Can MCDA Serve Ex-Post to Indicate ‘Winners and Losers’ in Sustainability Dilemmas? A Case Study of Marine Spatial Planning in Germany

Abstract

Multi-criteria decision analyses (MCDAs) have been developed to support and evaluate decision-making on multi-layered problems. The benefit lies in creating transparency, among other benefits, especially in tackling divergent stakeholder interests. Within the energy transition, area shortage can lead to sustainability trade-offs, calling for the reconciliation of planning processes and satisfactory compromises. While ex ante MCDAs complement planning, the ex post consideration of processes has been less widely studied. Using a case study of offshore wind energy (OWP) within German marine spatial planning, we investigated the shifting weights of sustainability criteria and stakeholder interests. A multi-criteria approach (Preference Ranking Organization Method for Enrichment of Evaluations (PROMETHEE)) addressed how decision-making can be iteratively traced, and the winners and losers indicated in sustainability dilemmas, such as between climate and biodiversity implications. Findings illustrate that stakeholders are divided in the green-on-green dilemma. The ‘winners’ embrace the branches of energy and climate protection. It remains a question though for ‘losers’ how weighting decisions of sustainability goals can be detrimental, such as ‘good environmental status’, and what kind of balancing occurs. How compromises are found, such as through transparency and solid justification, is crucial in satisfactorily solving trade-offs for public interests. PROMETHEE makes revealing stakeholder constellations within policy dynamics feasible, though assuming there is the will to work multidisciplinarily within future planning decisions.

1. Introduction

The energy transition involves complex challenges [1]. Along the path to a climate-friendly policy, the expansion of renewable energy emerges as a key factor. It is not just technological solutions that need to be developed, such as those required for the switch to an energy system with more volatile and decentralized energy sources (e.g., wind energy, photovoltaics). In order to achieve the expansion targets, suitable land is also a basic prerequisite [2]. However, space is limited as there are many, and at times conflicting, claims and interests that require balancing [3]. The available area has become an ‘eco-currency’ [4], which becomes particularly noticeable, for example, in the case of wind energy [5]. As a result, deploying renewable energy presents itself as a socio-political issue as well. Stakeholders compete for the legitimacy of their actions and future organizational survival, often following conflicting goals [6]. Sustainability dilemmas arise when goals from ecological, economical, and societal concerns are embraced within an often vivid discussion arena [7,8]. It is a question of how to reconcile the demands of divergent stakeholders and, at the same time, realize the energy transition, while getting all concerns into view in a transparent and fair way. The task of spatial and environmental planning comes to the forefront when planning methods are used to realize an environmental social energy transition under suitable area and energy quantities.

An example of a negotiation process of competing sustainability goals among stakeholder interests and power dynamics can be seen in the update of marine spatial planning (MSP) in the German North and Baltic Sea. The maritime space is zoned into economically usable regions, which include the Exclusive Economic Zone (EEZ). As a consequence, the space is limited, and different demands have to be accommodated while stakeholder compete for locations [9]. In the wake of climate change movements, the expansion path for offshore wind energy (OWP) was increased to 40 GW by 2040 (§ 1 (1) Wind Energy at Sea Act (WindSeeG)), which is being retargeted due to climate protection and energy security circumstances [10]. Achieving energy goals is prominently driven by the climate protection narrative [11]. At the same time, there are requirements of the Marine Strategy Framework Directive (MSFD) as well as the EU Biodiversity Strategy 2030 that have gained in importance via the EU Green Deal. Among other interests and goals (e.g., military, resource extraction), securing shipping routes appears to be part of the policy agenda for economic performance and for the capacity of ports, as are fisheries’ requirements [12].

MSP in the EEZ must fulfill the coordination, balancing, and precautionary mandate between interests (e.g., fisheries, shipping, OWP, military, nature conservation, mineral resources) [13] (§ 1 (3) German Spatial Planning Act (ROG)). German MSP is considered to be pioneering and has been a model for other countries in the EU. It is described as a political process composed by numerous discursive struggles [9]. However, there is also the view that MSP is a post-political instrument, which seems to also be characterized by a process guided more by the needs of elite interests than by concern for public good, leading to depoliticization, which then fails to shift the regime towards sustainable development for continued economic progression [9]. Such a process arises, for example, when there is a lack of debate about the purposes of the planning process itself, leaving little room for alternatives, suspending opposition, and (inter-agency) power struggles [9,13,14,15]. This highlights the role of planning itself, especially given the need to resolve sustainability dilemmas under energy targets in striving for the energy transition. It appears necessary to investigate how far objections towards an equitable planning process can be methodically reviewed, examine the balancing of conflicting sustainability goals in the course of planning processes, and identify winners and losers.

Multi-criteria decision analysis (MCDA) is a systematic method that has been used to support the analysis of multiple alternatives within complex problems through multiple criteria [16]. The benefit of MCDA is its provision of a transparent and structured consideration of trade-offs in decision-making, among other benefits such as providing a flexible tool for mapping decision problems [17], evaluating the robustness of decisions [18], allowing stakeholder participation, and analyzing planning scenarios as well as their performance in a traceable way [18,19]. The use of MCDA methods in the environmental and energy fields has been growing [17,20,21,22], while tools for decision support in wind energy site selection have been developed, e.g., [23,24] for OWP [25,26,27,28,29,30,31,32].

While many ex ante MCDAs complement planning, the ex post consideration of planning processes in the field of wind energy has been less widely studied, cf. ex post MCDAs [33] for the food sector, [34,35] for agri-environment schemes, [36] for urban planning, and [37] for ex post MCDA. MCDA methods are generally applied for the ex ante evaluation of alternatives to minimize possible ‘wrong’ decisions, or when seeking to determine whether it is opportune to proceed with a proposition, cf. ex ante MCDAs [17] for renewable energies, [18] for nature conservation, [20] for environmental science, [22] for spatial MCDA, or [37,38] for sustainability assessment. In contrast, the ex post evaluation allows for the hindsight of completed processes, tracing their effects. This is used to identify problems and to highlight unexpected consequences, allowing future processes to learn from past concerns [38].

To track down planning processes about solving sustainability dilemmas under an energy and climate goal, the structured and participatory approach of MCDA in ex post appears eligible for two reasons. First, the evaluation to determine which concerns are prioritized in planning is described as a normative problem [39]. The criteria and interests stated in evaluation processes are predominantly not comparable, which means they are ‘incommensurable’ (e.g., financial gain of fishery vs. area for nature conservation). MCDA methods are based on the principle of incommensurability of criteria, i.e., their weak comparability [40]. Since a lack of a common measurement scale is recognized, both qualitative and quantitative information is linked to MCDA methods. Second, following the weighting of criteria and alternatives, the need for public participation has been increasingly recognized in recent developments of MCDA [41]. This enables the transparent integration of stakeholder values and a better understanding of the decision-making process (

Table 1. Procedures of a multi-criteria analysis (MCA) from step 1–4 and 7 (white) for supporting decision-making. The working steps of a multi-criteria decision analysis (MCDA), a form of MCA, involves complete value setting and weighting of criteria employed as well as an MCDA aggregation method to combine weights, such as Preference Ranking Organization Method for Enrichment of Evaluations (PROMETHEE), which is comprised of steps 1–8 (white and blue) according to [42].

Step 1	Setting the Context of the Decision
Step 2	Identifying alternatives
Step 3	Identifying targets and criteria (qualitative, quantitative)
Step 4	Describing the performance of the criteria within a performance matrix (points/weights)
Step 5	Assigning weights to each criterion
Step 6	Combining the weights of the criteria using an aggregation method (e.g., PROMETHEE)
Step 7	Visualization of the results (e.g., ranking)
Step 8	Sensitivity analysis

).

