Research Proposal Visser

Investigating the consequences of a large sand suppletion on the freshening of a nature reserve under different management strategies and climatic conditions Assessing the robustness of management strategies using the SEAWAT computer model

MSc thesis proposal Course: ESS-80836 (36 ECTS) Student: RW (Ruben) Visser – 930402894070 First supervisor: dr.ir. EJJ (Erik) van Slobbe Second supervisor: prof.dr.ir. SEATM (Sjoerd) van der Zee Third supervisor: ing. G (George) Bier September 25th, 2015 Earth System Science Group & Soil Physics and Management Group

Land

Introduction

important rest and feeding place for many types of birds (Noord-Holland, 2015).

Whereas many coastal regions around the world need to deal with salt water intrusion and a higher salt concentration of the groundwater, this research concerns a situation where, as a result of carrying out a climate change adaptation option, the groundwater is expected to become less saline. The Hondsbossche Zeewering, a sea dike that defends the Dutch coast near Petten (North-Holland), constructed in the 1870s and strengthened during the 50s and 70s of the previous century, had become a weak chain (“zwakke schakel”) in the Dutch coast defense system (Smit, van Duin, Henkens, & Slim, 2005). To strengthen it, this dike was not reinforced in a conventional way, but instead a building with nature (BwN) option was carried out: 30 million cubic meters of sand have been supplied at the sea-side of the dike, resulting in a new dune and beach area of an approximate width of 250m stretching along the length of the dike (HHNK, 2015c). This sand suppletion combines coastal protection with providing opportunities for nature development and recreational activities (HHNK, 2015b).

PROBLEM

Due to the precipitation surplus in the Netherlands, fresh water bodies are naturally formed in dune areas. Since fresh water has a lower density that saline seawater, the accumulated fresh water rests upon the heavier underlying salt water. The resulting fresh water bodies can extend to a depth of over 100 meters (Post, 2004). Right next to the dike, the nature reserve De Putten can be found in the polder. This nature reserve is part of the recently assigned Natura 2000 reserve Abtskolk & De Putten (Verburg, 2009). De Putten, itself a result of digging activities that took place in the 1950s, consists for a large part of open surface water (Smit et al., 2005). The presence of migratory dwarf geese was the main reason to appoint this area a Natura 2000 area (Verburg, 2009). Until now, salt seawater has been flowing underneath the dike, whereas fresh water is flowing into the polder from the dunes that are present at the southern border of the nature reserve (Smit et al., 2005). The transition from salt to fresh water gives rise to unique vegetation types to grow in this area. However, it is expected that the development of a fresh water body below the newly constructed dune area will influence the groundwater flow in the region, resulting in a gradual freshening of the water flowing underneath the dike into the De Putten (HHNK, 2015a; Smit et al., 2005). This might have consequences for the vegetation types that can grow in this area and this in turn might change the suitability of the nature reserve as an

Ruben Visser

The problem that will be dealt with in this research is the uncertainty regarding the influence of the newly constructed dunes on the water characteristics in the nature reserve De Putten. Part of this problem is the lack of knowledge about how this dune area can be managed in the coming decades and what the consequences of different management strategies on the groundwater flow will be. Another factor contributing to the problem is the uncertainty regarding the consequences of climate change and sea level rise. This local problem is imbedded in an overall lack of knowledge concerning the consequences of carrying out large scale sand suppletions on the regional groundwater characteristics.

AIM The overall aim of my research therefore is to describe a number of different strategies that could be used for the management of the dune area and subsequently investigate the consequences of these strategies on the dynamics of the fresh water body and the groundwater flow under different climate change and sea level rise scenarios. In this way I hope to be able to show which management strategies are most robust. It is my purpose to model the effects of the different management strategies on the groundwater flow from this year up to the year 2100, since this is also the time range exploited in the climate change scenarios I will be using.

