long andterm (Beltman that of et al.carr 1996). forest stronglyborder the higherincreasing sandy areas over of Pleistocenic time. Byorigin, the endIn of the the course 1980s, of 30 years the (1985-2015), area athad least lost 45 ponds most of itse.g. Rottigeearly Meente,-succession De Wieden, stages, De Weerribben while and thesome were mid restored-succession or created, stronglyspecies increasing-rich thefens share were maintainedVechtplassen area. by a mowing regime (Figureof 3). open-water These ponds systems in the area. did, Initial however, monitoring show signs of acidification, which is an indicationactivities of indicated moving that forwardrecolonization in of succession the ponds to In a study of historic aerial photographs of the 52-ha was slow and differed quite substantially between transitional fen. This process was probably enhanced because of atmospheric study area in the eastern Vechtplassen area spanning different ponds, which was hypothetically attributed to depositiona 50-year time periodof NH (1937-1987),y and fourNO successionalx (Paulissen etdispersal al. 2014), issues, to changesbut particularly in the water chemistry by surfaceand waterstages were pollution identified, and i.e. open eutrophication water, tall reed, short with P,to herbivorywhich byled invasive to increased musk rats (Beltman growth et al. 2011b, of fast - growingfen meadow andSphagnum carr forest (Bakker species et al. 1994). such Many as S.Sarneel squarrosum et al. 2011). and acidification of its surroundingsponds in the Westbroek/Tienhoven (Kooijman polder and were Paulissen still 2006). At the end of the 1990s, Water flow and water chemistry brownmossopen water in 1937, fens but terrestrialized had become rather overgrownreadily in by Sphagnum within a few years’ time. Also,a period the of 20-40 characteristic years, with the share brownmoss of open water ScorpidiumThe Westbroek scorpioides, polder is a flat whicharea with wasdrained probably decreasing and that of carr forest strongly increasing fenlands, situated close to a hill ridge of sandy very common in the 16th century but had already become rare in the 1980s over time. By the end of the 1980s, the area had moraines. The drainage occurs through parallel (Faberlost most ofet its al. early-succession 2017), has stages, now while disappearedsome ditches altogether. about 50 m apart, The which State date backForest from the Service, whomid-succession owned species-rich and managed fens were maintained this fen by aarea, era ofdecided peatland reclamation to re-excavate in the 14th-15 tha centuries number of mowing regime (Figure 3). These systems did, however, (Borger 1992). The general direction of surface water show signs of acidification, which is an indication 98flow, fed by rain water and groundwater discharge, is of moving forward in succession to transitional fen. towards the south-west. Groundwater discharge from This process was probably enhanced because of the sandy ridge toward the polder used to result in a

atmospheric deposition of NHy and NOx (Paulissen et water chemistry in the drainage ditches and the turf al. 2014), but particularly by surface water pollution and ponds which was favourable for the mesotrophic plant eutrophication with P, which led to increased growth communities mentioned. The water discharging in the of fast-growing Sphagnum species such as S. squarrosum polder Westbroek from this aquifer is relatively low in

57 This lake has a good water quality which is chemically rather similar to the groundwater in the aquifer. The regional water regime controlled by pumping stations and weirs was modified in such a way that the water from the Breukeleveense Plas was directed towards Westbroek in dry summer periods. This has improved the situation in the polder to the point that no signs of negative impacts on the submerged vegetation in the ditches are visible any longer.

a consequence, signs of deterioration of the quality of the submerged vegetation in ditches were observed in the early 1980s and there was increasing concern that the species-rich turf ponds would become negatively affected as well. The observed effects of the polluted river water in dry summer periods led to a decision by the water board to change the surface water source to feed the polder Westbroek/Tienhoven in the early 1990s; the new source was the Breukeleveense Plas, a shallow lake just 1 km west of the polder (Figure 4). This lake has a good water quality which is chemically rather similar to the groundwater in the aquifer. The regional water regime controlled by pumping stations and weirs was modified in such a way that the water from the Breukeleveense Plas was directed towards Westbroek in dry summer periods.

Figure 4. Water chemistry in the Westbroek polde ditches This has improved the situation in the polder to

in summer 1983. The Stiff diagrams give the proportional the point that no signs of negative impacts on the Figure 4. Water chemistry inconcentrations the Westbroek as polde% of total ditches electrical in summer postive 1983. and negative The Stiff diagramssubmerged give vegetation in the ditches are visible any charge. Water types dominated by Na, K and Cl occur in the the proportional concentrations as % of total electrical postive and negative charge. Waterlonger. types dominated by Na, K and Cl occursouthern in thepart southern of the polder, part while of the in polder,the northern while part in the the northern part the This major change in the water management of the water is relatively rich in Ca waterand HCO is relatively3. rich in Ca and HCO3. polder has prevented the precious plant communities This major change in nutrients,the water chloride management and sulphate, of butthe rich polder in calcium has preventedin the turfthe ponds to be affected by the polluted river precious plant communities in the turf ponds to be affected by the polluted river and bicarbonate ions. water. The water chemistry in the turf ponds remained water. The water chemistry in the turf ponds remained favorable even in the However, these mesotrophic to moderately eutrophic favorable even in the 1980s, when the polluted water 1980s, when the polluted water was penetrating the ditch system in dry systems became disturbed by polluted surface water was penetrating the ditch system in dry summers. The summers. The slow flow rates in the exchange between turf ponds and ditches, as well as the timely measuresince the of1960s. changing As groundwater the surface resources water in the source to theslow polder flow rates in the exchange between turf ponds and have played a positive hillrole. ridge The were current increasingly water used chem for drinkingistry of water the Westbroekditches, turf as well as the timely measure of changing the ponds is characterizedabstraction, by thelow amount ammonium, of groundwater phosphate discharged inand surfacesulphate water source to the polder have played a positive concentrations, which theis polderfavorable decreased for andthe more development surface water of supply initial, submergedrole. The current water chemistry of the Westbroek turf was needed to keep water levels sufficiently high during ponds is characterized by low ammonium, phosphate 100 dry periods in summer. This surface water originated and sulphate concentrations, which is favorable for the

from the Vecht river and, ultimately, from the Rhine development of initial, submerged stages of the brown and was polluted with nitrogen and phosphorus. It moss succession series (Geurts et al. 2008), although also had high concentrations of potassium, chloride this has not yet happened. and sulphate (Beltman et al. 1988, Beltman & Verhoeven 1988, Beltman & Rouwenhorst 1991). Restoration: digging of existing and As a result of the surface water supply, the water new ponds chemistry in the polder Westbroek/Tienhoven changed As indicated above, the terrestrialization process considerably in a non-favorable direction, as can be associated with the succession from open-water plant seen in the composition of the supply water in the communities to floating fens and finally Alder and 1980s. As the direction of surface water flow during Willow forest, had resulted in a situation in which the such dry periods was reversed, i.e. towards the north- species-rich phases became under threat by progressive east, ditches in the southern parts of both polders acidification and forest expansion (Figure 5). showed a strongly modified water chemistry with high Helped by the National Nature Policy Plan sodium, potassium and chloride and low calcium and (1990), the major nature conservation agencies in bicarbonate (Figure 4, Beltman & Rouwenhorst, 1991) As Westbroek/Tienhoven, Staatsbosbeheer (SBB) and Natuurmonumenten (NM), started to make plans for:

58 (1) Enhancing the management of high-quality species- Figure 6 shows a schematic overview of the rich floating fens by consistent annual mowing in restoration actions that were designed in the 1990s; this the summer and by removing young trees and tree figure worked as an initial framework for the activities. seedlings; Along the way, variations of the exact width, bank (2) Removing the top soil in pastures and abandoned shape (90o with ground level or less steep) and depth agricultural fields to restore species-rich wet (standard 1 m) were tried out, resulting in a number of grassland vegetation; ponds with different specifications, in particular in the (3) Excavating part of the turf ponds that had area with ‘new’ ponds in the northern section of the terrestrialized into late-succession Alder forests, ponds shown in Figure 5. while another part remained untouched to protect As will be shown during the excursion, these the final successional stages as well. restoration activities have resulted in a very promising vegetation development, with difference in plant These long-termstages restorationof the brown activities moss were plannedsuccession seriescommunity (Geurts structure et and al .speed 2008), of succession although in ponds this and overseenhas not by a yteamet happened. of experts from the Nature with different morphometry. conservation agencies (SBB and NM), scientists (in These developments are being monitored and will particular Boudewijn Beltman from Utrecht University) add to the knowledge base for maintaining the actions stagesand landof the management brown contractors.moss succession This team has series (Geurtsneeded foret conservational. 2008), of althoughthe complete this succession has workednot yRestoration: etfrom happened. 1990 until 2010 diggingand the resulting of existingof and these terrestrializingnew ponds fen communities. Some landscape now shows a drastic increase of the number illustrative examples:

of open-waterAs indicated turf ponds, above, while all theother terrestrializationstages of the • processSpecies-rich associated floating mats with(Carex rostrata,the succession Potentilla successionfrom are open also well-represented.-water plant During communities the 20 to floatingpalustris, fens Calla palustris)and finally have establishedAlder and in smallWillow Restoration: digging of existing and new ponds years offorest, carrying hadout this resulted work, a lot in of a experience situation had in whichpatches the species locally and-rich slowly phases expand became from the banks under threat by progressive acidification and forest expansion (Figure 3). As indicatedto be built up above, regarding the the technicalterrestrialization aspects of the process• associatedSome ponds are with fully theovergrown succession with emergents fromexcavations open-water with modern, plant heavycommunities equipment, insteadto floating fens(Typha and latifolia, finally T. angustifolia) Alder and after Willow 25 years, while forest,of the had hand resulted labor in former in a centuries,situation in ain peaty which the speciesothers-rich (mostly phases deep and became with steep under banks) have not threatenvironment. by progressive acidification and forest expansionshown (Fig muchure vegetation 3). development yet.