Therefore, we aim to explore how MCDA can be used to illustrate the power dynamics of involved stakeholders and their sustainability goals in planning processes, and to highlight consequences for balancing trade-offs and the role of planning itself. Particularly in light of a scarcity of space to satisfy all interests in MSP, the question of weighting processes comes under scrutiny, as does the role of planning in the overall structure of political dynamics and the extent to which it complies with a democratic planning mandate or post-political tendencies.

We introduce an ex post MCDA approach, employing the aggregation method ‘Preference Ranking Organization Method for Enrichment Evaluations’ (PROMETHEE) within the case study of the MSP update in the German North and Baltic Sea [43]. For an ex post-analysis, stakeholder values are derived from the completed planning process, i.e., the submitted statements. Given the recent debate within the policy arena about tackling sustainability trade-offs in MSP [4], here, we analyze three research questions using MCDA:

• In the planning process, i.e., in the marine spatial plan and drafts themselves, how have the weightings of the individual interests shifted?
• Who are the winners and losers in the course of the planning process?
• What reasons have led to a possible shift in power over time and what are the implications for fair and transparent planning processes?

2. Materials and Methods

2.1. Case Study Selection and Data

2.1.1. Case Study Selection: Marine Spatial Planning Process in Germany

As one of the earliest examples, German MSP in the EEZ, initiated in 2004 and established in 2009, appeared as part of mediating trade-offs between wind energy policies, especially those regarding nature conservation [13,44,45]. The update to the German MSP in 2021 was initiated by the responsible Federal Ministry of the Interior, Building and Community (BMI) with the support of the Federal Maritime and Hydrographic Agency (BSH) [46]. The update consists of a multi-stage planning process that required national and international consultation for a first draft plan [47] and, with respective feedback from public consultation, a second [48] and ultimately final plan [49] (Figure 1).

Detailed research on the consultation process was encouraged as the approval agency BSH endeavored for more than just a common or rather low quality process [13]. Particular attention was paid to a more comprehensive consideration of fundamental planning options [1,50], which had been criticized previously [51]. For the amendment, a scientific advisory board was also established to accompany the planning process with a limited yet independent mandate [52]. Based on initial consultations, but also on an internal analysis of what developments had occurred since the first marine spatial plan of 2009 [53], the planning team at the BSH initially developed three illustrative basic scenarios—business as usual, nature conservation, and climate protection—for spatial development in the German EEZ [1,50]. This supported a more structured scoping process, i.e., the early procedure of the Strategic Environmental Assessment (SEA) to detail the relevant scope of the upcoming assessment [54] and to initially register the expressed opinions of the involved stakeholders from the beginning.

As a result of the scoping meeting, the BSH found itself in a good position to present a first draft that merged all of the concerns. This initial draft was again open to comment, and was published publicly [13,47]. Statements could also be submitted for a second, adjusted draft [13,48], with the effect, however, that some key stakeholders in the planning process did not comment again. The BSH not only posted the statements received (consented by senders) on the planning process website, but a verbatim record of the discussions during the scoping meeting team was also made publicly available [55]. This transparency went beyond the practice in other, less ambitious planning processes. It already indicated how the discourses on the pathway towards the amended marine spatial plan were to unfold subsequently. We emphasize that alternative analysis, publicly available planning documents, and statements constitute a prerequisite of an ex post analysis, as it relies on available information to trace interests and spatial claims related to the process.

2.1.2. Data Input from the Case

To perform an ex post MCDA analysis, the two drafts as well as the final plan were used as alternatives to evaluate planning changes with regard to the value setting of stakeholders [46]. The final plan was used to compare which stakeholder interests were ultimately able to prevail. Geospatial data on area uses in the three plans were obtained from the BSH as vector data and were evaluated with a geographic information system (Q-GIS) for the criteria analysis [46]. This included using the Q-GIS field calculator to calculate the individual area sizes in square meters for each use (e.g., fishing and shipping). Further Q-GIS tools utilized in the analysis were the clipping and difference functions, which were applied to determine the individual areas of two different uses that overlapped with one another, such as shipping in nature conservation areas. The resulting layers and area sizes were then employed as a data basis for the MCDA. Additional information, such as the power capacity that can be installed within areas for OWP, was obtained from external studies [56].

As data input for the MCDA model and its Performance Matrix (Section 2.2), the contents of the three alternative plans need to be depictable. Some plan contents, such as the reference to the ‘good environmental status’, according to MSFD, cannot be expressed in quantitative terms. These references, i.e., qualitative values, are transformed into criteria within the MCDA model (Section 2.2.2.1) through a yes–no (y/n) scale (criterion applies or does not apply in the plans) or a 3-point scale. The scale reflects with ordinal levels, whether a criterion is lower, higher, or moderate in the plans (e.g., for safety and ease of shipping). The preference of the direction of the scales was translated into numerical values (yes = 1, no = 0, lower = 2, higher = 1, moderate = 0). Y/n-scales were chosen when criteria have been included or not in the plans (e.g., future arctic routes that are included or not). Similarly, 3-point scales were chosen when there were gradations between the plans, such as in terms of safety and ease of shipping. Numeric translation of the scales is linked back to the extent to which stakeholders rate a criterion to be maximized or minimized. For example, safety and ease of shipping, which need to be increased according to stakeholders, would support a lower value for the criterion, as lower values in the scale indicate a higher preference.

Results of data preparation were then transferred to a Performance Matrix common to MCDA [43]. To better capture the differences between the values for each criterion, the results were converted to percentages based on the total (see Performance Matrix, Supplemental Materials).

2.1.3. Stakeholder Selection from the Case

Stakeholders from different branches participated during the consultation period, providing 67 statements on the first draft, which decreased to 32 statements for the second draft. For the MCDA, 15 stakeholders were selected based on a clustering of stakeholders to their vested interests (e.g., from the energy, environmental, fishery, and raw material extraction branches, as well as agencies on different planning levels) (

Table 2. Number of statements towards the first draft and second draft within the consultation phase of the updating process of MSP.

	Energy Companies	Manufacturer, Project Developers, OWP Associations	Environmental Associations	Fishery Associations and Companies	Other Economic Associations and Companies	Research Facilities	(Federal) Agencies, Counties and Cities	Other
MSP-Draft-1	3	3	7	4	11	4	31	2
MSP-Draft-2	1	6	4	1	6	0	14	0

).

The selection of stakeholders was based on the following specification: Selected stakeholders should cover identified interest groups and have introduced statements to both of the drafts to identify their preferences for changing criteria within the planning stages. As not all stakeholders participated in both consultation phases, particularly federal ministries, a selection of federal agencies was used to cover potential change agents at the decision-making level (such as the Federal Ministry for Transport and Digital Infrastructure (BMVI), Federal Ministry for the Environment, Nature Conservation and Nuclear Safety (BMU), Federal Ministry for Economic Affairs and Energy (BMWi) and Federal Agency for Infrastructure, Environmental Protection and Services of the German Armed Forces (Bundeswehr)), as well as the Association of the Seawater Gravel Industry e.V. (ASGI) for capturing vested interests for raw material extraction (cf. statements: [57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81]). For the MCDA, it is assumed that those stakeholders are satisfied with plan changes in the planning process. In the sense of an ‘educated guess’, it can be postulated that it might have been obvious to refrain from submitting a second statement, especially if stakeholders agreed with the respective changes made to the draft, or possibly felt that their previous comments had been accepted to a large extent. For the final plan, it is concluded that the basic preferences of stakeholders for maximizing or minimizing criteria (such as area for nature conservation, area for OWP) have not changed significantly during the process (Figure 2).