RELEVANCE The first and foremost reason that the research I am proposing to conduct is relevant relates to the local circumstances in the study are: there is a general concern that the constructing of the dune area will result in a gradual freshening of De Putten (Smit et al., 2005). The results of one modelling study indicate that freshening may occur to a distance of 350 meters away from the dunes (HHNK, 2015a), but the rate at which this freshening will occur under different climatic conditions has not been quantified yet. As a consequence, it has also not been attempted so far to indicate which management strategies would perform well under future climatic conditions in limiting the fresh water fluxes towards De Putten. Knowledge on this subject might help preserve the valuable nature of De Putten. Preservation of nature reserves has social relevance, since nature areas provide ecosystem services, which positively contribute to human well-

Research Proposal ESS-80836

P a g e |1

being (Fisher et al., 2008; Kettunen & ten Brink, 2013). From a social perspective it is also worth investigating the possibilities of extracting some water from the fresh water body, especially so since it is expected that the availability of fresh water in the Netherlands might decrease as a result of salinization (Bresser et al., 2005; Oude Essink, van Baaren, & de Louw, 2010; Prinsen, Sperna Weiland, & Ruijgh, 2015). Although this research is focused on a specific area, the resulting insights may also be valuable for similar situations around the world. The effects of large sand suppletions on regional groundwater flow and the consequences for the land-use possibilities of nearbylying land are largely unknown, while carrying out such large scale suppletions of sand is gaining popularity around the world. The need to understand the consequences they might have on the availability of fresh water is of world-wide relevance, since the IPCC has high confidence that the problem of salinization will be aggravating in this century as a result of sea-level rise and climate change (IPCC, 2014). This research might make a modest contribution to increasing our insights regarding this subject.

THEORY For my research, I will draw on theories from different disciplines, including geohydrology, climate change adaptation, modelling. Some relevant background on these disciplines will be described in this section. In the study area, different geological formations can be found that originate from fluvial-, fluvioglacial, aeolian and marine deposits during the Pleistocene (Mulder, Steenbergen, & Werff, 1995). During the Holocene, thanks to warmer conditions, peat could develop, while also sand and clay were deposited (Mulder et al., 1995). At a depth of about 250 m, the Maassluis complex can be found (TNO, 2015). This formation can be regarded as the hydrogeologic base, which means that it can be assumed that no significant fluxes of water are going through this formation. For groundwater modelling, of all these geological formations, especially the conductivity (i.e. the ease with which water can flow through) is of great importance (TNO, 2015).

developed. The scenarios that will be used in this research are the KNM’14 scenarios that have recently been developed by the Dutch Meteorological Institute (KNMI), “to provide a scientific set of plausible, consistent and relevant future climate conditions, to be used as a reference framework for a multitude of society impact assessments” (Attema et al., 2014). These “society impact assessments” refer to another field of study relevant for my research, namely climate change adaptation. The conviction has grown that it is not sufficient to take actions that will mitigate climate change; it is necessary that we also try to adapt to changes that are already occurring and to effects that cannot be avoided anymore (IPCC, 2014). Research into climate change adaptation options has become a scientific field of itself and many different approaches have been developed that can guide the development and implementation of adaptation options. Uncertainty about future developments (of natural as well as of social, political and economic conditions) is an integral part of adapting to climate change. A clear distinction in dealing with this uncertainty is visible between those approaches that are based on a sound scientific understanding and detailed predictions of the future (often called “top-down approaches”) and those approaches that start with assessing the current vulnerability of a certain region or (eco)system and next assess the robustness of different strategies that can be followed to address the current risks (often called “bottom-up” approaches). Van Pelt and Swart (2011) call the first approach the “predict-then-act approach” and the latter the “assess-risk-of-policy framework”. The assess-risk-of-policy framework, see Figure 1, contributes to climate proofing, i.e. making sure that plans and investments are sustainable under a range of different future climatic conditions (Commission, 2007). This framework, being a risk-oriented approach, is assumed to be better at dealing with uncertainties than an approach in which predictions determine which actions to take (Ludwig, van Slobbe, & Cofino, 2014). Investigating the robustness of preferred strategies is an approach that is also followed by the Dutch Delta Committee (Deltacommissaris, 2015).