Fig. 5. The situation of open turf ponds (blue) in 2005. Fig. 5. The situation of open turf ponds (blue) in (Staatsbosbeheer).2005. (Staatsbosbeheer).

Helped by the National Nature Policy Plan (1990), the major nature conservation 59 agencies in Westbroek/Tienhoven, Staatsbosbeheer (SBB) and Fig. 5. The Natuurmonumentensituation of open turf ponds (NM), (blue) started in 2005. to (Staatsbosbeheer). make plans for :

Helped by(1) the Enhancing National Naturethe management Policy Plan (1990),of high -thequality major species nature- richconservation floating fens by agencies consistentin Westbroek/Tienhoven, annual mowing in the summerStaatsbosbeheer and by remo(SBB)ving youngand trees and Natuurmonumententree seedlings; (NM), started to make plans for:

(1) Enhancing the management of high-quality species-rich floating fens by consistent annual mowing in the summer and by removing young trees and tree seedlings; 101

101

Maintain than newly created ponds to already contain diaspores Existing turf species- left over in the seed bank of the shore or the pond ponds rich fens bottom (Bekker et al. 1998, Klimkowska et al. 2010). However, due to excavation activities a large part of Maintain the existing seed bank is removed and colonization by high-quality dispersal in space also becomes important. Arrival of forest propagules to new habitats is often a time-consuming process, with species that possess traits enhancing Pasture or Removal of long-distance dispersal arriving sooner and in greater maize field top soil numbers (Brederveld et al. 2011, Fraaije et al. 2015). However, also the distance between habitats and their Excavate Standard size, connectivity plays an important role: nearby source forested turf depth and populations facilitate colonization. In the successional ponds and shape pastures series relevant here, three main dispersal mechanisms determine the dispersal of plant species and the Banks less connectivity between habitat areas: dispersal by water, steep & wind, and waterfowl. shallow For species characteristic of aquatic and shoreline vegetation, and for the early-succession, aquatic Figure 6. Schematic overview of activities in the restoration phases of the series, species dispersal by water is of turf-pond vegetation in Westbroek/Tienhoven (Beltman 2016). particularly important. Water is widely available as a dispersal vector, and, perhaps even more importantly, • The restoration works have not resulted in adverse is likely to connect wet, suitable sites. Seeds of early- eutrophication and water quality has only improved. succession species may float for months, particularly • Mat formation is locally hampered by grazing in stagnant or slow-flowing water (Van den Broeket by non-native species (Muskrat, Red American al. 2005). Vegetative fragments may float for much crayfish). shorter periods of time, but may be highly effective in • Generally, invertebrate communities and birds have colonizing new areas (Boedeltje et al. 2003, Boedeltje improved strongly in the past 20 years. et al. 2004, Boedeltje et al. 2008). A disadvantage of the turbary system in regard to water dispersal is that Plant dispersal the water in the system hardly flows – the system At The large number of plant species associated with resembles a set of interconnected lakes rather than the succession from open water to carr forest and the a stream. Hence, dispersal by water is very slow and rapid turnover of these species during succession imply directed by wind, pushing the floating seeds downwind, that dispersal to new or restored ponds is essential. In particular in ditches and linear ponds (Soomers et al. natural landscapes with oxbow lakes such dispersal 2013, Sarneel et al. 2014). It has been shown that the would have been facilitated by the proximity of orientation of the fen pond complexes towards the existing populations serving as propagule sources and prevailing wind direction may affect dispersal distances by the ample availability of dispersal vectors water, to a considerable degree (Soomers et al., 2013). The fens wind and waterfowl. In the turf ponds, past dispersal in our study area are orientated in a NE→SW direction, capacities likely consisted also of dispersal by these so that wind- enhanced hydrochorous dispersal is same vectors, while the direct connections via surface expected to be quite effective considering the prevailing water to nearby propagule sources compensated for the south-westerly winds. Eventually, water-dispersed lack of flowing (river) water. In the current situation, seeds in such systems are likely to be deposited at where early-successional stages in the landscape are (north-eastern) shorelines at times of receding water sparse and far apart, dispersal limitation has become levels in spring, followed by a rapid germination and an issue likely to limit the re-colonization of newly locally high seedling densities (Sarneel & Soons 2012, created ponds. Ponds that had terrestrialized before Van Leeuwen et al. 2014). Another limiting factor to and are being re-excavated will have a better chance seed dispersal by water is that it only reaches sites

60 PhragmitesPhragmites and and Typha Typha, but, but also also carr carr forest forest trees trees (Alnus (Alnus, Betula, Betula, Salix, Salix) disperse) disperse veryvery well well by by wind. wind.

OtherOther problems problems with with terrestrialization terrestrialization. .

In InIn Inthe the past past few few years, years, it itbecame became clear clear that that succession succession in inthe the new new ponds ponds was was notnot very very successful successful yet, yet, despite despite improvement improvement of ofsurface surface water water quality. quality. This This may may havehave been been due due to todispersal dispersal problems problems (see (see above), above), or orlack lack of oftime, time, as as it ittakes takes aboutabout 50 50-60-60 years years for for new new rich rich-fens-fens to todevelop develop (Bakker (Bakker et etal. al. 1994; 1994; Faber Faber et etal. al. 2017).2017). We We tried tried to tospeed speed up up this this process process with with the the use use of offloating floating platforms platforms (Loeb (Loeb et etal. al. 2016; 2016; in inDutch), Dutch), but but this this did did not not really really help help (Figure (Figure 7). 7). When When protected protected fromfrom grazing grazing, aboveground, aboveground plant plant biomass biomass could could become become extremely extremely high, high, and and in in thethe best best case case still still mainly mainly consisted consisted of ofeutrophic eutrophic plant plant species. species. When When the the floating floating platformsplatforms were were not not protected, protected, the the vegetation vegetation was was eaten eaten by by Grey Grey goose goose (Anser (Anser anseranser) or) orAmerican American crawfish crawfish (Procambarus (Procambarus clarkii clarkii). ).Especially Especially the the latter latter species species is is becomingbecoming a reala real problem, problem, and and will will be be the the focus focus of ofa newa new OBN OBN-project-project starting starting this this year.year.

Figure 7. Experiment with floating platforms to speed up the Figurethat are 7.connected Experiment via surface with water floating so that platforms dispersal to speed up the terrestialization process from aquatic Figure 7. Experiment with floating platforms to terrestializationspeed up the process terrestialization from aquatic vegetation process to floatingfrom aquatic vegetatibetweenvegetati differenton onto tofloating floatingcatchments rich rich fensor fenshydrological (Loeb (Loeb et unitsetal. al. 2016;is 2016; inrich Dutch).in fens Dutch). (Loeb et al. 2016; in Dutch). not possible. The scale at which dispersal by water takes place is therefore highly site-specific, and dependent on speed up this process with the use of floating platforms the hydrological connections by surface water and the (Loeb et al. 2016; in Dutch), but this did not really help FurtherflowFurther velocity andinformation/further information/furtherdirection of the system. reading reading(Figure 7). When protected from grazing, aboveground Dispersal by wind is a common phenomenon in plant biomass could become extremely high, and in particular for later successional species. Increasingly the best case still mainly consisted of eutrophic plant terrestrial stages have increasingly higher proportions of species. When the floating platforms were not protected, BakkerspeciesBakker with SA, traits SA, facilitatingVan Van den den long-distance Berg Berg NJ dispersalNJ & &Speleers by Speleers the vegetation BP BP (1994). was(1994). eaten byVegetation GreyVegetation goose (Anser transitions anser)transitions or of of wind (Soonsfloatingfloating 2006). Thewetlands wetlandsadvantage in of inathis complexa mechanism complex of is of turbaries turbariesAmerican between crawfishbetween (Procambarus 1937 1937 and and clarkii). 1989 1989 Especially as asdetermined determined the that wet, fromunsuitablefrom aerial aerial patches photographs photographs can be bridged with and with suitable GIS. GIS. Vegetatio Vegetatiolatter species 114:161 114:161 is becoming-167.-167. a real problem, and will be the sites across the water can be connected. In addition, focus of a new OBN-project starting this year. this dispersal mechanism is not limited to catchment boundaries or hydrological units. A disadvantage is 105Further105 information/further reading that this process is not limited to wetlands but rather Bakker SA, Van den Berg NJ & Speleers BP (1994). Vegetation distributes seed across the landscape, so that dispersal transitions of floating wetlands in a complex of turbaries is not directed towards suitable sites (Soons 2006). The between 1937 and 1989 as determined from aerial scale at which dispersal by wind takes place is much less photographs with GIS. Vegetatio 114:161-167. site-specific than for water dispersal, and much more Bekker RM, Schaminee JHJ, Bakker JP, & Thompson K (1998). dependent on the traits of the species dispersing: plant Seed bank characteristics of Dutch plant communities. Acta species that release their seeds from a great height and Botanica Neerlandica 47:15-26. produce very light seeds (with a low terminal velocity) Beltman B (2016). Veenvormende zodden in Westbroek en may disperse their seeds over hundreds to thousands of Tienhoven. Anoda, Utrecht. meters (Soons et al. 2004a, Soons et al. 2004b, Soons et Beltman B, Barendregt A, Kooijman K, Smolders F. & G. Ter al. 2005). Typical emergent species such as Phragmites Heerdt G (2011a). Landschapsdoorsnede laagveenlandschap. in H. Van Dobben, A. J. M. Jansen, and N. A. C. Smits, editors. and Typha, but also carr forest trees (Alnus, Betula, Salix) Alterra/Unie van Bosschappen, Wageningen. disperse very well by wind. Beltman B, Duel H, Van der Bie M, Otten E & Rouwenhorst G Other problems with (1988). Ecohydrologie in polders: het Noorderpark. Landschap terrestrialization 1988:152-169. In In the past few years, it became clear that succession Beltman B, Omtzigt NQ, & Vermaat JE (2011b). Turbary Restoration Meets Variable Success: Does Landscape Structure in the new ponds was not very successful yet, despite Force Colonization Success of Wetland Plants? Restoration improvement of surface water quality. This may have Ecology 19:185-193. been due to dispersal problems (see above), or lack of Beltman B., & Rouwenhorst TG (1991). Ecohydrology and fen time, as it takes about 50-60 years for new rich-fens to plant distribution in the Vechtplassen area, the Netherlands. develop (Bakker et al. 1994; Faber et al. 2017). We tried to Proceedings IAHS meeting Vienna:199-213.