2.2. MCDA Model

With the MCDA model, criteria are defined based on prior data preparation of the plan contents (Section 2.1.2). Criteria are evaluated considering the plan alternatives and stakeholders’ preferences using an aggregation method known as PROMETHEE. How the PROMETHEE method was selected and how it operates is first explained in principle (Section 2.2.1). Based on this, precisely what data are entered when in two models is determined, which include a single stakeholder MCDA (Section 2.2.2) and a group MCDA that builds upon the prior model (Section 2.2.3).

2.2.1. Selection of MCDA Aggregation Method PROMETHEE

2.2.1.1. General Method for Single Stakeholder MCDA

Using a systematic literature review for MCDA, a viable ex post MCDA model was developed and analyzed according to the PROMETHEE method [37,82]. Compared to multi-criteria analysis (MCA), multi-criteria decision analysis (MCDA) methods are a form of MCA which contain the scoring of alternatives and criteria, combining them by weights to receive a ranking for each alternative [43]. Different aggregation methods might be used for this purpose. We developed an aggregation model within the group of Multi-Attribute Decision Making (MADM) models and outranking school [83]. For a detailed description of the overall MCDA model selection, see the Supplemental Materials.

The outranking aggregation method, PROMETHEE, emerges as one that is applicable for the purpose of the case study analysis. It includes visualization tools such as the Geometrical Analysis for Interactive Decision-Aid (GAIA) plane to picture stakeholder constellations [84] and is a common approach often used for the latter in the field of environmental management [82]. PROMETHEE has been developed by Brans and Vincke [85] to help decisionmakers perceive the difference between the evaluations of criteria. It requires only indifference and preference thresholds for different criteria, as this method is based on preference degrees (between 0, no preference, and 1, strong preference), e.g., compared to Elimination and Choice Translating Reality (ELECTRE) [83]. For an ex post analysis based on the case study, this method presents a suitable approach since the retrospective angle does not allow for an individually tailored research design for stakeholders asking about values, but enables a post hoc evaluation of their statements, not all of which are uniform in content. The statements allow to at least identify the preference or indifference of criteria as features in the planning stages by using preference functions based on the PROMETHEE method (see Section 2.2.2.3).

To represent stakeholder preferences, Brans and Vincke [85] developed six preference functions. The linear preference functions require the indifference threshold q (no preference at a certain value of a criterion, q = 0) and the preference threshold p (preference at a certain value of a criterion). As soon as the difference between the evaluations for a criterion is higher than the preference threshold, a strong preference exists (preference degree is 1) and vice versa (preference degree is 0). Qualitative criteria are mainly mapped via linear preference functions such as functions of type I (usual preference function, q = p = 0), type II (quasi criterion preference function), and type IV (level criterion preference function) and quantitative criteria via functions of type III (V-shape preference function, q = 0) and type V (linear preference function) [85,86] (see Supplemental Materials for charts).

Based on the preference functions, the PROMETHEE method is built upon pairwise comparison of the alternatives characterized by their specific criteria [87]. A pairwise comparison entails that each alternative is compared to the other alternatives (i.e., outranking). The criterion pairwise preference degrees are summarized in scores, measuring how much an alternative is preferred over or by all other alternatives. This is indicated by positive flows (preference over all alternatives), negative flows (average of all preference degrees of an alternative compared to one alternative), and the net flows (subtracting the negative flows from the positive ones) [83]. In detail, the positive flows indicate how an alternative is outranking the others, while a higher value indicates how much better the alternative is. The negative flows illustrate how an alternative is outranked by the other alternatives, i.e., the lower the value, the better the alternative [87].

Stakeholders may assign different levels of importance to individual criteria, i.e., they are weighted differently (Section 2.2.2.2). By associating a numerical value, global positive, global negative, and global net scores can be displayed, which cover all criteria. The ranking of alternatives is thus based on the global flows, and every alternative receives a certain ranking profile [87] (Section 2.2.2.4). While PROMETHEE I contains only the positive and negative flows and thus the incomparability of alternatives, PROMETHEE II includes the net flows and thus a complete ranking that excludes the incomparable status, which is employed for this case study. Using the Performance Matrix, global net flows can be complementarily deduced from criterion preference degrees by means of the weighted sum (

Table 3. Mathematical calculations within PROMETHEE (adopted from Lerche et al. [20]; for more details, see also Brans and Smet [87], Ishizaka and Nemery [83], and Brans and Vincke [85]).

Step	Mathematical Calculation	Description
Outranking relations	$\pi \left(a_1{b}_j\right)={\displaystyle \sum _{k-1}^k}{w}_{k}x{p}_k\left(a_1{b}_j\right) with j=\left(1,\dots ,n\right)$	$\pi \left(a,b_j\right)$ indicating the outranking relation of alternative a and to alternative $b_J$, preference values $p_k$, weights $w_k$, and criteria k
Outflow for each alternative	${\varphi }^{+}\left(a\right)=\frac{1}{n-1}{\displaystyle \sum _{j=1}^n}\pi \left(a,b_j\right)$	${\varphi }^{+}$ for outflow
Inflow for each alternative	${\varphi }^{-}\left(a\right)=\frac{1}{n-1}{\displaystyle \sum _{j=1}^n}\pi \left(b_j,a\right)$	${\varphi }^{-}$ for inflow
Netflow for each alternative	${\varphi }^{n\mathrm{e}t}\left(a\right)={\varphi }^{+}\cdot \left(a\right)-{\varphi }^{-}\cdot \left(a\right)$	${\varphi }^{n\mathrm{e}t}$ for netflow

).

2.2.1.2. General Method for Group Stakeholder MCDA

Group Decision Support System (GDSS) extensions for PROMETHEE have been developed [87,88] to embrace and compare the perceptions of several stakeholders with regard to the decision problem [83]. For PROMETHEE GDSS, every stakeholder scores and ranks the criteria given and thus each alternative using PROMETHEE II. The resulting rankings are then aggregated into a group ranking by considering each individual net flow as a single criterion, enabling a comparison of stakeholders’ preferences in relation to each other [83]. Various software tools exist for the aggregation of stated weighting and preferences. For this analysis, Smart Picker Pro [89] was used, which, according to Ishizaka and Nemery [83], experiences a good application. Figure 3 summarizes the different methods employed within the ex post MCDA model.

2.2.2. Single Stakeholder MCDA—Case-Specific Input Parameters

2.2.2.1. Criteria

Identified criteria for an MCDA need to be consistent [43], i.e., to encompass the affected concerns within MSP, as well as to capture the changes during the planning stages. An overarching approach with a view to the three pillars of sustainability ‘economy’, ‘ecology’, and ‘society’ therefore seemed suitable for structuring potential sustainability trade-offs from an interdisciplinary perspective. A literature analysis on sustainability conflicts in OWP and MSP was carried out to assess the extent to which the narratives in the statements reflect common negative and positive effects. The results were compared and mentioned criteria within the stakeholder statements.

Following the genesis of MSP, from the first and second draft to the final plan for MSP, this approach permitted an ex post analysis, revealing which issues may be relevant and which were actually voiced by stakeholders in the case study. Linking stakeholder perspectives on criteria back to indicated criteria in a literature analysis was considered useful, since some discourse analyses, such as those conducted by Weber et al. [8] and Jessup [90], have indicated that stakeholders do not always name all the issues that affect them, but may also strategically choose arguments, i.e., criteria, depending on the case constellation and policy context. This allows the MCDA to be made plausible and to assess the transferability of findings from the case study (Figure 4).