Since my research concerns future developments of a nature reserve, the effects of climate change have to be incorporated. Some changes in the climate have already been observed, others are expected to occur during this century and beyond (IPCC, 2014). To describe possible pathways along which climate change may occur in the future, scenarios have been Ruben Visser

Research Proposal ESS-80836

P a g e |2

incoming fluxes (often restricted to precipitation) over outgoing fluxes (mainly evapotranspiration). The recharge of aquifers however does not only consist of the vertical flux of water seeping into the aquifer, but also of fluxes of water going from one aquifer into an adjacent (under- or overlying) aquifer (Green et al., 2011).

EXPERIMENTAL DESIGN Here, I will introduce my research questions and the setup of the different phases of the research to be conducted.

RESEARCH QUESTIONS FIGURE 1 ASSESS-RISK-OF-POLICY FRAMEWORK - ADAPTED FROM VAN PELT AND SWART (2011)

When thinking about managing water resources, not only must natural conditions and changes therein be taken into account, but also (changes in) socioeconomic conditions. This is especially acknowledged in the Integrated Water Resource Management (IWRM) approach to water issues, which has “comprehensive natural resource planning as founding principle” (Ludwig et al., 2014). Water resources planners should aim to use scientific knowledge in order to integrate competing claims for fresh water, while at the same time political and framing issues must be acknowledged as well (Ludwig et al., 2014). This implies that it is of great importance for water managers to investigate projections of future demands for fresh water resources and for arable land, as well as stakeholder perceptions and policy objectives. In order to understand the remaining part of this proposal and the whole of my thesis report, some main concepts need to be understood. One central term in this research is “variable-density groundwater flow”, which is used to highlight the influences of density differences on groundwater flow (Jorgensen, 1902). In coastal areas, the groundwater flow is influenced by the presence of different water bodies which differ mainly with regards to their chemical composition: seawater contains more salt than fresh water and as result has a higher density (Pauw, 2015). This in turn results in the formation of freshwater bodies on top of the underlying salt groundwater. The size of these freshwater bodies is influenced by the height of the groundwater table, and the speed at which they develop depends on the recharge rate. The term “recharge” is used in hydrology to describe the amount of water entering a certain area or volume. In many cases, it is equivalent to precipitation excess, i.e. the surplus amount of Ruben Visser

Based on the description of the problem and the theoretical approach, I have formulated the following research question: How effective are different strategies in preserving the current status of nature reserve De Putten under changing climatic- and groundwater flow conditions? In order to be able to answer this main question, several sub-questions must be answered first: -

-

-

-

How well can the present groundwater characteristics and flow patterns be represented in the model to be constructed? Which management strategies could be exploited to preserve the status quo of De Putten? How do these strategies influence the volume of the fresh water body and the regional groundwater flow? How are the groundwater flow dynamics influenced by climate- and sea level changes? How do the water characteristics in nature reserve De Putten change under different scenarios and strategies?

METHODS In this research, I will take a risk-oriented approach to test the robustness of different strategies under different scenarios. To follow this risk-oriented approach, the “assess risk of policy”-framework as described by Van Pelt and Swart (2011) will be used (see figure 1). The first phase of my research will thus consist of clearly structuring the problem. This entails investigation of the KNMI’14 climate change scenarios and related uncertainties, current and (predictions of) future land use and water demands in the neighboring area, and relevant policy objectives. Also the uncertainties related to differences between measurements and model approximations must be