61 Beltman B, Van den Broek T, Van Maanen K & Vaneveld K laagveenpetgaten: Speerpunt voor natuurherstel in laagvenen. (1996). Measures to develop a rich-fen wetland landscape Rapport nr. 2016/OBN208-LZ, VBNE, Vereniging van Bos- en with a full range of successional stages. Ecological Engineering Natuurterreineigenaren, Driebergen. 7:299-314. Paulissen MP, Schaminee JH, During HJ, Wamelink G & Beltman B, & Verhoeven JTA (1988). Distribution of fen plant Verhoeven JTA (2014). Expansion of acidophytic late- communities in relation to hydrochemical characteristics in successional bryophytes in Dutch fens between 1940 and 2000. the Vechtplassen area. Pages 121-136 in J. T. A. Verhoeven, Journal of Vegetation Science 25:525-533. editor. Structure of vegetation in relation to carbon and nutrient economy. SPB Academic Publischers, The Hague. Sarneel JM, Beltman B, Buijze A, Groen R, & Soons MB (2014). The role of wind in the dispersal of floating seeds in slow- Boedeltje G, Bakker JP, Bekker RM, Van Groenendael JM, & flowing or stagnant water bodies. Journal of Vegetation Science Soesbergen M (2003). Plant dispersal in a lowland stream in 25:262-274. relation to occurrence and three specific life-history traits of the species in the species pool. Journal of Ecology 91:855-866. Sarneel JM, & Soons MB (2012). Post-dispersal probability of germination and establishment on the shorelines of slow- Boedeltje G, Bakker JP, Ten Brinke A, Van Groenendael flowing or stagnant water bodies. Journal of Vegetation Science JM & M. Soesbergen M (2004). Dispersal phenology of 23:517-525. hydrochorous plants in relation to discharge, seed release time and buoyancy of seeds: the flood pulse concept supported. Sarneel JM, Soons MB, Geurts JJ, Beltman B & Verhoeven Journal of Ecology 92:786-796. JTA. (2011). Multiple effects of land-use changes impede the colonization of open water in fen ponds. Journal of Vegetation Boedeltje G, Ozinga WA & Prinzing A (2008). The trade-off Science 22:551-563. between vegetative and generative reproduction among angiosperms influences regional hydrochorous propagule Soomers H, Karssenberg D, Soons MB, Verweij PA, Verhoeven pressure. Global Ecology and Biogeography 17:50-58. JTA, & Wassen MJ (2013). Wind and Water Dispersal of Wetland Plants Across Fragmented Landscapes. Ecosystems Borger GJ (1992). Draining-digging-dredging; the creation of a 16:434-451. new landscape in the peat areas of the low countries. Pages 131-172 in J. T. A. Verhoeven, editor. Fens and bogs in The Soons MB (2006). Wind dispersal in freshwater wetlands: Netherlands: Vegetation, history, nutrient dynamics and Knowledge for conservation and restoration. Applied conservation. Kluwer Academic Publishers, Dordrecht. Vegetation Science 9:271-278.

Brederveld RJ, Jaehnig SC, Lorenz AW, Brunzel S., & Soons Soons MB, Heil GW, Nathan R, & Katul GG (2004a). MB (2011). Dispersal as a limiting factor in the colonization of Determinants of long-distance seed dispersal by wind in restored mountain streams by plants and macroinvertebrates. grasslands. Ecology 85:3056-3068. Journal of Applied Ecology 48:1241-1250. Soons MB, J. H. Messelink JH, Jongejans E & Heil GW (2005). Faber AH, Kooijman AM, Brinkkemper O & B. Van Geel Habitat fragmentation reduces grassland connectivity for both B (2016). Palaeoecological reconstructions of vegetation short-distance and long-distance wind-dispersed forbs. Journal successions in two contrasting former turbaries in the of Ecology 93:1214-1225. Netherlands and implications for conservation. Review of Palaeobotany and Palynology 233: 77-92. Soons MB, Nathan R & Katul GG (2004b). Human effects on long-distance wind dispersal and colonization by grassland Fraaije RG, Ter Braak CJ, Verduyn B, Verhoeven JTA & Soons plants. Ecology 85:3069-3079. MB (2015). Dispersal versus environmental filtering in a dynamic system: drivers of vegetation patterns and diversity Van den Broek T, Van Diggelen R, & R. Bobbink R. (2005). along stream riparian gradients. Journal of Ecology 103:1634- Variation in seed buoyancy of species in wetland ecosystems 1646. with different flooding dynamics. Journal of Vegetation Science 16:579-586. Geurts JJM, Smolders AJP, Verhoeven JTA, Roelofs JGM & Lamers LPM (2008). Sediment Fe : PO4 ratio as a diagnostic Van Leeuwen CHA, Sarneel JM, Van Paassen J, Rip WJ & and prognostic tool for the restoration of macrophyte Bakker ES (2014). Hydrology, shore morphology and species biodiversity in fen waters. Freshwater Biology 53:2101-2116. traits affect seed dispersal, germination and community assembly in shoreline plant communities. Journal of Ecology Klimkowska A, Van Diggelen R, GrootjansAP & Kotowski 102:998-1007. W. (2010). Prospects for fen meadow restoration on severely degraded fens. Perspectives in Plant Ecology Evolution and Van Wirdum G, Den Held AJM, & M. Schmitz M (1992). Systematics 12:245-255. Terrestrializing fen vegetation in former turbaries in The Netherlands. Pages 323-360 in J. T. A. Verhoeven, editor. Fens Kooijman AM & Paulissen MPCP (2006). Acidification rates in and bogs in The Netherlands: vegetation, history, nutrient wetlands with different types of nutrient limitation. Applied dynamics and conservation. Kluwer Academic Publishing, Vegetation Science 9: 205-212. Dordrecht.

Loeb L, Geurts J, Bakker L, Van Leeuwen R. Van Belle J., Verhoeven JTA, & Bobbink R. (2001). Plant diversity of fen Van Diggelen J, Faber AH, Kooijman AM, Brinkkemper landscapes in The Netherlands. Pages 65-87 in B. Gopal, editor. O, Van Geel B, Weijs W. Van Dijk G Loermans L, Biodiversity in wetlands: assessment, function and conservation. Cusell C, Rip W & Lamers LPM (2016). Verlanding in Backhuys Publishers, Leiden, The Netherlands.

62 Peat meadowsPeat meadowsin The Netherlands in The; Netherlands; a millenniuma millennium of soil subsidence of soil. subsidence

Jos Verhoeven, Ab Grootjans Jos Verhoeven, Ab Grootjans.

Intensively drained agricultural Intensively drainedpolder areas agricultural polder areas eat meadows are drained peatlands that are Figure 1. Wetlands in The Netherlands in the year 100 AD (after Zagwijn 1975). Peat meadows are drained peatlands that are in use for agriculture. Peat in use for agriculture. Peat meadow areas in meadow areas in the Netherlands have mostly originated from drainage and the Netherlands have mostly originated from Figure 1. Wetlands in The Netherlands in the year 100 AD reclamation of extensive bogs and fens in the coastal plainFigure that 1 . usedWetlands to be in Thethe Netherlands in the year 100 AD (after Zagwijn 1975). P (after Zagwijn 1975). drainage and reclamation of extensive bogs and fens most important land cover until 1000 AD (Figure 1). Most of these reclamations inth the coastalth plain that used to be the most important started in the 12 – 15 century and have brought farmers centuries of relatively land cover until 1000 AD (Figure 1). Most of these high crop production and grazing, particularly in the initial phases. These th th ‘polders’ (diked reclamationsareas with starteda manipulated in the 12 – water15 century level) and have been in agricultural use for centuries.have Right brought from farmers the centuries start, these of relatively areas high have crop been subject to peat shrinkage and soilproduction subsidence and grazing,, which particularly has created in the initial the typical polder landscape with water levelsphases. in ditches These ‘polders’that are (diked higher areas than with the a manipulated land (Figure 2). water level) have been in agricultural use for centuries.

Right from the start, these areas have been subject to peat shrinkage and soil subsidence, which has created the typical polder landscape 108with water levels in ditches that are higher than the land (Figure 2). The rate of subsidence has risen strongly in the past 50 years as a result of deeper drainage and intensive

agricultural use and has increasingly been accompanied Figure 2. Typically Dutch peat polder landscape with the Figure 2. Typically Dutch peat polder landscape with the water higher than the land. by deterioration of water quality. There is concern that water higher than the land. Photo: Hans Joosten. Photo: Hans Joosten.

