Institutional misfit and environmental change: A ...

Science of the Total Environment 576 (2017) 599–608

Contents lists available at ScienceDirect

Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv

Institutional misfit and environmental change: A systems approach to address ocean acidification Julia A. Ekstrom a,b,⁎, Beatrice I. Crona c,d a

Natural Resources Defense Council, 111 Sutter St. Flr 20, San Francisco, CA 94104, USA University of California, Santa Barbara, CA, USA c Stockholm Resilience Centre, Stockholm University, Sweden d Global Economic Dynamics and the Biosphere, The Royal Swedish Academy of Sciences, Sweden b

H I G H L I G H T S

G R A P H I C A L

A B S T R A C T

• US laws show quantitative improvement in dealing with ocean acidification. • The analysis brings a quantitative dimension to comprehensive legislative analysis. • The results are consistent with governance literature verifying the utility of the approach. • The method presented can be used to evaluate governance of a less understood issue.

a r t i c l e

i n f o

Article history: Received 14 June 2016 Received in revised form 11 October 2016 Accepted 16 October 2016 Available online xxxx Editor: Dr. Simon Pollard Keywords: Governance Institutional fit Social-ecological system Human-environmental system Institutional interplay

a b s t r a c t Emerging environmental threats often lack sufficient governance to address the full extent of the problem. An example is ocean acidification which is a growing concern in fishing and aquaculture economies worldwide, but has remained a footnote in environmental policy at all governance levels. However, existing legal jurisdictions do account for some aspects of the system relating to ocean acidification and these may be leveraged to support adapting to and mitigating ocean acidification. We refine and apply a methodological framework that helps objectively evaluate governance, from a social-ecological systems perspective. We assess how well a set of extant US institutions fits with the social-ecological interactions pertinent to ocean acidification. The assessment points to measured legal gaps, for which we evaluate the government authorities most appropriate to help fill these gaps. The analysis is conducted on United State federal statutes and regulations. Results show quantitative improvement of institutional fit over time (2006 to 2013), but a substantial number of measured legal gaps persist especially around acknowledging local sources of acidification and adaptation strategies to deal with or avoid impacts. We demonstrate the utility of this framework to evaluate the governance surrounding any emerging environmental threat as a first step to guiding the development of jurisdictionally realistic solutions. © 2016 Elsevier B.V. All rights reserved.

⁎ Corresponding author at: Policy Institute for Energy, Environment and the Economy, University of California, Davis 1605 Tilia Street, Suite 100, Davis, CA 95616, USA. E-mail addresses: [email protected] (J.A. Ekstrom), [email protected] (B.I. Crona).

http://dx.doi.org/10.1016/j.scitotenv.2016.10.114 0048-9697/© 2016 Elsevier B.V. All rights reserved.

600

J.A. Ekstrom, B.I. Crona / Science of the Total Environment 576 (2017) 599–608

1. Introduction Ocean acidification (OA) has emerged on the global environmental agenda as an issue of acute concern (Kerr, 2010). The rising acidity of both surface and deeper waters (Gruber et al., 2012) threatens to negatively impact a range of biological and biogeochemical processes (Beaufort et al., 2011; Orr et al., 2005) and significantly alter marine ecosystems (Fabry et al., 2008) which in turn underpin a wide array of marine ecosystem services (Cooley et al., 2009). OA is also forecasted to have complex interactions with multiple other stressors making predictions for outcomes and impacts on ecosystem services highly uncertain (Boyd, 2011). Despite the apparent urgency of the issue and the role of the ocean in climate regulation, ocean acidification has remained a footnote in the development of climate change and related environmental policy at both international and national levels (Galland et al., 2012; Kim, 2012; Billé et al., 2013). While OA is a global concern, effects will be particularly discernable along coastlines characterized by upwelling or coral reefs. Such regional ‘OA hotspots’ therefore warrant not just a global, but a multilevel approach to governance. At the international level multilateral environmental agreements (MEAs) addressing OA are remarkably absent. Some have proposed that OA can be dealt with through UNFCCC, as both climate change and OA share the root cause of increasing CO2 in the atmosphere (Doney et al., 2009) but others argue that UNFCCC does not provide an adequate legal framework as OA is not an effect of climate change and as such falls outside the UNFCCC’s jurisdiction (Harrould-Kolieb and Herr, 2012; Kim, 2012). This absence of multilateral agreements for policy coordination among states observed for OA has been described as a nonregime (Dimitrov et al., 2007). There are multiple reasons why nonregimes emerge, such as the fact that interactions are not well understood scientifically, the impacts may be felt locally while sources are global, and interventions are likely to interact with a range of other environmental and non-environmental institutions. Combined these create the conditions for a perfect collective action dilemma, diluting the incentives for sovereign states to act, individually or in collaboration (Galaz et al., 2012). OA is an issue domain that exhibits all three of these characteristics. A regime formation around the issue is therefore unlikely in the near future. Because of this and the regional distribution of initial OA effects (Strong et al., 2014), local and national governance options need to be explored in parallel with international efforts (Biermann, 2015; Galaz et al., 2012; Ostrom, 2010). This paper therefore examines the regulatory landscape relating to OA at the federal level of the US. While sub-global levels may be better suited for developing adaptive responses, a key challenge remains: the problem of fit, referred to above (see Brown, 2003; Folke et al., 2007; Galaz et al., 2008; Young, 2008 for further discussion on the topic). Lack of fit between ecosystems and governing institutions (form hereon referred to as institutional gaps) has been the cause of significant environmental degradation worldwide (Barnes and McFadden, 2008; Folke et al., 2007). Given the novelty of OA as a policy domain, identifying institutional gaps is thus a first critical step in understanding the national governance landscape in place to address OA. However, analytical tools for examining both fit and identifying solutions have remained sparse (Vatn and Vedeld, 2012; Epstein et al., 2015) with the exception of some recent developments (Guerrero et al., 2015; Lebel et al., 2013; Bodin et al., 2014; Treml et al., 2015). This paper addresses some aspects of this scholarly gap by providing an analytical framework for examining both ecological-institutional fit and the institutions potentially important in addressing lack of fit. We begin by outlining the analytical framework in more detail, including a brief discussion of OA from a social-ecological system perspective. Next we describe how a systems approach can be used to analyze ocean acidification-related governance for purposes of developing a policy resolution across sectors and ecosystems. We then elaborate the methodology and present the results in two parts; analysis of fit