For the literature analysis, a keyword search (e.g., OWP planning, MSP, environmental, societal, impacts) was carried out in common databases (Semantic Scholar, Google Scholar, Web of Knowledge). A three-level target system with supporting criteria (n = 59 for ecology, n = 31 for society, n = 71 for economy) often applied in MCDA and PROMETHEE within the field of sustainability research was developed on the basis of the three sustainability pillars (e.g., Lerche et al. [20], Wu et al. [91], Abdel-Basset et al. [92]), having the sustainable siting of OWPs as the overall target (see criteria hierarchy in the Supplemental Materials).

Forty-four criteria were selected for the MCDA based on the literature review, the stakeholder statements, and the contents of the marine spatial plans in the planning stages, considering the co-use of different interests, especially those due to the scarcity of space (e.g., OWP, nature conservation, shipping). The selection of criteria focused on the main lines of conflict mentioned in the statements and partially complemented them with a more comprehensive impact analysis, according to criteria identified in the literature analysis (Figure 5).

2.2.2.2. Criteria Weights

Different stakeholder interests enter the MCDA; however, these interests are not always addressed by all stakeholders at the same time. The individual weighting was based on an analysis of the stakeholders’ statements and aimed at weighting the individual criteria as importance weights [20]. The importance of a criterion increases when it is weighted higher [87]. This approach can therefore be described as non-hierarchical, since only the criteria are weighed along the hierarchy of objectives and criteria, which characterizes MCDA methods. In contrast, within a stricter hierarchical MCDA approach, such as AHP, (sustainability) targets are also weighed to derive partial weights for individual criteria to help understand the structure of the problem [20].

In order to reflect the different weights of criteria to stakeholders, the Simple Multi-Attribute Rating Technique (SMART) was used, which assigns a score as a percentage weighting value to mentioned interests, i.e., criteria [93]. A two-stage approach was applied. The criteria for MCDA, which were named by the stakeholders in their statements, were considered (all equally) important, and a sum of 100 points was assigned to these criteria to calculate a percentage weighting value in the end in relation to the sum of all criteria. For criteria that were not named in the statements, it was assumed that they were not relevant for stakeholders; therefore, these received 0 points. The number of weighted criteria, i.e., criteria named in the statements, was then used to determine the percentage share and thus a weighting value of stakeholders for every criterion (

Table 4. Stakeholder-specific weights to criteria depending on whether criteria were mentioned or not within the consultation phases. Not mentioned criteria received a weight of 0%.

	Weight Mentioned Criteria in Relation to the Sum of MCDA Criteria (%)
TGO (Transmission grid operators)	4.76
BWO (Federal offshore association)	3.85
BUND (BUND Friends of the earth association)	5.88
JEA (Joint environmental associations)	4.35
SFA (State fishing associations Schleswig-Holstein)	6.67
CCI (Chamber of commerce and industry Schleswig-Holstein)	5.56
ASGI (Association of the seawater gravel industry)	16.67
MELUND (Ministry for Energy Transition, Agriculture, Environment, Nature and Digitalization Schleswig-Holstein)	5.23
MWATT (Ministry of Economics, Transport, Labor, Technology and Tourism Schleswig-Holstein)	8.34
MPA (Monument protection authorities of the coastal federal states Mecklenburg-Western Pomerania)	50
BMVI (Federal Ministry of Transport and Digital Infrastructure)	9.09
BMU (Federal Ministry for the Environment, Nature Conservation and Nuclear Safety)	9.09
BMWi (Federal Ministry for Economic Affairs and Energy)	8.34
BfN (Federal Agency for Nature Conservation)	7.69
BW (Federal Agency for Infrastructure, Environmental Protection and Services of the German Armed Forces)	50

).

2.2.2.3. Criteria Preference Functions

By using the preference functions by Brans and Vincke [85], the stakeholders’ preferences based on differences in criteria values are covered separately [20] (see Section 2.2.1.1). Based on references to the suitability of preference function III (V-shape-function) for quantitative values (e.g., area sizes for uses of the sea) and I (usual-function) for qualitative values, the adequate preference functions were chosen. Which of the two functions selected for each criterion is listed in the Supplemental Materials. Basically, the selection is grounded on the following capabilities of the functions to reflect the values of stakeholders towards criteria:

(1) The usual preference function (Type I) shows small differences between values particularly strongly, which is why this function is well suited for yes–no scales and 3-point scales, as used here. A criterion is met or is not in the plan alternatives. (2) Applying the linear function over the V-shape (Type III), the intensity of the preference increases linearly up to a certain value p, and, above p, the preference is strong (preference value). Using this function for quantitative criteria, a detailed indifference value is not needed, indicating the point at which certain value a criterion becomes significant for stakeholders. The V-shape function (Type III) is particularly suitable, since individual values for the criteria are not named in detail in the stakeholder statements; for example, the point at which CO₂ savings and the expansion of offshore wind energy in detail are most preferred. Rather, when employing the V-shape, generally, a linear preference can be assumed with increasing attribute values of a criterion (e.g., increasing CO₂ savings through the expansion of OWP is generally considered positive). The largest attribute value of the criteria is estimated here as value p, i.e., it is assumed that the largest attribute value is preferred (or rejected) in each case (

Table 5. Preference functions for the criteria.

Criteria No. (See Supplemental Materials)	Preference Function	Indifference Threshold	Preference Threshold
1–14, 17–23, 25, 27, 29–33, 35, 37–41	V-shape (Type III)	0	Highest attribute value of a criterion
15, 16, 14, 26, 28, 34, 36, 42–44	Usual (Type I)	q = p = 0	q = p = 0

).

The direction in which the preference functions need to be turned, i.e., whether higher attribute values are preferred or rejected by stakeholders, is specified in PROMETHEE, as well as whether a value should be minimized or maximized (Min/Max). These values could be determined based on a qualitative analysis of the stakeholders’ wording in their statements. Positive descriptions of criteria were recorded as meaning that these criteria should be maximized and vice versa (

Table 6. Analysis of wording in stakeholder statements to obtain preference values for minimizing or maximizing criteria.

Attitude towards Attribute	Example Key Words in the Statements	Direction in Which the Attribute Should be Modified
Attribute is satisfactorily implemented in the marine spatial plan	we support (the reduction in spatial uses for…/establishing more…) we appreciate	Min/max
Attribute is not satisfactorily implemented in the marine spatial plan	an adjustment is to be initiated zones should not be passed through a designation is considered reasonable here	Min/max
Attribute is not mentioned in the statements	no keywords in the statements	No modification

and Supplemental Materials for each criterion).

2.2.2.4. Results—Ranking, GAIA Plane, and Sensitivity Analysis

The ranking and the GAIA plane were plotted for each stakeholder, allowing to identify which plan alternative would meet the individual interests compared to the final plan that finally prevailed. Moreover, the GAIA plane shows which criteria and plan alternatives are opposed from the stakeholders’ point of view, as their parameters were implemented differently, be it well or poorly, in the plans [84]. For stakeholders such as environmental groups, the results are considered a ‘best-out-of-worst case’ selection, as these tend to reject all drafts due to wind energy use being prioritized too highly.

A sensitivity analysis was performed with Smart Picker Pro by checking for individual stakeholders (e.g., BMVI) what influence the weighting of a criterion has on the ranking using the walking weights approach in PROMETHEE (e.g., for criterion future arctic routes). If the ranking varies even with small weight changes, it indicates that the results are sensitive, i.e., that the results are valid only for this particular constellation of criteria, and that slight changes lead to different results (and vice versa for robust results) [83].