Figure 2. Typically Dutch peat polder landscape with the water higher than the land. 63 Photo: Hans Joosten. 109

109

The rate of subsidence has risen strongly in the past 50 years as a result of deeper drainage and intensive agricultural use and has increasingly been accompanied by deterioration of water quality. There is concern that climate climate change will further aggravate the problems in The western area encompasses parts of the provinces change will further aggravatthese areas ande thenecessitates problems adaptations in ofthese land use areasof Noord-and andnecessitates Zuid-Holland and Utrecht, whereas the adaptations of land useand and water water management. management. northern part is located in the provinces of Friesland, Between There are two regions with substantial Overijssel and Drenthe. The largest area in the western Between There are twoareas regions of peat meadows with insubstantial the Netherlands, areas i.e. the of peatregion meadows is characterized in bythe its location within a ring of Netherlands, i.e. the westernwestern peat peat meadow meadow area and the area northern and peat the northernurbanized peat and denselymeadow populated areas (the ‘Randstad’) area (Figure 3), whichmeadow also areashows (Figure the3), which current also shows area the current of peat soils,and is often covering called the less‘Green Heart’. The peat meadow than 20% of the originalarea peatlandof peat soils, coveringextent). less than 20% of the original areas are in use for intensive dairy production, but at peatland extent). the same time are visited by large numbers of people from the surrounding cities for recreation and leisure. In the western peat meadow region, current water tables are mostly around 60 cm below ground level, whereas in the northern peat meadows (Friesland) water tables are often kept down to 100-120 cm below ground level (see Figure 4). This reflects a greater

emphasis on agricultural targets and smaller interest of nature conservation values in the latter region. As a

result of the actions of private land owners (farmers), the lowest parcels in the landscape are the ones with the most intensive agricultural use. The water management in the peat meadow areas is complex and primarily geared towards the current agricultural use. To maximize agricultural production, water tables are kept low (0.6 – 1 m below ground level) by collecting drainage water in ditches and reservoirs eventually pumping it out of the polders. Current agricultural practice requires relatively dry field conditions in early spring to enable access by

heavy machinery and pasturing of dairy cattle. Land use Figureprivate 3. Peat soilland is The ownersNetherlands; western(farmers), area (left), the lowest parcels in the landscape are the ones planning programs in the 1960s-1990s have aimed at Figure 3. Peat soil is Thenorthernwith Netherlands; areathe (right) most (Rienkswestern intensive & Gerritsen area 2005). (left),agricultural 1= Zegveld,northern use.area (right) (Rienks & Gerritsen 2005). 1= Zegveld,2= Tjeukemeer, 2= Tjeukemeer, 3= Bovensmilde. 3= Bovensmilde.

The western area encompasses parts of the provinces of Noord- and Zuid- Holland and Utrecht, whereas the northern part is located in the provinces of Friesland, Overijssel and Drenthe. The largest area in the western region is characterized by its location within a ring of urbanized and densely populated areas (the ‘Randstad’) and is often called the ‘Green Heart’. The peat meadow areas are in use for intensive dairy production, but at the same time are visited by large numbers of people from the surrounding cities for recreation and leisure. In the western peat meadow region, current water tables are mostly around 60 cm below ground level, whereas in the northern peat meadows (Friesland) water tables are often kept down to 100-120 cm below ground level (see Figure 4). This reflects a greater emphasis on agricultural targets and smaller interest of nature conservation values in the latter region. As a result of the actions of Figure 4. Typical soil profile of a peat meadow in the Province of Friesland. The top 80 cm are well drained and show amorphic, very decomposed peat, while the layers deeper than 90 cm are clearly much 110less decomposed and show well structures plant remains.

64 Figure 4. Typical soil profile of a peat meadow in the Province of Friesland. The top 80 cm are well drained and show amorphic, very decomposed peat, while the layers deeper than 90 cm are clearly much less decomposed and show well structures plant remains.

The water management in the peat meadow areas is complex and primarily geared towards the current agricultural use. To maximize agricultural production, water tables are kept low (0.6 – 1 m below ground level) by collecting drainage water in ditches and reservoirs eventually pumping it out of the polders. Current agricultural practice requires relatively dry field conditions in early spring to enable access by heavy machinery and pasturing of dairy cattle. Land use planning programs in the 1960s-1990s have aimed at maximizing agricultural production by re-allotments and by deeper and more effective drainage.

As a result of the oxidation of peat after exposure to oxygen has been accelerated substantially in this period. Soil subsidence in organic soils is not restricted to The Netherland, but it is a world-wide problem. Soil subsidence may amount to 2 cm annually (in tropical areas) and generates enormous greenhouse

gas (GHG) emissions (20-40 t CO2 per ha and year annually) and eventually often loss of productive land (Joosten 2015). Subsidence and erosion of the peat soil provoke insufficient drainage, ponding water following soil compaction and drought prone top soils due to the hydrophobic nature of degraded peat (Zeitz & Velty 2002; Wallor & Zeitz 2016; Joosten et al. 2017). These adverse soil conditions raise costs of water management including a higher water demand for irrigation (Regional Water Authority Hunze en Aa’s 2018).

In The Netherlands the height of the water table in polders is fixed by the water authorities based on the requirements of agriculture, nature conservation or housing. Individual landowners often install small-scale pumps for additional

111 maximizing agricultural production by re-allotments hydrological management to keep the areas drained and and by deeper and more effective drainage. conventional agriculture possible. In coastal peatlands As a result of the oxidation of peat after exposure subsidence increases the risk of floods and salt water to oxygen has been accelerated substantially in this intrusion, which is especially relevant in the light of period. Soil subsidence in organic soils is not restricted climate change induced sea-level rise (Joosten et al. to The Netherland, but it is a world-wide problem. 2015; Hooijer et al. 2012). Soil subsidence may amount to 2 cm annually (in The maintenance of fixed water tables all year round tropical areas) and generates enormous greenhouse leads to an output of surface water through pumping gas (GHG) emissions (20-40 t CO2 per ha and year in winter and in wet periods in summer, as well as an annually) and eventually often loss of productive land input of surface water from the rivers (Rhine, IJssel) (Joosten 2015). Subsidence and erosion of the peat during dry summer periods. The river water flows into soil provoke insufficient drainage, ponding water the polders through inlets in the reservoirs (‘boezems’). following soil compaction and drought prone top As a result of climate change, the occurrence of more soils due to the hydrophobic nature of degraded peat severe droughts in summer is expected to become more (Zeitz & Velty 2002; Wallor & Zeitz 2016; Joosten et frequent, leading to an increase in the demand for fresh al. 2017). These adverse soil conditions raise costs of water from outside. Apart from soil subsidence and water management including a higher water demand water quality deterioration, the availability of fresh for irrigation (Regional Water Authority Hunze en Aa’s water in dry summers is another point of great concern 2018). in discussions on the future management of these In The Netherlands the height of the water table in areas. polders is fixed by the water authorities based on the requirements of agriculture, nature conservation or Challenges for the future after one housing. Individual landowners often install small- millennium of soil subsidence scale pumps for additional water level control. This In the year 1000 AD, the average level of fens and implies that there are various water level sections bogs in the western coastal plain was still more than (‘peilvakken’) within one polder. Decisions on the 2.5 m above mean sea level (see Figure 5). During the height of the water tables are made every 10-15 years, centuries thereafter, initial drainage and reclamation to adapt to subsidence of the ground level and to new operations led to the first areas of any importance on insights in optimal land use. The height of the water drained peat where agricultural crops were grown. The table in polders is fixed by the water authorities based drained soils started to subside, so that the agriculture went further down with the subsiding land. The reclamation became quite on the requirements of agriculture, nature conservation started to become hampered by very wet soils again. widespread since 1550, so that subsidence rates increased, which, together with th or housing. Individual landowners often install seaThe level invention rise, led of tothe a windsituation mill where in the current 15 century peat soilled levels are lower than 2.5 small-scale pumps for additional water level control. mto below new opportunitiessea level. The to expectations drain more areeffectively, that climate so that change will aggravate the situation because it will lead to faster sea level rise but also faster rates of Decisions on the height of the water tables are made subsidence.water levels went further down with the subsiding land. every 10-15 years, to adapt to the progressing subsidence The reclamation became quite widespread since 1550, of the ground level and to new insights in optimal land use. Peat losses are accelerated by ploughing and by addition of lime and fertilizers (which increase peat oxidation) and by wind and water erosion (by which bare peat is blown or washed away). The resulting lowering of the surface necessitates – in case of continued conventional exploitation – a continuous deepening of the drainage ditches. This again enhances peat oxidation, surface lowering, ditch deepening, etc. in what is known as ‘the vicious circle of peatland utilization’. The continuously lowering surface makes gravity drainage increasingly difficult and FigureFigure 5. 5.1000 1000 years years of sea of levelsea level rise and rise soil and subsidence soil subsidence in the Western in Netherlands. the Western Netherlands. necessitates more and more complex and expensive

Greenhouse gas emissions 65 The The peat meadow region is a major source of greenhouse gases (GHG).

Emissions of CO2 amount to 2.26 tons per ha for each cm of soil subsidence, so that they are directly proportional to the depth of the water table below soil surface. In areas which are intensively used for dairy production there is an

additional emission of N2O, whereas wet fen and marsh areas are subject to CH4 (methane) emissions. A recent compilation of emission studies at different scales in the peat meadows of The Netherlands in the framework of the research program ‘Climate Changes Spatial Planning’ has demonstrated that rewetting of previously drained peat meadows has a strongly reducing effect on the GHG

emissions, with reductions in CO2 by far outweighing increases in CH4 emission (Schier-Uijl 2010).