followed by analysis of agency involvement. We end with a discussion of the areas of poorest institutional fit and specific measured legal gaps needing policy attention now and provide some methodological reflections highlighting advantages and limitations of the approach. 2. Approach and methodology The broad spectrum of drivers behind OA and human and environmental impacts suggests that an analysis of institutional fit and gaps needs to be based on a systems conceptualization of the problem. The benefit of the systems approach employed here (and described below) is that it identifies linkages between key relevant elements in the social-ecological system reflected in existing policy documents. Such links (or rather the legal documents representing them) may not currently be aimed at addressing OA (e.g. links between fisheries and tourism) but could nonetheless be a good starting points for tackling gaps in OA governance (Billé et al., 2013). Because social-ecological systems are not all the same any analysis must be tailored to the specific geographic region in focus. To allow us to make scientifically grounded conclusions we focus on the OA impacts related primarily to shelled mollusks and the potential repercussions and responses generally expected across the United States. Our motivation is that most biological evidence thus far points to shelled mollusks (larval stage) being severely affected by ocean acidification (Kroeker et al., 2013; Waldbusser et al., 2013; Talmage and Gobler, 2010). Several regions in the US are already seeing the effects of increased acidification, especially along the West and Northeast coast. Multi-million dollar losses are projected with repercussion for both ecosystems and livelihoods in the US (Kennedy, 2009; Cooley and Doney, 2009), and worldwide (Cooley et al., 2009; Brander et al., 2012; Narita et al., 2012; Narita and Rehdanz, 2016; Fernandes et al., 2016), making this a salient case to showcase our analysis. The analytical framework has three parts; 1) building a model of a social-ecological system (SES) affected by a particular environmental change phenomenon, in this case OA; 2) identifying policy gaps associated with OA. This second step builds on methods developed by Ekstrom and Young (2009) developed in the context of marine policy and ecosystem based management (EBM). We show how this method can also be applied for other policy issues, such as OA, and how it can be used to assess regulatory development over time. 3) Thirdly we estimate which government organizations are responsible for implementing the regulatory instruments uncovered in step 2. This allows identification of the most likely barriers to institutional change by highlighting which organizations and legal instruments are currently involved and which may need to be involved in coordinating efforts to effectively prevent or alleviate further ocean acidification. Step 3 aims to provide a baseline of agency jurisdiction for the modeled SES system to assess the potential for filling the legal gaps through existing jurisdictions. Our focus is purely on jurisdictional and statutory issues, such as planning and mitigation, and as such does not address the issue of where future research funding should be directed for OA. 2.1. A framework for analyzing institutional fit and agency involvement Fig. 1 outlines the proposed framework for analyzing institutional fit and agency involvement and its effects on a particular ecosystem. The analysis consists of three steps, each outlined in more detail below. 2.1.1. Step1: constructing a social-ecological systems model The first step is to construct a social-ecological system model relating to, in this case, OA. The boundaries delineating our system of interest encompass both the sources of OA (carbon dioxide emissions, agriculture, urban development) all the way to expected responses of communities engaged in shelled mollusk fisheries and aquaculture. The utility of this systems conceptualization approach has been stressed by scholars aiming to integrate systematic and holistic analyses of sector-


601

Fig. 1. Framework for analyzing institutional fit and agency involvement in relation to global environmental change phenomena. The framework consists of three interrelated parts. The systems model forms the basis of identifying fit and areas of misfit, then followed by examining areas of misfit (legal gaps) for the potential existing agency involvement. For each part, the main research questions addressed are listed in bold, and the methodological approach outlined in bullets below.

based management problems (Juda and Hennessey, 2001). One type of conceptualization often used to evaluate environmental problems from a systems perspective is the Driver-Pressure-State-Impact-Response (DPSIR) model (see e.g. OA for the Scotian Shelf by (Curran and Azetsu-Scott, 2012; Eurostat, 1999). We represent this using six components (A) Drivers; (B) Pressure; (C) State; (D) Biological Impact; (E) Human Impact; and (F) Human Response (Fig. 2). Each component can be subdivided into multiple sub-components (Table S2) each represented by a term (e.g. fossil fuel, agriculture, etc., see Fig. 2b) and each potentially interacting with one or more other sub-components in the defined system (Fig. 2a, b). A total of 22 sub-components were initially identified through open-ended interviews with experts in relevant fields (biology, geochemistry, oceanography, ecology, anthropology) (see Ekstrom, 2008). These were subsequently reviewed to reflect the most updated scientific understanding of OA. Based on these interviews a symmetrical matrix of sub-components by sub-components was created with each cell indicating the existence (1), or not (0), of a direct linkage between the system's sub-components (Fig. 2b). The system can thus also be analyzed at a smaller scale by evaluating the proportion of sub-component linkages within and between the larger components. As with geographic mapping, small-scale evaluation of data is useful for providing broad overviews of a large area or dataset and can help detect patterns. Our model is a conceptual one aiming to capture key elements of importance to understanding and addressing OA. It is not meant to reflect social-ecological dynamics in detail, and thus directionality was not included because this cannot be evaluated in the subsequent text analysis. Following the DPSIR order the first category (principal Drivers) refers to the burning of fossil fuels and local sources contributing to OA. Emissions caused by fossil fuel consumption and the resulting increase in concentration of atmospheric CO2 create the Pressure of the mounting uptake of inorganic carbon into the ocean. Local pressures now recognized as exacerbating OA also make up this component (Boehm et al.,