Research Proposal ESS-80836

P a g e |3

identified in this phase of the research, so that they can be incorporated in later phases. This is already a time-consuming occupation, since it includes constructing the model, initializing it and comparing it with measured and previously modelled chloride concentrations. The modelling will be done in three steps, using Groundwater Modelling Systems (GMS). During the first step, only the groundwater flow patterns with a uniform density distribution will be modelled using MODFLOW (Harbaugh, 2005). The next step consists of adding information about chloride concentrations using MT3D. In the last step, the SEAWAT model (Langevin C.D., 2008) is used to iteratively combine the first two steps, thus using the chloride distributions to calculate the density-driven water fluxes, which in turn influence the chloride distributions, and so on. The first two steps should provide information on uncertainties between the model and measurements, whereas the last step should already give some indications on the fresh water fluxes that can be expected. This first phase of the research is clearly the most laborious. In the second phase, different management strategies that could limit the fresh water seepage rates will be identified. This phase consists of reviewing relevant literature. Next, the consequences of these strategies will be assessed over a range of possible future changes in climate and mean sea level. In this third phase, the modelling work is continued to calculate the changes in the groundwater flow under different strategies and under changing climatic conditions. The final phase in the “assess risk of policy”framework will consist of summarizing the findings concerning the robustness of the strategies.

DATA REQUIREMENTS To be able to make accurate projections of future groundwater dynamics, a model needs to be constructed that resembles the real situation as close as possible. Information on the topography of the study area must be obtained, since local height differences and the exact shape of the constructed dune area (especially its height) might have a large impact on the resulting groundwater flow dynamics. Also, detailed information about the geohydrology of the study area is crucial, so that the different layers in the model represent the different geologic formations accurately. Therefore, it is necessary to investigate the results of field measurements and modelling work, in order to obtain knowledge about the different formations and their hydraulic conductivity values. The needed hydrogeological information will be obtained from the DINOloket of TNO, Geologische Dienst Ruben Visser

Nederland (Geologic Service Netherlands) (TNO, 2015), and from published literature on this topic, e.g. Mulder et al. (1995). Needed as well is information about the initial chloride concentrations, which will be used as a proxy for salinity. This data is less abundantly available, but the results of some measurements can also be obtained from the DINOloket (TNO, 2015). The results of geophysical measurements (electromagnetic waves) as presented by Velstra (2013) could provide for an additional information source on the regional distribution of salt and fresh groundwater. Also, information about future precipitation and evapotranspiration values under different climate change scenarios will be needed as input into the model.

DATA ANALYSIS AND PRESENTATION Two distinct types of data will be collected in this research. The first type of data concerns social, political and ecological information that is relevant concerning the dune area and De Putten. This type of information will be collected from literature and will probably encompass Dutch and European water and nature policies, information about land and water use in the nearby region, etc. Also to be obtained from published documents is information about climate change scenarios and related uncertainties. The second type of data that will be collected will be the model-output. Direct model output is presented as states. In this case the relevant states would be the height of the water table and the salinity. With this state-data, I hope to obtain information on the changes over time of the following variables: -

The volume of the fresh water body The groundwater flow directions and rates The seepage rates of fresh water in De Putten

Initially, differences may be found to occur between measurements and the modelled reference situation. These differences give rise to uncertainties concerning the model projections of future developments. This uncertainty-measure needs to be incorporated; only changes and trends that extend beyond them can be called “significant”. Although most of my research time will be invested in the modelling work, the amount of data that can be collected might be rather small, since setting up a well-functioning model, following the three steps described above, takes time. However, I expect to be able in the end to present graphs showing e.g. the volume of the fresh water body, the fresh water seepage rates and the chloride concentrations of the surface waters in De Putten. Also, 2- or 3D figures will be presented, showing the model setup and, e.g.,

Research Proposal ESS-80836

P a g e |4

lines connecting points with the same chloride

concentration.