113

so that subsidence rates increased, which, together Effects on agriculture and urban areas with sea level rise, led to a situation where current peat The peat meadow regions are primarily in agricultural soil levels are lower than 2.5 m below sea level. The use. The dominant land use type is dairy production expectations are that climate change will aggravate the with less than 5% maize fields. In the western peat situation because it will lead to faster sea level rise but meadows (provinces of Noord- and Zuid-Holland and also faster rates of subsidence. Utrecht) the farmer incomes are at or just below the national average owing to the optimal spatial allotment. Greenhouse gas emissions There are opportunities for broadening the sources of The The peat meadow region is a major source of income by providing facilities for recreation, health care

greenhouse gases (GHG). Emissions of CO2 amount and nature conservation. In the northern peat meadows to 2.26 tons per ha for each cm of soil subsidence, (provinces of Friesland and Drenthe), the dairy farmers so that they are directly proportional to the depth are among the strongest in The Netherlands and will of the water table below soil surface. In areas which see agriculture as their main economic function in the are intensively used for dairy production there is an future. It is to be expected that the tendency towards

additional emission of N2O, whereas wet fen and larger farms will be continued. The requirements for

marsh areas are subject to CH4 (methane) emissions. fresh water will become imminent at times that the A recent compilation of emission studies at different availability of high-quality fresh water is limiting at the scales in the peat meadows of The Netherlands in the national level. The accelerated soil subsidence will lead framework of the research program ‘Climate Changes to increased differences in elevations within polders, Spatial Planning’ has demonstrated that rewetting which will create major problems for dike maintenance. of previously drained peat meadows has a strongly It will also generate enormous cost for upgrading the reducing effect on the GHG emissions, with reductions infrastructure in urban areas, such as for instance in the

in CO2 by far outweighing increases in CH4 emission city of Gouda, situated in the ‘Green Heart’ of Holland. (Schier-Uijl 2010). This city is well known for its cheese. But one could say Figure 6 shows that total GHG emissions expressed that the success of Gouda cheese has left present and

as CO2 equivalents were lowest in the rewetted polder future generations with enormous unpaid bills. Horstermeer, mostly because restored peat formation fixed carbon rather than releasing it as in the two Effects on nature polders where drainage had continued. Stopping There are two types of nature reserves in the peat

fertilizer use did reduce the N2O emissions (polder meadow area, both of which are strongly controlled Stein in comparison with polder Oukoop), but that only by human management; (i) the peat meadows marginally affected the GHG balance. characteristic bird fauna, and (ii) peat-forming fen ecosystems (Verhoeven et al. 2010). Peat meadow nature reserves are characterized by species-rich grassland communities and dense meadow bird populations. These meadows have high water tables and low fertilizer additions and are not in commercial agricultural use (Van de Riet et al. 2010). Large fractions of the European Godwit and Lapwing populations are breeding in these meadows, which has created a special responsibility according to EU legislation. Shallow ponds and lakes are inhabited by a range of plant communities characterizing the succession from open water to marsh habitat to fen or carr vegetation. In the western peat meadows, the national government has allocated 12,000 ha of (semi-)natural Figure 6. Greenhouse Gas (GHG) emissions in three peat areas to the Ecological Main Structure (EHS). Both Figure 5. Greenhouse Gas (GHG) emissions in three peat meadow polders with different land use meadow polders with different land use and water table types of nature reserves described above are part of and water table management (Schier-Uijl 2010). management (Schier-Uijl 2010). EHS. For thousands of hectares, this objective still

Figure 5 shows that total GHG emissions expressed as CO2 equivalents were lowest in66 the rewetted polder Horstermeer, mostly because restored peat formation fixed carbon rather than releasing it as in the two polders where drainage had continued. Stopping fertilizer use did reduce the N2O emissions (polder Stein in comparison with polder Oukoop), but that only marginally affected the GHG balance.

Effects on agriculture and urban areas

The peat meadow regions are primarily in agricultural use. The dominant land use type is dairy production with less than 5% maize fields. In the western peat meadows (provinces of Noord- and Zuid-Holland and Utrecht) the farmer incomes are at or just below the national average owing to the optimal spatial allotment. There are opportunities for broadening the sources of income by providing facilities for recreation, health care and nature conservation. In the northern peat meadows (provinces of Friesland and Drenthe), the dairy farmers are among the strongest in The Netherlands and will see agriculture as their main economic function in the future. It is to be expected that the tendency towards larger farms will be continued. The requirements for fresh water will become imminent at times that the availability of high-quality fresh water is limiting at the national level. The accelerated soil subsidence will lead to increased differences in elevations within polders, which will create major problems for dike maintenance. It will also generate enormous cost for upgrading the infrastructure in urban areas, such as for instance in the city of Gouda, situated in the ‘Green Heart’ of Holland. This city is well known for its cheese. But

114

The Agenda for the western peat meadow areas proposed to make water management schemes leading for land use decisions (‘function follows water table’), implying that land uses have to be adapted to the water tables arising from a water management scheme that is most sustainable. This is in contrast to current practice, where water management is always adapted to the needs of farmers, nature conservation and settlements. This new policy is still in the discussion and planning phase; it is definitely controversial in some parts of the region. As indicated earlier, the problems identified for these regions (soil subsidence, but also drought, flooding, eutrophication and salinization) are not caused by climate change but they are certainly accelerated and aggravated by it.

In the Netherlands intensive discussions are taking place about the future of the peatland landscape, see graphics below. Currently large peatland areas are subject to drainage-induced subsidence (“bodemdaling”) leading to ever increasing height differences between deeply drained and pumped, intensively used agricultural lands on the one hand (left picture foreground) and low intensity agriculture, settlement and nature conservation sites with much higher water levels on the other hand (left background).

It is clear that continued pumping and subsidence (“loslaten”, left picture) is impossible, but what are the alternatives: continuing conventional land use but raise the water levels somewhat to slow down subsidence (“remmen”, central picture), or stopping subsidence completely by raising the water level to at or over the surface and changing land use towards wet livelihoods, including paludicultures, floating solar energy, and wet tourism (right picture).

Figurehas to be 6. realized Three by differentbuying land scenarios and modifying for thefut ure waterFigure management 7. Three different scenariosof the Dutchfor future peat water polders. Left businessland and water as management.usual, mid: In slow the northern-down peatsoil subsidence,management and ofright the Dutch stop peat soil polders. subsidence Left business by as rigorous rewetting (https://www.provincie-utrecht.nl/onderwerpen/alleusual, mid: slow-down-onderwerpen/bodemdaling/ soil subsidence, and right stop soil meadow areas, the nature areas have been identified subsidence by rigorous rewetting (https://www.provincie- and separated from the farming areas at a higher scale, utrecht.nl/onderwerpen/alle-onderwerpen/bodemdaling/). Iresultingn the inpast large areasdecade, of either stu completedies agriculturalhave been carried out to identify the problems in the peatuse or completemeadow nature areas protection. when In the current western areas, land usethat landand uses water have to managementbe adapted to the water will tables continue inthere a arecontext more transitions of climate between thechange two. (Woestenburgarising from 2009 a water, managementSchier-Uijl scheme et al. that 201is most4 ). At theWater same quality time, issues aremeasures related to eutrophication have been designedsustainable. Thisto isimprove in contrast tothe current situation, practice, which couldand to the decrease leaching of sulfate the fromrate the ofpeat soilmeadows. subsidence where waterand management enhance is wateralways adapted quality to the without These sulfates are formed by oxidation of pyrite, which needs of farmers, nature conservation and settlements. is quite common in most peat soils on the coastal plain. 116This new policy is still in the discussion and planning High sulfate concentrations lead to mobilization of phase; it is definitely controversial in some parts of the phosphates from lake and ditch sediments and lead region. As indicated earlier, the problems identified to loss of mesotrophic plant communities (Lamers et for these regions (soil subsidence, but also drought, al. 1998, Smolders et al. 2006, 2010, Lamers et al. 2015, flooding, eutrophication and salinization) are not Geurts et al.2008). caused by climate change but they are certainly accelerated and aggravated by it. Policies and adaptation strategies In the Netherlands intensive discussions are taking Water The national Dutch planning policy for peat place about the future of the peatland landscape, see meadow areas is reflected in the policy documents graphics below. Currently large peatland areas are ‘Nota Ruimte’ (VROM, 2004) and the ‘Programma voor subject to drainage-induced subsidence (‘bodemdaling’) de westelijke veenweidegebieden’ (VROM, 2009). The leading to ever increasing height differences Nota Ruimte calls for conservation of the peat soil between deeply drained and pumped, intensively and the peat meadow landscape. Vulnerability for soil used agricultural lands on the one hand (left picture subsidence is given as the main criterium for regional foreground) and low intensity agriculture, settlement development. Grassland-based dairy production is and nature conservation sites with much higher water an important economic function which is required to levels on the other hand (left background). conserve the typical peat meadow landscape. However, It is clear that continued pumping and subsidence farmers must respect the ecological and cultural- (‘loslaten’, left picture) is impossible, but what are the historic value of this region. Areas with rapid soil alternatives: continuing conventional land use but raise subsidence or saltwater intrusion require a new water the water levels somewhat to slow down subsidence management with higher water tables and responsible (‘remmen’, central picture), or stopping subsidence authorities should consider these options. The Agenda completely by raising the water level to at or over the for the western peat meadow areas proposed to make surface and changing land use towards wet livelihoods, water management schemes leading for land use including paludicultures, floating solar energy, and wet decisions (‘function follows water table’), implying tourism (right picture).