2015; Gobler et al., 2014; Melzner et al., 2012; Wallace et al., 2014) and include nutrient runoff from land-based discharge (Sunda and Cai, 2012), upwelling (Boehm et al., 2015; Fabry et al., 2008; Feely et al., 2010; Sunda and Cai, 2012) and low aragonite river water discharge (Salisbury et al., 2008; Waldbusser and Salisbury, 2014). State in our case is the reduced pH and carbonate availability in the ocean, scientifically referred to as ‘acidification’ (Caldiera and Wickett, 2003). This sequestration changes the chemistry of the ocean surface water, eventually producing a decrease in carbonate concentration (Doney et al., 2009; Orr et al., 2005). Biological research has shown that this reduction in carbonate saturation can harm marine biota (Biological Impacts), especially the larvae of organisms that use carbonate to develop shells, e.g. shelled mollusks, such as clams, mussels, oysters and other bivalves (Kroeker et al., 2013; Waldbusser et al., 2015). Such biological effects in turn have negative implications for capture fisheries, aquaculture and possibly also tourism (Ekstrom et al., 2015; Mathis et al., 2015) (i.e. Human Impacts). The Response component includes responses which are primarily aimed at dealing with the effects of OA. Reducing the sources of global OA and sources that locally amplify regional conditions are also response strategies to reduce or mitigate OA impacts. These strategies are addressed in our analysis through the Drivers and Pressures components, which is consistent with Curran and Azetsu-Scott (2012). Duplicating those as part of the Response component would therefore constitute a form of unnecessary redundancy in the textual analysis presented below. We indicate this feedback of mitigation measures to divers (as a human response) through the dotted arrow in Fig. 2. 2.1.2. Step 2: analysis of institutional fit This step identifies gaps in legal instruments in place to address OA and relies on a form of text analysis of legal documents developed in Ekstrom and Young (2009). It is based on the idea that management institutions should reflect the nature and functionality of a relevant

602


Fig. 2. From conceptualization of ocean acidification to network representation of the system. Panel (a) is a conceptual diagram of the ocean acidification problem showing the acidification process from the source of carbon emitters through the predicted impacts on organisms and people. Letters correspond to the components in the DPSIR model linking the source of OA to the projected impacts and response. Panel (b) is the matrix display of the network of DPSIR components. Each term under the respective component represents a sub-component. Outlined with bold lines, linkages are organized into ‘blocks’ based on their respective components; Panel (c) is another way of visualizing the inter-component linkages (gray cells in matrix) with each node representing a modeled component (D. Drivers; P. Pressure; S. State; BI. Biological Impact; HI. Human Impact; R. Human Response).

ecosystem (Costanza and Folke, 1996; Folke et al., 2007; Young, 2002a). Law is the primary medium by which government agencies exercise control across multiple sectors so by testing if relationships between system components are represented or not in legal documents the gap analysis reveals whether formal regulation and law potentially address the environmental issue at hand. While we recognize that such gaps can sometimes be filled by informal arrangements and ‘de facto’ regulations our analysis is limited to legal documents. A ‘measured’ legal gap in ocean governance is defined here as a situation where laws and regulations do not address a linkage between two sub-components of a system. We assess the existence of linkages between sub-components both within and between components, recognizing that we can only determine the presence of a potential recognition of the link in law. Furthermore, the SES model does not include direction of links given that the text analysis cannot determine directionality of the links. Linkages can include interactions among species and/or habitats, or with

biophysical conditions and the sources, human stressors, or with human responses to the conditions. The analysis consists of three steps: a. Creating a term-by-document matrix of laws to produce a law network, where sub-components of the modeled system are represented as terms in law (the compilation of laws evaluated is described in Table S1 and terms queried are listed in Table S2). Linkages in the law network (from hereon referred to as “observed legal linkages”) are instances of co-occurrences between sub-components in the systems model in legal texts. These include connections across and within the components.

Data was generated by reviewing legal documents of federal law. Those referring to at least one key term representing any sub-


component were included in the analysis resulting in a total of 505 legal documents (319 regulatory, 186 statutory) for 2013 and 725 legal documents (669 regulatory, 56 statutory) for 2006. Term-by(legal) document matrices were created based on co-occurrence of subcomponents summed across all documents for each of the two years using MINOE (v.1.1) (Ekstrom et al., 2010a). Cells in these matrices were counts representing the total number of co-occurrences found in the document collections searched for each year, as demonstrated to be most effective in Ekstrom et al., 2010b. For more details on the matrix creation see Supplementary materials. b. Identifying links between sub-components modeled in the system but not accounted for in law (“measured gaps”). The co-occurrence matrix, structured the same as the SES model matrix (Fig. 2b), was compared to the latter to identify institutional gaps and potential fit between institutions and the SES. These gaps are referred to as measured legal gaps to indicate that they are derived from text analysis, not qualitative legal analysis. c. Calculating the measure of fit (M), which is the ratio of the sum of observed co-occurrences and the total number of modelled SES links. A high score, therefore, indicates a higher fit between the institutions and the conceptually modeled OA SES. M score represents the proportion of the perceived scientific linkages that are captured by co-occurrences in the policy documents. This does not indicate whether policy targets of these documents are filled. Additionally, this metric reveals potential fit rather than absolute fit as the ‘legal linkages’ represented by co-occurrences are derived from text analysis of laws and regulations rather than interpretative, qualitative analysis.