PROJECT PLANNING

FIGURE 2 PROJECT PLANNING

The planning of my project is graphically presented in figure 2. This planning will only be deviated from when this is inevitable. Some phases of the project however, especially both colloquia, the discussion of the draft report and the examination, will need to take place on a specific date which has not yet been agreed upon. Therefore, these might take place in a week prior or after the week that is indicated. Writing the thesis report is not only limited to four weeks in February, rather, parts of the report will be written during the whole period that research is carried out, since the time spent on waiting for model results is thus spent more efficiently. I have a specific aim of finishing that part of the report that contains background information near the end of November.

PROBLEMS AND HOW TO DEAL WITH THEM There are several possible problems I might encounter during my research. Firstly it might occur that the first phase of the research takes too much time, leaving less time than desirable for the other phases. This can be anticipated by working simultaneously on the first and the second phases. This will be possible, since the time that it takes for the model to run and produce results can be spent on conducting the literature review that makes up the second phase. Ruben Visser

Another problem might arise as a result of my aim to model the variable-density groundwater flow from now until 2100. If the situation occurs that there have not been any observable changes in groundwater flow and/or seepage rates at the end of this period, it might be desirable to run the model for a longer time in order to investigate the changes in groundwater flow until 2150 or 2200. This will yield some problems, since the climate change scenarios I will be using only provide information about changes in the climate until 2100 (Attema et al., 2014). I would then propose to extrapolate the changes to the new end-date. As a result of differences that are bound to occur between the measurements that have been done in the area and the modelled current situation, a certain degree of uncertainty must be incorporated in the modelling results. Consequently, the differences between the scenarios might be obscured to some degree by this uncertainty. This problem can only be minimized by using the most accurate data that is available. High quality data concerning the geohydrology of the study area is available from the Dutch Geological Survey as well as from other sources (Mulder et al., 1995). A lack of data might occur concerning the actual topography of the study area. Especially relevant is the height of the dunes. When no data file containing this information can be

Research Proposal ESS-80836

P a g e |5

found, the height will be approximated based on literature. For the second step in the modelling, chloride concentration values are needed. In case either the quantity or quality of these cannot be found, I will approximate them by relying on previously conducted research on the groundwater characteristics in the region. If this would lead to a too coarse and therefore inappropriate study of the real situation, this research could still be valuable for reasons of showing the relationships between management strategies and climate change scenarios on changes in groundwater flow patterns in coastal zones.

LITERATURE REFERENCES Attema, J., Bakker, A., Beersma, J., Bessembinder, J., Boers, R., Brandsma, T., . . . Haarsma, R. (2014). KNMI’14: Climate Change scenarios for the 21st Century–A Netherlands perspective. In B. van den Hurk, P. Siegmund, & A. K. Tank (Eds.). De Bilt, Netherlands. Bresser, A. H. M., Berk, M. M., Born, G. J. v. d., Bree, L. v., Gaalen, F. W. v., Ligtvoet, W., . . . Veraart, J. A. (2005). The effects of climate change in the Netherlands (pp. 112). Bilthoven: MNP. Commission, E. (2007). Adapting to climate change in Europe–options for EU action. Green Paper, from the Commission to the Council, the European Parliament, the European Economic and Social Committee and the Committee of the Regions, Brussels, 29, 2007. Deltacommissaris, S. (2015). Deltaprogramma 2016 - Werk aan de delta. Fisher, B., Turner, K., Zylstra, M., Brouwer, R., Groot, R. d., Farber, S., . . . Harlow, J. (2008). Ecosystem services and economic theory: integration for policy-relevant research. Ecological Applications, 18(8), 2050-2067. Green, T. R., Taniguchi, M., Kooi, H., Gurdak, J. J., Allen, D. M., Hiscock, K. M., . . . Aureli, A. (2011). Beneath the surface of global change: Impacts of climate change on groundwater. Journal of Hydrology, 405(3), 532-560. Harbaugh, A. W. (2005). MODFLOW-2005, the US Geological Survey modular ground-water model: The ground-water flow process: US Department of the Interior, US Geological Survey Reston, VA, USA. HHNK. (2015a). Oog voor de natuur. Retrieved 25-9-2015, 2015, from http://www.kustopkracht.nl/ HHNK. (2015b). Over Kust op Kracht. Retrieved 25-92015, 2015, from http://www.kustopkracht.nl/ HHNK. (2015c). Veilige Kust van Zand. Retrieved 25-92015, 2015, from http://www.kustopkracht.nl/ IPCC. (2014). Climate change 2014: impacts, adaptation, and vulnerability (C. B. Field, V.R. Barros, D.J. Dokken, K.J. Mach, M.D. Mastrandrea, T.E. Bilir,, K. L. E. M. Chatterjee, Y.O. Estrada, R.C. Genova, B. Girma, E.S. Kissel, A.N. Levy,, & P. R. M. S. MacCracken, and L.L. White Eds. Vol. 1). Ruben Visser