67 In the past decade, studies have been carried out to been tested for efficacy and cost effectiveness, it is identify the problems in the peat meadow areas when quite complicated to evaluate which combination of current land use and water management will continue measures is optimal in a particular area. in a context of climate change (Woestenburg 2009, Schier-Uijl et al. 2014). At the same time, measures have Paludiculture; an example of been designed to improve the situation, which could harvesting wet crops while fighting decrease the rate of soil subsidence and enhance water soil subsidence quality without harming the most important economic The concept of ‘paludiculture’, in which new (wet) crops functions in the area, i.e. agriculture and recreation, are being cultivated in rigorously rewetted peat polders mostly ‘ecotourism’ linked to nature values. has been developed in Eastern Germany is well known These measures can be subdivided into two major in some parts of Germany (Wichtmann et al. 2016), but categories, i.e. (1) measures affecting thewater system, in the Netherlands it is still in a stage of early scientific mostly by modifications of physical water control experiments and small scale pilot studies (Geurts & structures and of the water regime and (2) measures Fritz 2018). Paludiculture is a form of productive land affectingland use and economic functions. Some major use that allows rewetted peatlands to produce food, examples of such measures are: fiber and energy. Typical crops grown in paludicultures (‘paludicrops’) are perennial crops that thrive on (1) Water system waterlogged or flooded soils, for exampleSphagnum a. Flexible water level regime; increasing water levels (peat moss), Phragmites (reed), Typha (cattail), Salix in ditches; larger units with equal water levels; (willow) and Zizania (wild rice). larger shallow ponds and lakes. These measures can In 2015 the international, transdisciplinary research prevent extremely low water levels in dry summers project CINDERELLA, was started to investigate and slow down subsidence. the productive use of rewetted peatlands and the b. Buffer zones bordering ditches and canals, wetlands simultaneous restoration of ecosystem services, for water quality improvement. including reduction of greenhouse gas (GHG) emissions c. Drainage systems (tubing) below the water table and land subsidence, water and nutrient retention, and (‘under-water drainage’), which deliver water to the water purification. The main objective was to extend central parts of grassland parcels so that water levels the scientific base for a sustainable, productive land use do not become extremely low in summer. This is a of wet peatlands and making alternative uses accessible measure with a claimed subsidence reduction of to farmers and land authorities. 30%, but this is based on limited data and is still We will report here some results from a pilot study in disputed. Zegveld, which we will visit.

(2) Land use and economic functions Zegveld: experimental a. Robust and climate-proof crops; salt-tolerant crops; paludiculture plots with energy crops. high and low water tables b. ‘Green-blue services’, i.e. farmers allow water to be (2015-2018) stored in reservoirs to be used in dry period or to prevent flooding events in villages and cities. Renske Vroom, Jeroen Geurts (Radboud University c. Maintaining biodiversity in the landscape by Nijmegen) funding farmers for measures stimulating meadow birds and botanical diversity. n 2015 experimental paludiculture plots were d. Measures to stimulate recreational use (swimming, Iestablished on the experimental farm of the camping at the farm, hiking). Veenweide Innovatiecentrum (VIC) in Zegveld (Province e. Building residential areas in a robust, ‘water-proof’ of Utrecht). First 5-10 cm of the topsoil of the peat way. meadow were removed to build little dikes around two In practice, Regional Adaptation Strategies will basins of 10x120m. Very young cattail plants (Typha consist of packages of such adaptation and mitigation latifolia) were planted by hand in 8 random plots of 48 measures. Although most of the measures have m². In other plots, reed (Phragmites australis), willow

68

Figure 8. Left: plots drying out in July 2015 (groundwater Figure(Salix alba)7. Left:, and silvergras plots drying (Miscanthus out giganteus) in July were2015 (groundwater level > 30-40 cm below the surface). FigureFigure 7.7. Left:Left: plotsplots dryingdrying outout inin JulyJuly 20152015 (groundwater(groundwaterlevel > 30-40 cm below levellevel the >> surface). 3030--4040 Right: cmcm stable belowbelow high the thewater surface).surface). Right:planted. stable Water high was suppliedwater levelfrom the in adjacentJuly 2016 ditch (20 by cmlevel above in July the 2016 surface). (20 cm above Pictures the surface). by PicturesJeroen by Geurts Right:Right: stable stable high high water water level level in in July July 2016 2016 (20 (20 cm cm above above the the surface). surface). Pictures Pictures by by Jeroen Jeroen Geurts Geurts. .. a pump. In the beginning plots were regularly drying Jeroen Geurts. out in summer to more than 40 cm below the surface InInIn October OctoberOctober 2016 20162016 the thethe water waterwater level levellevel in in inthe thethe northern northernnorthern basin basinbasin was waswas lowered loweredlowered to toto 20 2020 cm cmcm below(Figure the7). In Augustsurface water (‘drought levels fluctuated treatment’) between andat water the levels connection of + 20 cm. In spring pipe 2017, was water closed. levels The below0below and +20 the thecm and surfacesurface from October (‘drought(‘drought onwards, water treatment’)treatment’) levels regularly andand the droppedthe connectionconnection to 30-40 cm below pipepipe surface waswas in closed.theclosed. TheThe reason for choosing lower water levels was the poor performance of Salix (little reasonwerereason definitely forfor raisedchoosingchoosing to 20 cm lowerabovelower the water watersurface. levelslevels ‘drought waswas the treatment’the poorpoor and performance performancetherefore water was of pumpedof SalixSalix in (little(little heightheightheightBecause increase increase increaseof the drying and and periodsand yellow yellowyellow and subsequent leaves) leaves)leaves) and andand the again thethe very in veryvery June poor to poorpoor raise performance the performanceperformance water level above of the ofof Miscanthus surface MiscanthusMiscanthus atatweedat water waterwater development, levels levelslevels only of ofof 20%+ + +20 of 20 20the cm. Typhacm.cm. In plants InIn spring springspring 2017, 2017,2017,for two water waterweeks.water Afterlevels levelslevels that, regularlywater regularlyregularly levels were dropped dropped droppedmaintained to toto 30 3030- -- 4040survived40 cm cmcm below the belowbelow first surface month surfacesurface of2015, in in inwhereasthe thethe ‘drought ‘droughtmore‘drought than treatment’ treatment’treatment’at -20 cm again. and andand therefore thereforetherefore water waterwater was waswas pumped pumpedpumped 75% of the Phragmites and Salix survived. Therefore, all All species except for Salix grew better at higher water inin inagain againagain in in inJune JuneJune to toto raise raiseraise the thethe water waterwater le levellevelvel above aboveabove the thethe surface surfacesurface for forfor two twotwo weeks. weeks.weeks. After AfterAfter that,experimental water paludiculture levels were plots, maintainedexcept for the high at -20levels cm compared again. to the dry conditions in 2015, with that,densitythat, waterplots,water were levelslevels reestablished werewere in maintainedspringmaintained 2016 with atat --2020average cmcm plant again.again. heights of about 160 cm in 2016 (Figure 9). Allnew, species more vital except Typha plants for and Salix Phragmites grew rhizomes, better Survivalat higher rates were water also higher, levels especially compared for Typha and to the AllandAll reusedspeciesspecies Salix andexceptexcept Miscanthus forfor plants SalixSalix from grew grewthe old betterbetterMiscanthus atat higherhigher, although waterwater Miscanthus levelslevels density comparedcompared still decreased to to thethe dry conditions in 2015, with average plant heights of about 160 cm in 2016 dryexperiment.dry conditionsconditions Water levels inin in 2015,20162015, (first withyearwith of replantedaverageaverage andplantplant its biomass heightsheights production ofof wasaboutabout very low. 160160 Plant cm cmdensity inin 20162016 (Figur(Figurplots)(Figur weree ee8 8maintained).8). ).Survival SurvivalSurvival at +20 rates cmratesrates in allwere werebasinswere also(Figure alsoalso higher, higher,higher,of T. angustifolia especially especiallyespecially increased for forfor toTypha moreTyphaTypha than and and50and plants/m Miscanthus MiscanthusMiscanthus2, , ,, althoughalthough8).although Miscanthus MiscanthusMiscanthus density densitydensity still stillstill decreased decreaseddecreasedwhereas and and andT. latifolia its itsits biomass even biomassbiomass produced production production 80-100production new shoots was waswas very veryvery 2 2 low.low.low.In OctoberPlant PlantPlant 2016 density densitydensity the water oflevel ofof T.in T. theT. angustifolia northernangustifoliaangustifolia increased perincreasedincreased m . The number to toto of more Phragmitesmoremore than than shootsthan 50increased 5050 plants/m plants/mplants/m ,2 2,, basin was lowered to 20 cm below the surface (‘drought more than 10 times in the first growing2 season. whereaswhereaswhereas T. T.T. latifolia latifolialatifolia even eveneven produced producedproduced 80 8080-100--100100 new newnew shoots shootsshoots per perper m mm.2 2.The. TheThe number numbernumber of ofof Phragmitestreatment’) and shootsthe connection increased pipe was closed. more The than 10Due times to the inhigher the water first levels, growing weed development season. PhragmitesreasonPhragmites for choosing shootsshoots lower water increasedincreased levels was the moremore poor thanthan was 1010 much timestimes lower in inthan thethe in 2015: firstfirst only growinggrowing 2% cover in season. season.the Dueperformance to the of higher Salix (little water height increase levels, and weed yellow developmentTypha and Salix plots,was 4% much in the Miscanthus lower than plots, in 2015: Dueleaves)Due to andto the the very higherhigher poor performance waterwater levels, levels,of Miscanthus weedweed developmentdevelopmentand 13% in de Phragmites waswas much muchplots. Instead lowerlower of weeds, thanthan inin 2015:2015: onlyonlyonly 2% 2%2% cover covercover in in inthe thethe Typha TyphaTypha and andand Salix SalixSalix plots, plots,plots, 4% 4%4% in in inth ththe eeMiscanthus MiscanthusMiscanthus plots, plots,plots, and andand 13% in de Phragmites plots. Instead of weeds, floating algae beds developed in 13%13% inin dede PhragmitesPhragmites plots.plots. InsteadInstead ofof weeds,weeds, floatingfloating algaealgae bedsbeds developeddeveloped69 inin thethethe plots plotsplots with withwith low lowlow vegetation vegetationvegetation cover covercover ( Miscanthus((MiscanthusMiscanthus and andand Salix SalixSalix).). ).T. T.T. latifolia latifolialatifolia had hadhad the thethe highesthighesthighest winter winterwinter biomass biomassbiomass in in inthe thethe first firstfirst year yearyear (2.8 (2.8(2.8 ton tonton dm/ha), dm/ha),dm/ha), followed followedfollowed by byby Ph PhPhragmitesragmitesragmites

119119119

(1.9 ton dm/ha) and T. angustifolia (1.0 ton dm/ha; Figure 3). Miscanthus biomass was very low (45 kg dm/ha) and therefore not suitable to grow under permanent wet conditions.