The fit metric, M, is assessed for subsets of the matrices (‘blocks’) for a finer scale analysis of measured gaps between and within components (see Supplementary material for more detail and sub-component fit assessment and M score interpretation). A block is the set of linkages between two components or a set of linkages within a single component. Blocks, indicated by bold outlines in the matrix view of the conceptual SES model in Fig. 2b, are sets of linkages. This provides an organized way to look in more detail (than the whole system view) at where the measured legal gaps are. For example, one ‘block’ is made up of 15 modelled linkages between subcomponents in Pressure and State, whereas there are three linkages that make up the block of linkages between the Pressures and Biological Impacts. To assess changes in institutional fit over time the same analysis is run on data from multiple points in time to evaluate any change in fit over time. In this study this was done for law compilations related to OA from 2006 and 2013. The year 2006 was chosen as a starting point since OA largely emerged as a problem in the public sphere after the publication of a report by the Royal Society in 2005 (Raven et al., 2005). 2.1.3. Step 3: analysis of agency involvement As in step 2, the approach to estimate agency involvement within the SES model is based on text analysis of legal documents. It identifies what agencies are formally involved in the relevant topics1 and which agencies (empowered by which laws) would need to coordinate to address the social-ecological linkages identified in the systems model by calculating the degree of agency involvement (Agency Involvement Measure, AIM) using the key term frequency in documents associated with each agency (see Supplementary material for more detail). The measure is composed of raw counts of the total number of key terms in the set of documents that are under the responsibility of a given agency. For example, laws under the Environmental Protection Agency (USEPA) contain the term pH nearly 2000 times in the laws examined, 1 A topic can be either an individual modeled sub-component or a larger component composed of sub-components.

603

whereas pH is mentioned fewer than 50 times by laws under the Department of Commerce (US-DOC). Although this frequency does not translate directly into legal definitions of jurisdictions, it does provide a semi-quantitative and objective metric as a first step for approximating what agencies are involved in what topics thus revealing areas of potential functional overlap (Ekstrom et al., 2009, 2010b). Functional overlap across agencies at one institutional level (US federal) occur “when two or more statutes or regulations separately cover intersecting activities” (Ekstrom et al., 2009). We use this relative measure of agency responsibility (AIM) as a way to identify what agencies could coordinate to fill existing gaps, and to provide initial guidance in identifying areas of potentially problematic interplay if gaps are filled haphazardly, without strategic incorporation of all agencies with existing legal jurisdiction over the linked sub-components. 3. Results Results are presented in the following order: (1) calculation of institutional fit scores (M) for the SES model and legal documents from 2006 and 2013 at the scale of the whole system and for system subcomponents; (2) assessment of improvement of institutional fit over time and identification of remaining measured legal gaps; (3) analysis of measured agency involvement surrounding the gaps identified in the previous step. 3.1. Degree of institutional fit There are 95 modeled linkages (gray cells Fig. 2b, listed in Table S4) between sub-components in our system model (Fig. 2b). The degree of fit evaluates how many of the modeled linkages between subcomponents are accounted for in the laws. The degree of fit (M) of 2006 data was 0.31 on a scale of 0–1, where 1 is the maximum possible fit (see Table S3, Supplementary material, for scale interpretation). Analysis of fit using 2013 laws returned a score of 0.38, thus indicating a slight improvement in fit between institutions and modeled SES system linkages over time at the level of the whole system. Next we evaluated fit at the scale of individual ‘blocks’ of the SES model (see Fig. 2b where bolded outlines surround each block). At this scale, fit was lowest for blocks comprising relationships associated with human response to OA, all of which received M scores of poor fit in both years (≤0.33) (Fig. 3). Areas of notably high institutional fit are sub-component linkages within the Human Impacts component (i.e. impacts on humans) and between the Human Impacts and Biological Impacts, the latter of which improves by 2013 leaving no gaps (M = 1.0). 3.2. Improvement of (sub-component) institutional fit over time and remaining legal gaps Based on the text analysis, fifty percent (8 of 16) of the blocks had higher M scores in 2013 than in the 2006 laws and seven of the measured gaps between sub-components in 2006 were filled by 2013 (Fig. 3). Three of the blocks showing improved potential fit are associated with Drivers, and four are associated with State. Other notable changes include the link between ocean acidification and pH (sub-components of State), acknowledged by 2013, but not in 2006. Similarly carbonate saturation state and ocean acidification are both acknowledged (sub-components within State), neither of which were addressed in 2006. This arguably reflects the increasing scientific understanding of the issue. The lack of measured links between pH-monitoring and ocean acidification-monitoring (State and Response sub-components) in 2006 is also filled in the 2013 laws. Based on the text analysis of co-occurrences, the measure of overall institutional fit improves between the year 2006 and 2013; however, the text analysis reveals that some measured gaps persist. For the 2013 data set we found 69 measured legal gaps. Some of these are

604


Fig. 3. Degree of fit based on laws from 2006 (a) and 2013 (b). Overall network fit (M) scores are noted next to each year. Network diagrams show M score results of block-based fit analysis. Each node represents a modeled component: D. Drivers; P. Pressure; S. State; Ib. Biological Impact; Ih. Human Impact; R. Human Response. Degree of institutional fit for intercomponent connections is represented by color and thickness of the lines connecting components. Using the same format for line representation outlines of nodes indicate the degree of fit (M) for links within components (blocks).