For ranking visualization, alternatives that have a positive net score are displayed in blue, whereas alternatives with a negative net score are shown in red. Using the descriptive ranking, the contribution of each criterion for each plan alternative to the final score is illustrated [87].

Within the GAIA plane, the decision problem between plans and stakeholder values is visualized [83,84]. While the blue bullets show the marine spatial plans as alternatives from the first and second plan drafts to the final plan, the arrows show the relations of the criteria by a vector [94]. The black arrow is the decision stick, i.e., the aggregated compromise for the decisionmaker. The closer the plan alternatives are to each other, the more similarities they have and vice versa. This is also reflected for the criteria and their direction with respect to the plans. The criteria’s (i.e., the arrows) position shows which criteria are compatible and which are in conflict [94]. The closer a criterion is to a plan alternative, and when a criterion is pointing to the alternative, the better the performance of the criterion in that plan, as perceived by the respective stakeholder, is. The length of the arrows indicates the influence of the criterion with respect to the plan alternatives, i.e., the more distinguishable a criterion is compared to the plan, the longer the arrow and vice versa (e.g., only weighted criteria are displayed in the GAIA plane) [83,84,87].

2.2.3. Group MCDA and Sensitivity Analysis—Case-Specific Input Parameters

While the single stakeholder MCDAs represent the preferences for the planning stages for each stakeholder, they cannot represent the relationships among stakeholders. For this, a group MCDA is needed, which reflects the positions of the stakeholders in their preferences to each other and for the plan alternatives in the overall picture (see Section 2.2.1.2). The group MCDA shows which stakeholders would ultimately prefer which draft and/or final plan and how the stakeholders’ interests thus conflict with each other or resemble their demands. The group MCDA serves as a stakeholder analysis to identify specific stakeholder groups as well (see Section 2.2.2) [95]. For this purpose, the results of the single stakeholder MCDAs are combined, and a group MCDA is conducted. The overall picture is accomplished considering each individual net flow of the single stakeholder MCDAs as a new individual criterion.

The results of each stakeholder’s ranking of the first and second draft and the final plan have been transferred to a three-step ordinal scale (1,2 and 3), resembling the ranking of the plans as a data basis for the Performance Matrix for the group MCDA (see Supplemental Materials). In order not to exaggerate very small differences in preferences for single plans, and therefore distort the ranking results, score values with an equal second and third decimal digit received the same value of ordinal scale (e.g., for 0.11 and 0.116). It was assumed for the group MCDA that all stakeholders selected in this case are weighted equally, i.e., the weighting is 6.67% for each stakeholder (n = 15). Thus, underpinning this assumption is that all stakeholder opinions on the plans are given equal weight in the decision-making process.

A sensitivity analysis for the group MCDA was performed as well using the walking weights approach in PROMETHEE to analyze the impact of a possible higher weighting of certain stakeholders on the outcome, i.e., the preference between the three plan alternatives. For this purpose, stakeholders that are assumed to be major players were weighted at just about 50%, such as the BMVI and the BMWi as well as the BMU and BfN. Thus, other stakeholders have received a lower weight in the analysis in relation to the total sum, i.e., 100%. The results were visualized in the graphical ranking, the descriptive ranking, and the GAIA plane.

3. Results

3.1. Shifts of Weighting of Vested Interests within the Planning Stages

3.1.1. Shifting Prioritized Interests during the Planning Process

The initial results indicate that the MCDA enables the examination of decision-making processes and policy dynamics retrospectively. As a long-term implementation of marine policy goals and principles, throughout the three planning stages of MSP, the planning alternatives mainly vary in their spatial designations (Figure 6).

Applying the Performance Matrix illustrates that the share of areas for sustainability criteria has changed in the process, particularly through the establishment of co-uses to realize shared interests on the same sites (see Performance Matrix, Supplemental Materials). While the first draft tends to designate greater areas solely for nature conservation as well as fewer areas for wind energy and shipping, the second draft shows opportunities for co-uses, as is the case with nature conservation that is combined with OWP and shipping. For the final plan, minimal changes were made to the second draft, minimally enlarging shipping areas (some within nature conservation areas) and areas for infrastructure projects, which are shown for illustrative purposes.

Differences in area allocations are manifested by variants. The share of individual areas has not changed significantly in quantitative terms; however, instead, the qualitative effects from co-uses are considered more. This effect is illustrated when compared to the previous basic planning scenarios prior to the planning process (see Section 2.1.1) [96]. Moreover, the share of areas for OWP and shipping have been increased through the introduction of a new planning category that includes the conditional use of the respective areas when ecological reports are available [48,49].

In addition, applying the Performance Matrix illustrates that the share of available space to bird migration has increased from the second draft to the final plan. As a result, the potential for conflict with overlaying areas for wind energy also increases. Furthermore, the larger share of wind energy is combined with the co-use of areas for loons and harbor porpoises. Disturbance from OWP in areas for loons is accepted and is moving to a stricter planning category (from reserved to priority areas) [48,49]. Sand and gravel extraction locations remain unchanged throughout the process and, consequently, end up in co-use with nature reserves as well. During the planning process for the second draft and final plan, the reference to the ‘ecosystem approach’ and ‘good environmental status’ under the MSFD was curtailed. Thus, throughout the three planning phases, OWP, nature conservation, and shipping are essentially the uses that are continuously adjusted to make room for one another.

3.1.2. Stakeholder Participation and Voiced Interests

The preceding literature review indicates both positive and negative effects on sustainability criteria (e.g., nature conservation, acceptance) potentially caused by OWP that may be more multi-faceted than that communicated by stakeholders in German MSP. Stakeholders have predominantly introduced criteria from the fields of ecology (nature conservation interests) and economy (interests of the offshore industry, shipping), whereas criteria from the field of society are almost completely absent, such as criteria concerning the fairness and acceptance of the process (see Supplemental Materials). This shows that the societal criteria relevant to the planning procedure were initially perceived positively by stakeholders, and are therefore not highlighted in statements in the first place (e.g., just as planning documents are published on the Internet; therefore, this is not a concern). The perceived fairness of weighing, which involves stakeholders when interests are negotiated, might have been affected just later in the process.

Moreover, stakeholders appear to carry their own advocacy into the planning process, especially since sustainability criteria that stakeholders could potentially represent were not always included. The potentially positive impacts of OWP on fish stocks (artificial reef effect [97]) and climate change mitigation effects are not appreciated by environmental groups, as further OWP would contrast with the goals of ‘good environmental status’ under MSFD. Instead, offshore industry associations adopt the narrative of conservation effects of OWP to push for wind energy expansion (see Performance Matrix, Supplemental Materials). Arguments are usually not substantiated with corresponding data and/or quantified in values.