Figure 9. Left: Average plant heights1 in September 2015 (dry 1 However, the fodder value and nutrient composition Figure 8.conditions), Left: Average 2016 (water plant level +20 heights cm) and 2017 in (waterSeptember level 2015 (dry conditions), 2016 (water level +20 was different at each harvest time, 2which is important cm) and+20 2017cm and -20(water cm). Right: level Dry biomass+20 yieldcm 2 andin January -20 and cm). Right: Dry biomass yield in January and for choosing the most suitable biomass application. SeptemberSeptember 2017 2017 (water (water levellevel +20 +20 cm and cm -20 and cm). 1no-20 data cm). 1no data for Salix and Miscanthus in 2017; 2no 2 Fodder value and nutrient content was highest in data forfor Miscan Salix andthus Miscanthus in September. in 2017; no data for Miscanthus in September. spring, with a raw protein content of 127 g/kg dm and For that reason only Typha and Phragmites aplots digestion were coefficient monitored (VCOS) of 63 in% (Pijlman the growinget al. floating algae beds developed in the plots with low subm.). At the end of the growing season a large part of season of 2017. T. latifolia biomass was two times higher at high water levels vegetation cover (Miscanthus and Salix). T. latifolia had the nutrients was stored in the rhizomes already and (9.2 tonthe highestdm/ha) winter than biomass at in low the first water year (2.8 levels ton (4.4the ton aboveground dm/ha). biomass Moreover, contained more plants fibers and were on averagedm/ha), followed30 cm by Phragmitestaller and (1.9 ton produced dm/ha) and almostless moisturetwo times (40-60%, more depending flowers on the weather). (19 vs 11 per m2T.). angustifolia Phragmites (1.0 ton dm/ha; produced Figure 3). Miscanthus5.5 ton dm/haTherefore, and biomassdid not from differ a winter between harvest is better the dry and wetbiomass plots. was very lowHowever, (45 kg dm/ha) plants and therefore were not moresuitable than for e.g. isolation20 cm material taller or bedding. under wet suitable to grow under permanent wet conditions. Biomass production in the Phragmites plot was conditions, whereas plant density was almost 20% higher under dry conditions. For that reason only Typha and Phragmites plots were low (2.8 ton dm/ha) and did not differ between the monitored in the growing season of 2017. T. latifolia harvested and unharvested side of the plot. However, Our resultsbiomass wasshow two timesed higherthat atan high earlywater levels harvest plantin Maydensity iswas possiblehigher on the harvestedfor Typha side (165, after which (9.2an ton equivalent dm/ha) than at lowamount water levels of (4.4 tonbiomass dm/ canshoots/m be2 ) thanharvested on the unharvested at the side end(148 shoots/ of the growingha). Moreover,season. plants Multiple were on average harvests 30 cm taller are and alsom 2).possible, The most likely but reason th fore the total low biomass cumulative biomassproduced production almost two timesof 10 more ton flowers dry (19 mattervs 11 per perproduction hectare compared was to not the other dependent Phragmites plots on is the 2 time ofm ). thePhragmites first produced harvest 5.5 ton dm/ha(Figure and did 9 not). Biomassthat these yield plants weredecreased originally grown 30 -from40% seeds when and differ between the dry and wet plots. However, plants not from rhizomes. plants werewere more thanfirst 20 cmharvested taller under wet in conditions, October or January, due to senescence and 2 physicalwhereas dama plantge density after was storms.almost 20% higherThe underaverage dry number of shoots was 73 plants/m in 2016conditions. and did not increase in comparison with 2015. The average plant height was 180Our cm, results which showed is that 50 an cmearly harvesthigher in Maythan is the year before. possible for Typha, after which an equivalent amount However,of biomass the can fodder be harvested value at the end and of the growingnutrient composition was different at each harvestseason. time, Multiple which harvests areis alsoimportant possible, but the for total choosing the most suitable biomass cumulative biomass production of 10 ton dry matter application.per hectare Fodder was not dependent value onand the timenutrient of the first content was highest in spring, with a raw proteinharvest content (Figure 10).of Biomass 127 yieldg/kg decreased dm 30-40%and a digestion coefficient (VCOS) of 63 % (Pijlmanwhen et plants al .were subm.). first harvested At inthe October end or January,of the growing season a large part of the nutrientsdue to wassenescence stored and physical in thedamage rhizomes after storms. already and the aboveground biomass The average number of shoots was 73 plants/m2 in 2016 contained more fibers and less moisture (40FigureFigure- 60%,9. 10.Biomass Biomass depending yield yield in incase case of of multiple onmultiple the harvests harvests weather). ofof T.T. latifolia (dry weight). First harvest in the and did not increase in comparison with 2015. The respectivelatifolia (drymonths. weight). Second First harvestharvest in theOctober respective of the months. same year. For September 2017, yield of the Therefore, biomass from a winter harvestharvested is better side of the suitable plot is shown. for e.g. isolation average plant height was 180 cm, which is 50 cm higher Second harvest in October of the same year. For September material or bedding. than the year before. Biomass2017, yield production of the harvested in the side Phragmites of the plot is shown.plot was low (2.8 ton dm/ha) and did not differ between the harvested and unharvested side of the plot. However, plant 2 70 density was higher on the harvested side (165 shoots/m ) than on the 2 120 unharvested side (148 shoots/m ). The most likely reason for the low biomass production compared to the other Phragmites plots is that these plants were originally grown from seeds and not from rhizomes.

Zegveld: establishment of 0.4 ha cattail (Typha latifolia) for fodder

Monique Bestman, Jeroen Pijlman (Louis Bolk Institute)Jeroen Geurts (Radboud University Nijmegen).

In July 2016, a 0.4 ha parcel (ca. 25x160 m) of cattail (Typha latifolia) was established in Zegveld. The purpose of this parcel is to grow enough biomass for feeding experiments with dairy cows, and to get more experience with weed control, fertilization, harvesting moment versus biomass production and nutritional values, and harvesting techniques on larger parcels. Young Typha plants were obtained from a professional grower. In July at planting, plants were 30-60 cm high. Right after removing the top soil layer, the surface was dry, which made it possible to use a small tractor for pulling an adapted vegetable planting machine followed by a light trailer (Figure 10). During the season a water level of on average 20 cm was realized using a solar-powered water pump (Solar Portmann) equipped with a water level sensor.

The plants grew well: in September (only 2 months after planting) 9-16 new shoots per m² were counted with plant heights up to 80 cm. At 8 and 30 November the amount of grown biomass at ±5-10 cm above water level was estimated at 371 kg ha-1 on both moments. The plants were not fertilized nor weed control was necessary; the water level and clean soil at establishment created good conditions against possible weeds.

121

Zegveld: establishment of 0.4 ha cattail (Typha latifolia) for fodder

Monique Bestman, Jeroen Pijlman (Louis Bolk Institute)Jeroen Geurts (Radboud University Nijmegen).