especially noteworthy and of more concern than others. For example, measured gaps involving sub-components ‘ocean acidification’ and ‘carbonate saturation’ persist in both 2006 and 2013 laws. Modeled linkages connecting OA (State) to Human Impacts are largely absent from law, creating a poor fit in this block. Modeled linkages connecting the local drivers to the problem of OA (Pressure-State) are also absent. Another example is the absence of any measured legal linkage between the sub-component of eutrophication (State) and any of the subcomponents pH, ocean acidification or saturation state (also State variables). This shows a measured legal gap in acknowledging the relationship and role of eutrophication as part of exacerbating OA or altering of seawater pH even though this has been acknowledged as potentially important for OA effects (Cai et al., 2011). Other gaps of concern include the lack of acknowledged links between carbon dioxide emissions (Pressure) and pH, OA, or saturation state (State) – again showing a complete absence of acknowledging the driving pressure behind OA. Finally, the low presence of key terms discussing human responses (adaptation measures) to ocean acidification should be noted. 3.3. Assessing agency involvement Identifying gaps in the institutional fit provide a springboard for future action. The next stage involves investigating how to fill the existing gaps in governance. As a first step, the agency involvement analysis aims to identify the government agencies currently involved in legislation associated with each of the modeled system components. Thus it should not be seen as a guideline for decisions regarding research allocation relating to OA. Of the 32 federal government departments and agencies investigated in this analysis, 14 agencies were involved in one or more of the modeled system components investigated. Fig. 4 presents the agency involvement for each individual sub-component. For example, Fig. 4a contains the keyword count for each sub-component of the Driver category. Each bar represents an agency; the height represents the agency’s estimated involvement in the component (based on term frequency), and bar colors represents the individual sub-components. Across all DPSIR components three federal departments stand out in terms of involvement; the US Department of Agriculture (USDA), the

Environmental Protection Agency (EPA), and the Department of Commerce (DOC) (in decreasing order of involvement). The USDA's primary involvement is in agriculture, which is a sub-component in the Driver component of the SES model. The EPA's involvement is more diversified among sub-components of different components of the SES system. The agency dominates the Pressure and State components and shares nearly equal involvement with the DOC in the responsibility for laws that deal with the Biological Impact). Besides its strong (but secondary) involvement in the Biological Impact component, DOC is the primary government agency involved in the laws relating to the sub-components comprising the Human Impact and Human Response to OA. To identify which agencies may be best placed to address remaining measured legal gaps we start by examining the areas of critical concern in the block-based analysis of institutional fit from 2013 laws (Fig. 3, and further described in Section 3.2). These included measured legal gaps in the blocks of: Response – Biological Impact, and Human Impact – State, Pressure –State, and within State (Fig. 5). As shown in Fig. 5 for the Pressure-State block (where the model SES connects CO2 emissions, nutrients, upwelling and river outflow to OA) the EPA is the primary agency involved in both of these components, and to a lesser degree the USDA who is primarily involved in watershed and nutrient sub-components of the Pressure component. This suggests the EPA may have the legal jurisdiction to cover the measured gaps in law and thus be the singularly best placed agency to come up with a strategy to fill the Pressure-State gaps identified in our analysis. For the Human Impact – State block, the DOC is largely involved in the Human Impact component, whereas the EPA is largely involved in the State component. Therefore, this points to the likely need for coordination between these two agencies in order to improve the fit in this area. The very low fit within the State block at first suggests a need for mainly the EPA to improve fit. However, upon closer examination of agencies involved in each sub-component, a different interpretation arises. The DOC is most strongly involved (based on term count) in eutrophication, whereas the EPA is more heavily involved in pH. For the Response-Biological Impact block several agencies appear to be involved in the management of both components, creating a potentially more complicated coordination scheme to fill the potential legal gaps.


605

Fig. 4. Agency Involvement Measure (AIM) in federal law (year 2013) for each modeled component. Each chart contains the AIM of each of the DPSIR components (A. Driver; B. Pressure; C. State; D. Impact (biological); E. Impact (human); and F. Human response). Colors correspond to those used for components in Figs. 2 and 3. For acronyms see Supplementary material.

To explore what agencies could fill the individual measured legal gaps, we use the same method as for the block-based results, but look at the agency involvement calculated for individual sub-components rather than for the sum across entire components. This more fine scale analysis of the measured legal gaps and the agencies involved can help avoid misinterpretation of the agency involvement results, which can occur when analyzed at the level of entire components. For example, as noted above the overall view of the State component’s agency involvement (Fig. 4c) shows the EPA as the primary involved agency. However, when viewing just a single sub-component within State, ‘eutrophic’, we find that the DOC and DOI are relatively more highly involved (based on term counts) than the EPA, while for subcomponent ‘ocean acidification’ the DOC is the only agency with a law that accounts directly for this issue. 4. Discussion Ocean chemistry confirms what will happen if we continue on our business-as-usual path of fossil fuel burning and other carbon emitting activities (Beaufort et al., 2011; Doney et al., 2009; Fabry et al., 2008; Gruber et al., 2012; IPCC, 2014; Orr et al., 2005). A holistic ecosystembased approach is crucial to tackling the multi-scale and multi-sector environmental problem like ocean acidification (Cash and Moser,