Cambridge, United Kingdom and New York, NY, USA: Cambridge University Press. Jorgensen, D. G. (1902). US Geological Survey Professional Paper: US Government Printing Office. Kettunen, M., & ten Brink, P. (2013). Social and Economic Benefits of Protected Areas: An Assessment Guide: Routledge. Langevin C.D., T., D.T., Dausman, A.M., Sukop, M.C., Guo, W. (2008). . Ludwig, F., van Slobbe, E., & Cofino, W. (2014). Climate change adaptation and Integrated Water Resource Management in the water sector. Journal of Hydrology, 518, 235-242. doi: 10.1016/j.jhydrol.2013.08.010 Mulder, J., Steenbergen, v. T., & Werff, v. d. M. (1995). De bodemgesteldheid van het herinrichtingsgebied Bergen-Schoorl; resultaten van een bodemgeografisch onderzoek en de geschiktheidsbeoordeling voor de bloembollenteelt en weidebouw. (pp. 118). Wageningen. Noord-Holland, P. (2015). Beheerplan Natura 2000 Abtskolk & De Putten. Haarlem. Oude Essink, G. H. P., van Baaren, E. S., & de Louw, P. G. B. C. W. F. (2010). Effects of climate change on coastal groundwater systems: A modeling study in the Netherlands. Water Resources Research, 46(10), n/a-n/a. doi: 10.1029/2009wr008719 Pauw, P. S. (2015). Field and Model Investigations of Freshwater Lenses in Coastal Aquifers.pdf>. Post, V. E. A. (2004). Groundwater salinization processes in the coastal area of the Netherlands due to transgressions during the Holocene. Vrije Universiteit, Amsterdam, The Netherlands. Prinsen, G., Sperna Weiland, F., & Ruijgh, E. (2015). The Delta Model for Fresh Water Policy Analysis in the Netherlands. Water Resources Management, 29(2), 645-661. doi: 10.1007/s11269-014-0880-z Smit, C., van Duin, W., Henkens, R., & Slim, P. (2005). Casus Hondsbossche Zeewering. Een verkenning van de ecologische effecten van verschillende kustverdedigingsvarianten in de omgeving van de Vereenigde Hargeren Pettemerpolder. Alterrarapport, 1194. TNO. (2015). DINOloket - Data en Informatie van de Nederlandse Ondergrond. 2015, from https://www.dinoloket.nl/ Van Pelt, S. C., & Swart, R. J. (2011). Climate Change Risk Management in Transnational River Basins: The Rhine. Water Resources Management, 25(14), 3837-3861. doi: 10.1007/s11269-011-9891-1 Velstra, J. (2013). Zoet en zout grondwater Hondsbossche Zeewering, binnenduinrand en polder: interpretatie van SkyTEM metingen: AcaciaWater. Verburg, w. g. G. (2009). Natura 2000-gebied Abtskolk & De Putten. Programmadirectie Natura 2000 Retrieved from http://www.synbiosys.alterra.nl/natura2000/gebi edendatabase.aspx?subj=n2k&groep=8&id=n2k16 2&topic=introductie.

Research Proposal ESS-80836

P a g e |6