n July 2016, a 0.4 ha parcel (ca. 25x160 m) of cattail I(Typha latifolia) was established in Zegveld. The purpose of this parcel is to grow enough biomass for Figure 10. Planting the Typha plants and view on the parcel on the 2nd of September 2016 feeding experiments with dairy cows, and to get more (Pictures: Monique Bestman). experience with weed control, fertilization, harvesting moment versus biomass production and nutritional At the 18th of January 2017, during a period of frost, all Typha plants were mown values, and harvesting techniques on larger parcels. at ±5-10 cm above water level with a brush cutter, while walking over the ice. Young Typha plants were obtained from a professional The cut biomass was removed, but biomass yields were not determined. In May grower. In July at planting, plants were 30-60 cm high. Right after removing the top soil layer, the surface 2017, the field was fertilized with 150 kg N and 150 kg K in the form of coated was dry, which made it possible to use a small tractor urea and potassium nitrate, and in July and August the field was fertilized three for pulling an adapted vegetable planting machine times with each time 20 kg N in form of coated urea. followed by a light trailer (Figure 11). During the season Figure 11. Planting the Typha plants and view on the parcel Figure 10. Planting the Typha plants and viewon theon 2nd the of September parcel 2016on (Pictures:the 2nd Monique of September Bestman). 2016 a water level of on average 20 cm was realized using a (Pictures: Monique Bestman). solar-powered water pump (Solar Portmann) equipped cut biomass was removed, but biomass yields were not with a water level sensor. determined. In May 2017, the field was fertilized with At the 18th of January 2017, during a period of frost, all Typha plants were mown The plants grew well: in September (only 2 months 150 kg N and 150 kg K in the form of coated urea and afterat planting)±5-10 9-16 cm new above shoots perwater m² were level counted with apotassium brush nitrate,cutter, and whilein July and walking August the over field wasthe ice. withThe plant cut heights biomass up to 80 wascm. At removed,8 and 30 November but biomassfertilized threeyields times were with each not time determined. 20 kg N in form ofIn May the2017, amount the of grown field biomass was at fertilized ±5-10 cm above with 150 coatedkg N urea. and 150 kg K in the form of coated waterurea level and was estimatedpotassium at 371 kgnitrate, ha-1 on both and in July andIn June August 2017, the Typhathe plantsfield flowered was fertilized and their three Figure 10. Planting the Typha plants and view on the parcel on the 2nd of September 2016 moments. The plants were not fertilized nor weed pollen were harvested by a company that grows (Pictures:times withMonique each Bestman). time 20 kg N in form of coated urea. control was necessary; the water level and clean soil at predatory mites for application in greenhouses. After Atestablishment the 18th ofcreated January good conditions 2017, againstduring possible a period that, of Typha frost, was allharvested Typha twice plants (June and were September) mow n atweeds. ±5-10 cm above water level with a brushto make cutter, silage for while feed experiments. walking Atover 23 June, the the ice. At the 18th of January 2017, during a period of frost, biomass yield was 6.81 ton dry matter ha-1. At 19 The cut biomass was removed, but biomass yields were not determined. In May all Typha plants were mown at ±5-10 cm above water September, biomass yield was only 1.94 ton dry matter 2017, the field was fertilized with 150 kg N and 150 kg K in the form of coated level with a brush cutter, while walking over the ice. The ha-1 (Figure 12). Therefore the total yield in 2017 was 8.75 urea and potassium nitrate, and in July andton Augustdry matter ha-the1. Afterfield the was first fertilizedharvest in June, three it was timesFigure 12. with Typha each harvest timeusing a 20two wheeledkg N inreaper-binder form of coatedobserved urea. that regrowth was mainly from new shoots. at the 19th of September 2017 (Picture: Monique Bestman). Further information/further reading

Figure 11. Typha harvest using a two wheeled reaper-binder at the 19th of September 2017 (Picture: Monique Bestman).

In June 2017, the Typha plants flowered and their pollen were harvested by a company that grows predatory mites for application in greenhouses. After that, Typha was harvested twice (June and September) to make silage for feed experiments. At 23 June, the biomass yield was 6.81 ton dry matter ha-1. At 19 September, biomass yield was only 1.94 ton dry matter ha-1 (Figure 11).

Figure 11. Typha harvest using a two wheeled reaper-binder at the 19th of September71 2017 (Picture: Monique Bestman). 122

In June 2017, the Typha plants flowered and their pollen were harvested by a company that grows predatory mites for application in greenhouses. After that, Typha was harvested twice (June and September) to make silage for feed -1 Figureexperiments 11. Typha. harvestAt 23 June,using a thetwo biomasswheeled reaper yield- binderwas 6.81at the ton 19th dry of Septembermatter ha 2017. At 19 -1 (Picture:September, Monique Bestman).biomass yield was only 1.94 ton dry matter ha (Figure 11).

In June 2017, the Typha plants flowered and their pollen were harvested by a company that grows predatory mites for application in greenhouses. After that, Typha was harvested twice (June and September)122 to make silage for feed -1 experiments. At 23 June, the biomass yield was 6.81 ton dry matter ha . At 19 September, biomass yield was only 1.94 ton dry matter ha-1 (Figure 11).

122

Arciniegas G, & Janssen R, (2012). Spatial decision support for Regional Water Authority Hunze en Aa’s (2018). collaborative land use planning workshops. Landscape Urban https://www.hunzeenaas.nl/about/voldoendewater/Paginas/ Planing. Droogte.aspx http://dx.doi.org/10.1016/j.landurbplan.2012.06.004 Schier-Uijl AP (2010). Flushing meadows. The influence of Arciniegas G, Janssen R, & Omtzigt N, (2011). Map-based management alternatives on the greenhouse gas balance of fen multicriteria analysis to support interactive land use allocation. meadow areas. PhD thesis Wageningen University. International Journal of Geographical Information Science, pp. 1-17. Schrier-Uijl, A. P., Kroon, P. S., Hendriks, D. M. D., Hensen, A., Van Huissteden, J., Berendse, F. & Veenendaal, E. M. Arciniegas GA, Janssen R, & Rietveld P, (2012). Effectiveness of (2014). Agricultural peatlands: towards a greenhouse gas sink-a collaborative decision support tools: Results of an experiment. synthesis of a Dutch landscape study. Biogeosciences, 11(16), Environmental modelling & Software, doi:10.1016/j. 4559 envsoft.2012.02.021 Smolders AJP, Lamers LPM, Lucassen ECHET, Van der Velde G Geurts JJM, Smolders AJP, Verhoeven JTA, Roelofs JGM. & & Roelofs JGM (2006). Internal eutrophication: ‘How it works Lamers LPM (2008). Sediment Fe:PO4 ratio as a diagnostic and and what to do about it’, a review. Chemistry and Ecology 22: prognostic tool for the restoration of macrophyte biodiversity in 93-111. fen waters. Freshwater Biology 53: 2101-2116. Smolders AJP, Lucassen ECHET, Bobbink R, Roelofs JGM & Geurts JJM & Fritz C (eds.) (2018). Paludiculture pilots and Lamers LPM (2010). How nitrate leaching from agricultural experiments with focus on cattail and reed in the Netherlands. lands provokes phosphate eutrophication in groundwater fed Technical report Radboud University Nijmegen, Institute for wetlands: the sulphur bridge. Biogeochemistry 98: 1-7. Water & Wetland Research (IWWR). Van de Riet BP, BarendregtA, Brouns K, Hefting MM, & Hooijer, A., Page, S., Jauhiainen, J., Lee, W.A., Lu, X.X., Idris, Verhoeven JTA (2010). Nutrient limitation in species-rich A., Anshari, G. (2012). Subsidence and carbon loss in drained Calthion grasslands in relation to opportunities for restoration tropical peatlands. Biogeosciences, 9, 1053–1071. in a peat meadow landscape. Applied Vegetation Science 13: 315-325. Janssen R, Verhoeven JTA, Arciniegas G, & Van de Riet B. (accepted), ‘Spatial evaluation of ecological qualities to Van den Akker JJH, Jansen PC, Querner EP (2011). De huidige support interactive design of land use plans’, Environment and en toekomstige watervraag van veengronden in het Groene Planning B: planning and design. Hart: een verkenning van het effect van onderwaterdrains. Alterra-rapport 2142. Joosten, H. (2015). Peatlands, climate change mitigation and biodiversity conservation: An issue brief on the importance of Verhoeven JTA, Barendregt A Van de Riet BP (2010). peatlands for carbon and biodiversity conservation and the role Possibilities for nature dvelopment in drained peat polders (in of drained peatlands as greenhouse gas emission hotspots. Dutch, with English summary. Landschap 27 (3): 157-166. Nordic Council of Ministers. Vermaat JE, Harmsen J, Hellmann FA, Van der Geest HG, De Joosten, H., Gaudig, G., Tanneberger, F., Wichmann S. & Klein JM, Kosten S, Smolders AJP, Verhoeven JTA, Mes Wichtmann, W. (2016). Paludiculture: sustainable productive RG & Ouboter M. (2016). Annual sulfate budgets for Dutch use of wet and rewetted peatlands. In: Bonn, A., Allott, T., lowland peat polders: The soil is a major source through peat Evans, M., Joosten, H., Stoneman, R. (Eds), Peatland restoration and pyrite oxidation. Journal of Hydrology 533: 512-522. and ecosystem services: Science, policy, and practice, Cambridge University Press, Cambridge, 339-357. VROM (2004). Nota Ruimte. Ministerie voor Ruimtelijke Ordening en Milieubeheer. Staatscourant 2006. Joosten, H., Tanneberger, F. & Moen, A. (2017). Mires and peatlands of Europe. Status, distribution and conservation. 780 VROM (2009). Programma westelijke veenweidegebieden. pages. Schweizerbart Science Publishers, Stuttgart. Ministerie voor Ruimtelijke Ordening en Milieubeheer.

Kosten S (2011). Een frisse blik op warmer water; over de invloed Wallor, E. & Zeitz, J. (2016). How properties of differently van klimaatverandering op de aquatische ecologie en hoe je de cultivated fen soils affect grassland productivity—A broad negatieve effecten kunt tegengaan, STOWA -rapportnummer investigation of environmental interactions in Northeast 2011-20, ISBN 978.90.5773.524.0. Germany. Catena, 147, 288-299.

Lamers LPM, Tomassen HBM & Roelofs JGM (1998). Sulfate- Wichtmann, W., C. Schröder & H. Joosten (2016). Paludiculture induced eutrophication and phytotoxicity in freshwater – productive use of wet peatlands. Schweizerbart Science wetlands. Environmental Science and Technology 32: 199-205. Publishers, Stuttgart.

Lamers LPM, Vile MA, Grootjans AP, Acreman MC, Van Woestenberg M (2009). Waarheen met het Veen? Kennis voor Diggelen R. Evans MG, Richardson CJ, Rochfort L, keuzes in het westelijk veenweidegebied. Alterra, LEI, CLM, Kooijman AM, Roelofs JGM & Smolders AJP (2015). Royal Haskoning, Universiteit Utrecht, Vrije Universiteit. Ecological restoration of rich fens in Europe and North America: from trial and error to an evidence-based approach. Biological Zeitz, J. & Velty, S. (2002). Soil properties of drained and rewetted Reviews 90: 182–203. fen soils. Journal of Plant Nutrition and Soil Science, 165(5), 618-626.

72