2000; Pahl-Wostl, 2009; Young et al., 2007; Klinger and Newton, 2016).This case study introduces an analysis of institutional fit and agency involvement and applies it to an emerging global environmental problem. The analysis is grounded in two key challenges identified as crucial for achieving robust governance regimes (Folke et al., 2007; Vatn and Vedeld, 2012); the problem of fit between institutions and the ecosystem they are designed to govern (Galaz et al., 2008; Young, 2002a), and the challenge occurring when one institution interacts with and affects the ability of other another institution to function (Young et al., 2008). Our analysis of fit based on 2013 data showed an overall institutional fit score of 0.38. This was an improvement from 2006 (0.31) but shows that many potential legal gaps remain and that institutional alignment with social-ecological processes is still far from optimal. The increasing institutional fit observed over time is primarily related to processes contained within the Driver and State components. This is largely the result of the FOARAM Act (2009), and the recent US carbon emission reduction laws and executive order related to climate change. As such it is a good example of how the analysis captures human responses to OA which feedback to affect primarily drivers. While the overall increase in fit in 2013 is not dramatic, it shows a growing acknowledgement of connections between processes that were not previously addressed. The remarkably high fit for connections within the Human Impacts

606


Fig. 5. Areas of poor institutional fit and agencies involved in each relevant component. The network diagram is the DPSIR framed system model based on results from the 2013 laws (from Fig. 3b) Agency Involvement graphs show the agencies involved in individual components making up the component dyads exhibiting the poorest institutional fit.

component and between the Human Impacts and Biological Impacts is also noteworthy and encouraging and may indicate a promising policy development in terms of adaptations to ocean acidification. Identifying specific gaps in how existing institutions acknowledge key sub-component linkages in the SES model is a way to highlight areas needing policy attention to increase the institutional fit (Folke et al., 2007; Galaz et al., 2008). Regarding how OA is addressed at the federal level in the US, the key measured legal gaps remaining in 2013 (Table S4) include (i) Links between (human) Response and Biological Impact; (ii) Links connecting local drivers to the problem of OA (linkages within State component and those between Pressure and State components); and (iii) Links between OA (State) and expected Human Impact; and (iv) No formal institutional arrangement even recognized the process of ‘ocean acidification’ by 2006 which is remarkable. However, by 2013 sub-components ‘ocean acidification’ and ‘carbonate saturation state’ do show up in law, showing improvement in the formal recognition of these concepts over time. Identifying potential gaps in the institutional fit between laws and SES interactions is a first critical step in understanding a national governance landscape in place to address OA, and provides a basis for understanding how to fill these gaps to improve governance. It is critical that new policies or laws are developed within the context of existing legislation in order to avoid creating more of the unintended overlaps already typical of the fragmented system of governance (Biermann et al., 2009; Ekstrom et al., 2009; Zelli and van Asselt, 2013). The measured agency involvement analysis presented here identifies agencies involved in legislation associated with each of the modeled system components (and sub-components). This analysis can provide policymakers with a preliminary indication of what agencies currently have legal jurisdiction over relevant processes and also which agencies could (and potentially should) be involved in developing strategies to fill remaining legal gaps relevant to ocean acidification. As such it could provide a first

step in understanding institutional interplay which is important to account for when adding new regulations or policy issues to an already crowded institutional space (Young, 2002b). Our analysis of agency involvement shows that three agencies (USDA, EPA, and DOC) are most frequently involved in regulations pertinent to the OA DPSIR model. However, as shown above, analysis of agency involvement at the scale of the whole SES (examining only the DPSIR components) can be misleading. Analysis of sub-components provides a more nuanced understanding of agency involvement, revealing that sub-components within State are not dealt with by the EPA alone, but also by the DOC and DOI. When two sub-components (or components) are managed by very different sets of agencies, this may indicate a coordination challenge that needs to be addressed to fill the gap. The examples of low institutional fit in the OA system explored here demonstrate that in some cases it is more useful to rely on agency involvement from specific sub-components, whereas in other cases the component-based measure of involvement is appropriate and useful. In summary, this study shows that while ocean acidification is finally recognized in US law, it is only sparingly so and given the risks this environmental threat creates for both ecosystems and human communities dependent on them, more action needs to be taken at the federal level, especially involving different management sectors (land-based water quality, air quality, fisheries). So far the main improvements are apparent in the increased laws to reduce global carbon emissions and the beginning of monitoring OA. However, the connection of OA to more local drivers is something that is clearly absent from law thus far, yet the links between the DPSIR model elements (and its subcomponents) currently recognized in law constitute an obvious and likely salient avenue to address OA governance at the local level. As top-down policy is often unfeasible due to political deadlocks, bottom-up initiatives at sub-national scales can be a way to generate the public awareness and consequently political momentum needed to confront