3.2. Winners and Losers in the Course of the Planning Process

3.2.1. Trade-Offs between Sustainability Criteria within Stakeholder MCDA

Applying MCDA ranking helps in resolving which stakeholders would prioritize which planning stage, i.e., drafts or the final plan of MSP under modified criteria. Comparing the ranking scores, some stakeholders show a stronger preference towards specific planning stages, as corresponding criteria values were changed that emerge as more important, i.e., those that have been named in the statements and thus weighted in the MCDA for those stakeholders (see Supplemental Materials for ranking scores for each stakeholder). The stakeholders most affected during the planning process appear to be public authorities, followed closely by environmental associations and economic interests, such as shipping (e.g., BfN (Federal Agency for Nature Conservation), BMU (Federal Ministry for the Environment, Nature Conservation and Nuclear Safety), BMVI (Federal Ministry of Transport and Digital Infrastructure), BMWi (Federal Ministry for Economic Affairs and Energy), BUND (BUND Friends of the earth association), CCI (Chamber of commerce and industry Schleswig-Holstein), JEA (Joint environmental associations), MELUND (Ministry for Energy Transition, Agriculture, Environment, Nature and Digitalization Schleswig-Holstein), MWATT (Ministry of Economics, Transport, Labor, Technology and Tourism Schleswig-Holstein), SFA (State fishing associations Schleswig-Holstein), TGO (Transmission grid operators)).

For stakeholders with low preferences between one of the plans, the criteria that appear important have not changed decisively (such as for BW (Federal Agency for Infrastructure, Environmental Protection and Services of the German Armed Forces) as being interested in military areas that have (at least not publicity) not been modified during the planning process, MPA (Monument protection authorities of the coastal federal states Mecklenburg-Western Pomerania) in marine monument protection, ASGI (Association of the seawater gravel industry) basically in sea water gravel extraction).

Using the GAIA plane, stakeholders display individual internal trade-offs between criteria and how they have been incorporated into the process. Focusing on the group MCDA in this analysis, which is based on the single stakeholder MCDAs, the opposing values and interests, which turn out in detail in the stakeholder-specific consideration, are explained here by way of an example including stakeholders and the environmental branches and federal agencies. In the environmental sector, the environmental association BUND (BUND Friends of the earth association) would prioritize the first draft plan, especially since criteria such as sediment relocation and conditional priority areas for wind energy are reduced by an overall lower level of wind energy development. Shipping lanes through nature reserves are not planned in the first draft. For ministries at the federal and state levels, different preferences and perceived trade-offs can be identified. The BMVI (Federal Ministry of Transport and Digital Infrastructure), which covers the needs of the transportation sector, considers its own interests to be rather fulfilled with the second and final draft, especially due to increased shipping routes through conditional area uses, which are to be opened for shipping after further assessment. In future, arctic routes would also be better covered. The BMWi (Federal Ministry for Economic Affairs and Energy), having economy and energy transition combined in one institution, also recognizes that sustainability criteria are better fulfilled with the second draft and final plan. In particular, the needs are met by the path to greater wind energy, including the opening of nature conservation areas and areas for research (citation: ‘It is therefore mandatory that co-use take place on areas currently allocated to other uses as well as offshore wind energy’, [69] (p 2). For a comprehensive illustration of all stakeholder MCDAs and GAIA planes, please see the Supplemental Materials.

3.2.2. Prioritized Stakeholder Interests Compared to the Final Plan within Group MCDA

The group MCDA illustrates the share of stakeholder interests within the green-on-green-dilemma (nature vs. climate protection). Under an equal weighting of the pursued sustainability criteria on the part of the stakeholders, the criteria of both the first and second drafts (more nature conservation vs. higher wind energy expansion path and shipping) would be almost equally preferred to the final plan (slightly higher co-use of shipping and nature conservation). Although the second draft and the final plan are similar in their attribute values of criteria, the final plan is generally more rejected as the scores show a negative value range. Yet, relatively small values indicate that preferences for either one of the plans tend to be rather small, as differences between the plans are rather small as well, being elevated more in qualitative manner (e.g., the impacts of shipping in nature conservation areas) (Figure 7).

Taking the GAIA plane for group MCDA indicates whether stakeholders show similar or opposite tendencies in their preferences for planning stages (arrangement of arrows to plan alternatives), similarity of plan alternatives (proximity to each other), and the extent of influence of the respective stakeholder (criterion) regarding the comparison of alternatives. The more different the stakeholders’ preferences (criteria) towards the alternatives are, the longer is the arrow. Stakeholders are plotted on top of each other in the GAIA plane, showing similar preferences on criteria and plans, enabling to pinpoint multiple parties with different views. Thus, stakeholders may be categorized into four parties, revealing (unusual) alliances relating to the three planning alternatives.

One party comprises the stakeholders who would rather prefer the first draft plan (‘best-out-of-worst case scenario’). Among them are representatives of the environmental sector, fisheries, and the sea water gravel industry. Stakeholders see individual conflict lines especially with increased wind energy development (BUND (BUND Friends of the earth association), JEA (Joint environmental associations), SFA (State fishing associations Schleswig-Holstein), ASGI (Association of the seawater gravel industry), MELUND (Ministry for Energy Transition, Agriculture, Environment, Nature and Digitalization Schleswig-Holstein), and MPA (Monument protection authorities of the coastal federal states Mecklenburg-Western Pomerania)).

For the second party, the second draft is more preferred. In particular, stakeholders see the greater expansion path for wind energy and shipping, for example, as well as the planning designation of infrastructure projects such as the Fehmarnbelt crossing (MWATT (Ministry of Economics, Transport, Labor, Technology and Tourism Schleswig-Holstein), TGO (Transmission grid operators), BWO (Federal offshore association), and BMVI (Federal Ministry of Transport and Digital Infrastructure)).

Within a third party, stakeholders of business branches tend more towards the final plan, with slightly higher shares of wind energy and shipping, which also overlap with nature conservation (BMWi (Federal Ministry for Economic Affairs and Energy), CCI (Chamber of commerce and industry Schleswig-Holstein)). Federal environmental institutions show themselves as a fourth party split in the green-on-green dilemma between species conservation and climate protection goals and are thus situated between the planning alternatives of the first and second draft plan (BMU (Federal Ministry for the Environment, Nature Conservation and Nuclear Safety), BfN (Federal Agency for Nature Conservation)).

Although the second draft plan and the final plan have similar criteria attribute values, the second draft would thus be more likely to satisfy some stakeholder interests. The sensitivity analysis on the example of the stakeholder BMVI (Federal Ministry of Transport and Digital Infrastructure) for traffic issues shows that the criterion of the future arctic routes tends to shift the preference towards the second draft and final plan, whereby navigation through nature conservation areas is a criterion that is less present in the second draft plan and thus fulfills the stakeholder interests more. Some short arrows using the GAIA plane for group MCDA display that stakeholder interests are more diverse regarding the plan alternatives (e.g., for BMU (Federal Ministry for the Environment, Nature Conservation and Nuclear Safety), BfN (Federal Agency for Nature Conservation)), indicating the sustainability dilemma within the planning stages (e.g., conservation vs. climate change mitigation through wind energy) (Figure 8).

At the same time, contrasting the final plan to stakeholder interests reveals the winners and losers of the planning process. Stakeholders from business and industry are more likely to achieve their pursued goals with the final plan, this is indicated, for example, for the BMWi (Federal Ministry for Economic Affairs and Energy). In the case of institutions representing both nature and climate protection interests, winners can thus be identified for the latter. In opposition are stakeholders such as environmental associations, which would prioritize the first draft in the MCDA within a ‘best-out-of-worst case scenario’ view and thus experience a weakening of their target interests in the further planning stages.

Thus, the descriptive ranking shows that, overall, the BMWi conveys the greatest preference for the final plan, followed by economic representatives (CCI (Chamber of commerce and industry Schleswig-Holstein)) and representatives of the offshore industry (BWO (Federal offshore association)). In contrast, there are representatives of the environmental sector who would rather see their own interests represented in the first draft, such as the agencies BfN (Federal Agency for Nature Conservation) and MELUND (Ministry for Energy Transition, Agriculture, Environment, Nature and Digitalization Schleswig-Holstein), as on the part of nature conservation. Other economic stakeholders are thus on a similar line, although for different reasons but similar interests through the perceived conflicts of objectives with increasing OWP to own area uses (such as fishing and gravel mining). In contrast, TGO (Transmission grid operators) for transmission grids, for example, see the first draft as a result of unfulfilled interests for developing OWP (Figure 9).