environmental issues such as ocean acidification in a holistic manner (Ostrom, 2010). Efforts initiated on the west coast and Gulf of Maine (California and North East Coastal Acidification Networks, C-CAN and NE-CAN respectively) exemplify such ground-up efforts in early stages. Whether these types of efforts and other state-level work can fill federal-level legal gaps, as suggested by some (e.g. Kelly and Caldwell, 2012; Kelly et al., 2011; Strong et al., 2014), may depend on the disciplinary interest and OA vision of those leading the initiatives. Though portions of the findings of this analysis may not be surprising to environmental policy experts (e.g. no law addressed ocean acidification specifically in 2006, but does now), the contribution of these findings is twofold. First, the analysis brings a semi-quantitative dimension to comprehensive legislative analysis and can be applied to any alternative conceptual human-environmental system of interest to a researcher or decision-maker. For example, one may add components or even new categories to the system model related to ocean acidification, components such as temperature and other climate changerelated feedbacks with a role in ocean acidification. Second, the gaps analysis presented in this paper reveals results consistent with governance literature verifying the utility of the approach. It is possible that the technique used in this paper could be even more instrumental in finding gaps in a system with less obvious gaps than those revealed for ocean acidification. 4.1. Methodological reflections While the methods showcased here demonstrate the power of text mining large numbers of legal documents to generate information about institutional fit and potential legal gaps, agency involvement and legal jurisdiction related to ocean acidification some reflections on methodological possibilities and limitations are warranted. First, not all management can be deduced from analysis of laws, and not all laws are implemented equally across jurisdictions. There is also a risk of capturing false positives and false negatives as the text analysis relies on presence/absence of key terms coinciding. Finally, we included only links (between the DPSIR model components) for which there is wellestablished scientific evidence. We recognize that this excludes potential responses which have not yet been fully documented by the scientific community and as such risks underestimating potential legal gaps. The approach is therefore not meant to replace qualitative examination and interpretation of law, but should rather be seen as a first step which can guide the direction of subsequent qualitative analysis (further limitations explored in Ekstrom et al., 2010b). We also recognize that misleading results could originate from problems related to more fragmented decision-making within individual agencies than imagined, and from high fragmentation of laws with only a few of the laws being implemented. These issues could be addressed through strategic discussions between policymakers about a select group of topics. Furthermore, analyses presented address only a fraction documents that could generate baseline governance data, neglecting e.g. budget allocations, spatial delineation of relevant laws and regulations, court cases, meeting notes, management plans, memorandums of agreement, and many others. This would provide a more qualitative evaluation of how the ecosystem component links are de facto addressed above and beyond legal instruments. Despite these limitations, and given the ease with which the textual analysis can be done and the surprisingly rich data on existing legal linkages it can provide (and hence also the potential gaps), we believe it can provide a good and efficient tool to flag up possible institutional misfit that needs further attention by both policy-makers and the academic community, for OA as well as other policy domains. Acknowledgements The 2006 ocean acidification fit portion of the analysis was originally part of JE's dissertation, chaired by Professor Oran Young and funded by

607

the California Council for Environmental Quality Initiative, California Sea Grant, Ocean Protection Council and the National Center for Ecological Analysis and Synthesis. We appreciate Hannah Torres, student at Florida State University, for helping compiling agency authority for 2013 statutes. Additionally, we thank Daniel Spiteri who helped clean and process the laws. This work was presented at the International Human Dimensions Program conference (IHDP) 2009 in Bonn (travel supported by IHDP) and at Resilience 2014 in Montpellier (travel supported by Natural Resources Defense Council). The input of BC was made possible by funding from the Erling-Persson Family Foundation through the Global Economic Dynamics and the Biosphere Program at the Royal Swedish Academy of Sciences. Appendix A. Supplementary data Supplementary data to this article can be found online at doi:10. 1016/j.scitotenv.2016.10.114. References Barnes, C., McFadden, K.W., 2008. Marine ecosystem approaches to management: challenges and lessons in the United States. Mar. Policy 32, 387–392. Beaufort, L., Probert, I., de Garidel-Thoron, T., Bendif, E.M., Ruiz-Pino, D., Metzl, N., Goyet, C., Buchet, N., Coupel, P., Grelaud, M., Rost, B., Rickaby, R.E.M., de Vargas, C., 2011. Sensitivity of coccolithophores to carbonate chemistry and ocean acidification. Nature 476:80–83. http://dx.doi.org/10.1038/nature10295. Biermann, F., 2015. Earth system governance. Encycl. Glob. Environ. Gov. Polit. Biermann, F., Pattberg, P., van Asselt, H., Zelli, F., 2009. The fragmentation of global governance architectures: a framework for analysis. Glob. Environ. Polit. 9:14–40. http:// dx.doi.org/10.1162/glep.2009.9.4.14. Billé, R., Kelly, R., Biastoch, A., Harrould-Kolieb, E., Herr, D., Joos, F., Kroeker, K., Laffoley, D., Oschlies, A., Gattuso, J.-P., 2013. Taking action against ocean acidification: a review of management and policy options. Environ. Manage. 52:761–779. http://dx.doi.org/10. 1007/s00267-013-0132-7. Bodin, Ö., Crona, B., Thyresson, M., Golz, A.-L., Tengo, M., 2014. Conservation success as a function of good alignment of social and ecological structures and processes. Conserv. Biol. 28:1371–1379. http://dx.doi.org/10.1111/cobi.12306. Boehm, A., Jacobson, M., O’Donnell, M., Sutula, M., Wakefield, W., Weisberg, S., Whiteman, E., 2015. Ocean acidification science needs for natural resource managers of the North American west coast. SCCWRP 867. Boyd, P.W., 2011. Beyond ocean acidification. Nat. Geosci. 4:273–274. http://dx.doi.org/ 10.1038/ngeo1150. Brander, L.M., Rehdanz, K., Tol, R.S.J., Van Beukering, P.J.H., 2012. The economic impact of ocean acidification on coral reefs. Clim. Chang. Econ. 03:1250002. http://dx.doi.org/ 10.1142/S2010007812500029. Brown, K., 2003. Integrating conservation and development: a case of institutional misfit. Frontiers in Ecology and the Environment 1 (9), 479–487. Caldiera, K., Wickett, M.E., 2003. Anthropogenic carbon and ocean pH. Nature 425, 365. Cai, W.-J., Hu, X., Huang, W.-J., Murrell, M.C., Lehrter, J.C., Lohrenz, S.E., Chou, W.-C., Zhai, W., Hollibaugh, J.T., Wang, Y., Zhao, P., Guo, X., Gundersen, K., Dai, M., Gong, G.-C., 2011. Acidification of subsurface coastal waters enhanced by eutrophication. Nat. Geosci. 4, 766–770. Cash, D., Moser, S., 2000. Linking global and local scales: designing dynamic assessment and management processes. Glob. Environ. Chang. 10, 109–120. Cooley, S.R., Doney, S.C., 2009. Anticipating ocean acidification’s economic consequences for commercial fisheries. Environ. Res. Lett. 4:24007. http://dx.doi.org/10.1088/ 1748-9326/4/2/024007. Cooley, S., Kite-Powell, H., Doney, S., 2009. Ocean acidification's potential to alter global marine ecosystem services. Oceanography 22:172–181. http://dx.doi.org/10.5670/ oceanog.2009.106. Costanza, R., Folke, C., 1996. The structure and function of ecological systems in relation to property-rights regimes. In: Hanna, S.S., Folke, C., Maler, K.-G. (Eds.), Rights to Nature: Ecological, Economic, Cultural, and Political Principles of Institutions for the Environment. Island Press, Washington, D.C., pp. 13–34. Curran, K., Azetsu-Scott, K., 2012. Ocean acidification. State of the Scotian Shelf. Atlantic Coastal Zone Information Steering Committee, Dartmouth, NS. Dimitrov, R.S., Sprinz, D.F., DiGiusto, G.M., Kelle, A., 2007. International nonregimes: a research agenda. Int. Stud. Rev. 9:230–258. http://dx.doi.org/10.1111/j.1468-2486. 2007.00672.x. Doney, S.C., Fabry, V.J., Feely, R.A., Kleypas, J.A., 2009. Ocean acidification: the other CO2 problem. Ann. Rev. Mar. Sci. 1:169–192. http://dx.doi.org/10.1146/annurev.marine. 010908.163834. Ekstrom, J.A., 2008. Navigating Fragmented Ocean Law in the California Current: Tools to Identify and Measure Gaps and Overlaps for Ecosystem-Based Management. University of California, Santa Barbara, Doctoral Thesis. Ekstrom, J.A., Young, O.R., 2009. Evaluating functional fit between a set of institutions and an ecosystem. Ecol. Soc. 14. Ekstrom, J.A., Young, O.R., Gaines, S.D., Gordon, M., McCay, B.J., 2009. A tool to navigate overlaps in fragmented ocean governance. Mar. Policy 33:532–535. http://dx.doi. org/10.1016/j.marpol.2008.11.007.