These trends of stakeholder power within the planning stages are also shown by a sensitivity analysis to check the influence of public authorities on the ranking result. If BMVI (Federal Ministry of Transport and Digital Infrastructure) and BMWi (Federal Ministry for Economic Affairs and Energy) were weighted approximately 50% and thus had a greater influence on the plan decision, the second draft and final plan would be given greater preference. For institutions that are divided in their fields of responsibility with regard to climate protection and nature conservation, it also appears that climate protection interests have been given priority, in particular also through a stronger co-use of wind energy with nature conservation areas.

4. Discussion

4.1. The results

4.1.1. Possible Reasons for a Shift of Power over Time

In retrospect, the ex post MCDA analysis assisted in showing the potential winners and losers of trade-offs in a sustainability dilemma [2]. In this case, stakeholders vying for economic interests, such as shipping and wind energy, have increasingly been able to assert their interests at the perceived cost of co-benefits with nature conservation. Reasons that may have led to a higher weighting of these certain concerns may be linked to socio-political circumstances. Wind energy concerns may have been increasingly implemented in the planning process to address societal interests in climate change mitigation [98]. In addition to the German Climate Protection Act, a ruling by the Federal Constitutional Court at the same time as the planning process stated the need to catch up on climate protection targets. The constitutional complaint was filed by Fridays for Future (Federal Constitutional Court Germany, 1 BvR 2656/18, 1 BvR 288/20, 1 BvR 96/20, 1 BvR 78/20), among others, and generated discussion about climate protection efforts on the part of the German government.

At the same time, the EU has raised expansion targets, creating a need for improvements [11]. Nevertheless, the question of the distribution between renewable energy in the energy mix is still open, whereby possible conflicts of onshore wind energy with residents and the different competence and expansion will of the federal states in federalism can be arguments to produce more renewable electricity from OWP, also through a possible greater efficiency of the turbines. However, the cumulative effects on the marine environment may not yet have been sufficiently researched [99], though effects may be felt to be ‘further away’.

Moreover, the fact that areas for shipping and wind energy are being increasingly kept free through MSP might raise the question of whether economic interests, such as for the offshore industry, are pushed under the narrative of climate protection. Shipping, although ensuring the movements of goods, is also an emitter of CO2 emissions, causing possible trade-offs. For Germany as a business location, the possible shortage of goods due to the COVID-19 pandemic [100,101], competition over future arctic routes [102], and the Suez Canal blockage in 2021 [103] may have increased the importance of shipping goods to ports. However, based on the statement submitted, there are fewer stakeholders from the shipping industry, and concerns also appear to have been represented through regulatory agencies.

The interests of nature conservation are backed by EU and national goals, such as the biodiversity strategies and the EU Green Deal. Nevertheless, these targets are perceived to have been set back, even though, in the long term, possible penalties from the EU can be expected if the targets are not met [104,105]. Institutions representing nature conservation interests, which have often joined forces with climate protection interests, yielded to the latter, suggesting that other nature conservation stakeholders may have had less weight in the negotiation process. However, environmental groups have only been representing interests against OWP and not against other economic uses such as shipping; therefore, environmental advocacy may have been lacking in this area.

4.1.2. Implications of Policy Dynamics for Planning

The iterative, largely transparent planning process through publicly available documents and alternatives may be regarded as a credible and accountable one. Not only because it may be appropriate for a cross-sectoral planning agency that also has approval authority. It also centers the planning agency into a decent and fair-minded moderating position—presumably a legitimate reward for the pursuit of many aspects that characterize early and fair participation processes.

At the same time, departmental coordination occurred at the federal level during the planning process, resulting in the second draft, which introduced changes that were received with surprise by some stakeholders, as voiced and discussed in a public technical discussion in a committee of the German Bundestag [44]. With departmental coordination, although many interests brought forward have been taken note of [106], at the same time, it appears that interests have been balanced once again at the regulatory level. This may indicate that, in the concluding planning steps, the process became ever more politicized. The environmental association NABU considers commercial interests to have been imposed by incentives from the Federal Ministry of Economics (BMWi) [104]. Bjärstig et al. [107] found for the Swedish case that, in a discourse around wind energy, governmental agencies together with wind entrepreneurs determined the framing, with NGOs and experts having smaller influence.

Such a process might raise the question of viable compromises when one or more parties may remain dissatisfied or when the actual objective is diluted by too much consideration. By opening conservation areas for other uses such as shipping, wind energy, and raw material extraction, a heightened potential for conflicts may result as well. The actual decisions on the priority of a use, and thus potential conflicts, could be shifted to the approval level. It may be that ‘illusionary compromises’ have been made that might be difficult to reach at the approval level. Possibly, co-uses thus could hamper the efficacy of spatial planning in favor of the sectoral approval of an authority’s discretion.

The process of weighing up why which targets are preferred or withdrawn apparently is partly perceived as being non-transparent by stakeholders [44]. Yet, spatial planning is also described as being part of the formative administration, whereby sounding out sustainability goals is a case of administrative decisions [44]. This assignment and understanding of spatial planning being rather an administrative act at the federal level raise the question of planning itself, as well as undergoing public consultation. Although public consultation does not necessarily have the goal of actually implementing all interests expressed, transparent and well-founded decisions seem to be relevant to gain the acceptance of stakeholders [108]. Still, transparency and a justified weighing of decisions appears to be a significant step to achieve acceptance among stakeholders [108].

Deferring nature conservation goals in favor of climate protection compared to the first draft can, however, be understood as a negotiation result [109]. At the same time, MCDA also shows that stakeholders can be grouped and divided between the first and second drafts, with the result that compromises seem to have been insufficiently considered between concerns such as nature conservation and climate protection. A clear preference of predominant stakeholder interests to the final plan is not present in this case study. Therefore, the task of planning to identify and weigh alternatives can be set, notably regarding the consultation processes towards the final plan, which evoked seemingly no other substantial amendments to the plan.

Yet, given sustainability dilemmas, which are also described as wicked problems in the context of the energy transition [110], finding satisfactory compromises for all parties involved can be a difficult task [111]. Charging post-political planning to the instrument of MSP as a process that, among other, favors negotiation in favor of interests for elites, therefore, takes a double-edged view. On the one hand, there is no one-fits-all solution [109]. Each stakeholder has their own values and methods for developing solutions [111]. On the other hand, there may be compromises that can be considered to have been found if various stakeholders recognize partial interests to be at least represented when the planning process allows transparent decision-making for better acceptance.

Not all sustainable criteria have entered the planning process as adjusting screws, such as nature conservation when compared to areas representing the interests of military, raw material extraction, and maintaining the status quo of shipping, which are linked to, among others, economic interests [44]. If only a few interests are considered variable, the planning leeway with which to identify sufficient areas for specific uses under compromise also may be diminished. Kiker et al. [111] describes conservation concerns as sometimes lacking advocates, as these are often ethical values whose value is subjectively captured and not directly monetized. Quantifying ecosystem services as being discussed [112] might enable a balanced comparison to other sustainability goals, for example, by capturing potential penalties if conservation targets are not met in the long term. However, given that monetization also poses challenges and the lack of standardized approaches, at least MCDA is able indicate, based on the quantification of criteria, which interests are actually being increasingly taken up in planning stages.

4.2. Ex Post MCDA for Tracking down Policy Dynamics

4.2.1. Opportunities

The methodical analysis with ex post MCDA pinpoints that a retrospective view of planning processes enables the portraying of overarching stakeholder relationships, interests, and potential policy dynamics in weighing interests as a thread to perceived fairness and satisfaction [44]. A feature of MCDA as a structuring tool is that the Performance Matrix allows changes in plan content to be quantified according to stakeholder interests [43]. PROMETHEE has offered a useful application, notably due to its overall easy handling and the opportunities to visualize sustainability criteria during the planning stages [113]. The ranking proved useful to map stakeholders’ preferences between plans as a ‘best-out-of-worst case scenario’, i.e., which planning stages are preferred compared to the final plan outcome. The advantage of MCDA lies in its ability to deal with two types of uncertainty, such as parameters for quantification and different stakeholder preferences [114]. PROMETHEE also has the ability to not fill in and weight all criteria when a distinctive value cannot be defined [115].

Nevertheless, it is also important to consider that other MCDA methods might have yielded a different result, e.g., as a result of different weighting methods of the criteria [116]. Løken [117] found that a non-salient MCDA method emerged for energy planning; therefore, it would have been useful to compare the results with other MCDA methods. Anastasiadou [118] considered applying two hybrid MCDA methods in particular as they are advantageous for a robust result [116,119]. When principally comparing MCDA methods, PROMETHEE showed robust results compared to multi-attribute utility theory (MAUT), ELECTRE, and the dominance-based rough set approach (DRSA), with DRSA being the easiest to use, followed by PROMETHEE and MAUT, and finally ELECTRE [39].

Overall, however, roughly the same results can be achieved with different MCDA methods, such as the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS), Multicriteria Optimization and Compromise Solution (VIKOR), and PROMETHEE [21,120]. Asadabadi et al. [121] posited a negative evaluation about the commonly applied AHP method [122,123] within companies, since decision-makers would rather notice errors in the ranking intuitively. Yet, especially the weighting, this represents an essential step which can potentially influence the result significantly [124,125], although an MCDA application does not mean that decision-makers necessarily have to rely on the results [20,126]. Compared to unaided evaluation, however, MCDA was often perceived as helpful and led to different decision results [126]. It is the group MCDA approach that allows societal problems to be better identified and therefore managed in order to pursue public good interests [35,127].

4.2.2. Challenges

The selection of criteria turns out to be a key issue that can have a significant influence on the result of MCDA. According to Viaggi et al. [38], a core challenge can be to understand and interpret the influence of criteria from the overall result. Along with the case study analysis, it also emerges that, with the sum of the criteria, the visualization in the GAIA plane becomes increasingly less legible, especially when plan alternatives display smaller differences. The literature analysis therefore proves helpful to clarify that the discourse about the statements in the consultation process seemed to be more constrained in the case study than in other processes of planning OWP and in MSP. The findings on the role of societal interests in planning decisions, such as transparency, argumentation and weighting of concerns, suggests that, with ex post MCDA, these interests can be reflected upon and addressed yet anew [104,105].

Yet, more robust and quantitative parameters could be chosen for an additional way forward if they are again recoupled with the stakeholders of the planning process [38,111]. This would also allow arguments to actually be quantified for once if stakeholders could be explicitly asked about each attribute value of a criterion, such as handling volume in shipping as a criterion. Moreover, possible difficulties in using MCDA can also be seen in the possible need for instruction for method application, especially for choosing between the available methods, such as between full aggregation and outranking MCDAs (e.g., AHP and PROMETHEE, or a method integration [119]), as well as for understanding their statistical basis [38]. For a layperson, MCDAs may be considered difficult; in practice, it appears advisable to consult an expert to support stakeholders in the ex post analysis, for example, in a support group [38].

True stakeholder participation, also in the ex post analysis, however, requires the will for multidisciplinary work [35]. In ex post analysis of decision-making using MCDA, this can pose an obstacle, along with communication problems [35]. There may be pressure to conform, dominant personalities determining the weighting process, and ambiguous responsibilities [32]. Therefore, stakeholder participation to readjust criteria and weights can also be perceived as being too challenging, time-consuming, and costly to engage in dialogue with stakeholders [32].

However, considering that the expansion path for OWP is to be increased once again with the new coalition agreement of the German government, and thus the securing of areas is in the sustainability dilemma with other uses, in particular areas for conservation, possible obstacles such as time restrictions should not be the decisive criterion. It should be planned for in good time, especially through the possible dialogue of stakeholders with decision-makers. The description of plans and decision-making may often be perceived as imperfect, leaving unclear which goals are addressed and how they are addressed [36]. Therefore, an extension of the ex post MCDA to include feedback from stakeholders or an ex ante consideration when updating MSP would potentially appear appropriate for dealing with sustainability trade-offs [2,128,129].

5. Conclusions

This article has shown how ex post MCDA might allow for reviewing the potential policy dynamics of planning processes, as in the case of MSP in the German North and Baltic Sea. In developing and analyzing an MCDA model based on the outranking approach PROMETHEE, both findings of tackling and weighting sustainability criteria between different planning stages were obtained, as well as the approach of using MCDA as an evaluation tool itself.

In this case, it emerged that sustainability criteria shifted in weights in the spur of moment during planning stages, with nature conservation goals apparently being deferred in the process. During the planning process, the share of areas was increased through co-use, such as for OWP and shipping, allowing for multiple area use with nature reserves. Using group MCDA based on the ranking and the GAIA plane shows stakeholders to be divided in the green-on-green dilemma based on their submitted statements within consultation phases (i.e., between climate protection–conservation). The winners in retrospective appear to be, in particular, stakeholders from the economic and climate protection sector, such as OWP and shipping. The interests of the environmental branches could have been more strongly satisfied with the first draft than they are in the final plan.

Socio-political engagement for climate protection indicates that this might have had an influence on a political departmental vote on MSP, among others, leading to fostered energy paths. This challenges the democratic planning mandate, and it is a question how balancing sustainability interests might come at other costs, whether it appears intentional, and what kind of balancing is taking place in democratic processes. Given that there may not be a one-size-fits-all solution under increasing spatial pressure, the role of how compromises are found, such as through transparency and justification within planning decisions, may emerge as crucial in solving trade-offs and to maintain the satisfaction of social problem solutions and to pursue public good interests. Thus, charging MSP as an instrument of a post-political process that, among others, serves only elite interests is a double-edged sword under the dilemma of compromise. However, considering all sustainability criteria as adjusting screws may appear as key.

Using MCDA served to be feasible in ex post planning analysis as the findings coincided with the following social discussions. PROMETHEE as an outranking approach assisted in facing the challenge of limited weighting possibilities, because no uniform weights of criteria are designated in stakeholder statements, as would have been possible through structured stakeholder interviews. It is therefore conceivable that MCDA may also be tested ex ante in the course of MSP’s further updating, questioning whether decision-making is supported in a social-political arena, particularly by quantifying the arguments of stakeholders and outranking criteria.

Societal competing interests in the energy transition, also discussed as wicked problems, may need supportive instruments such as MCDA for further democratic negotiation, which support transparency and trust between stakeholders. MCDA, however, presupposes that there is a willingness to work in a multidisciplinary manner. As this case study shows, planning might also be understood as executing official decision-making, meaning that a transparent planning process that is open to change may become increasingly important. A further analysis of MCDA might help to support more transparent, democratic decision-making when tackling the siting of renewable energy, including onshore wind energy and photovoltaics.