608


Ekstrom, J.A., Lau, G.T., Spiteri, D.J., Cheng, J.C.P., Law, K.H., 2010a. MINOE: a software tool to analyze ocean management efforts in the context of ecosystems. Coast. Manag. 38: 457–473. http://dx.doi.org/10.1080/08920753.2010.498400. Ekstrom, J.A., Lau, G.T., Cheng, J.C.P., Spiteri, D.J., Law, K.H., 2010b. Gauging agency involvement in environmental management using text analysis of laws and regulations. I/S - J. Law Policy Inf. Soc. 6, 189–210. Ekstrom, J.A., Suatoni, L., Cooley, S.R., Pendleton, L.H., Waldbusser, G.G., Cinner, J.E., Ritter, J., Langdon, C., van Hooidonk, R., Gledhill, D., Wellman, K., Beck, M.W., Brander, L.M., Rittschof, D., Doherty, C., Edwards, P.E.T., Portela, R., 2015. Vulnerability and adaptation of US shellfisheries to ocean acidification. Nat. Clim. Chang. 5:207–214. http:// dx.doi.org/10.1038/nclimate2508. Epstein, G., et al., 2015. Institutional fit and the sustainability of social-ecological systems. Curr. Opin. Environ. Sustain. 14, 34–40. Eurostat, 1999. Towards Environmental Pressure Indicators for the EU - Environment and Energy. EU Bookshop n.d. Fabry, V.J., Seibel, B.A., Feely, R.A., Orr, J.C., 2008. Impacts of ocean acidification on marine fauna and ecosystem processes. ICES J. Mar. Sci. 65, 414–432. Feely, R.A., Alin, S.R., Newton, J., Sabine, C.L., Warner, M., Devol, A., Krembs, C., Maloy, C., 2010. The combined effects of ocean acidification, mixing, and respiration on pH and carbonate saturation in an urbanized estuary. Estuar. Coast. Shelf Sci. 88: 442–449. http://dx.doi.org/10.1016/j.ecss.2010.05.004. Fernandes, J.A., Papathanasopoulou, E., Hattam, C., Queirós, A.M., Cheung, W.W.W.L., Yool, A., Artioli, Y., Pope, E.C., Flynn, K.J., Merino, G., Calosi, P., Beaumont, N., Austen, M.C., Widdicombe, S., Barange, M., 2016. Estimating the ecological, economic and social impacts of ocean acidification and warming on UK fisheries. Fish Fish. http://dx.doi. org/10.1111/faf.12183Fernandez et al. 2016. FOARAM, 2009. Federal Ocean Acidification Research and Monitoring Act (FOARAM) (USA). Folke, C., Pritchard, L., Berkes, F., Colding, J., Svedin, U., 2007. The problem of fit between ecosystems and institutions: ten years later. Ecol. Soc. 12, 30. Galaz, V., Hahn, T., Olsson, P., Folke, C., Svedin, U., 2008. The problem of fit among biophysical systems, environmental and resource regimes, and broader governance systems: insights and emerging challenges. Institutions and Environmental Change Principal Findings, Applications and Research Frontiers. MIT Press, Cambridge, MA. Galaz, V., Crona, B., Österblom, H., Olsson, P., Folke, C., 2012. Polycentric systems and interacting planetary boundaries — emerging governance of climate change–ocean acidification–marine biodiversity. Ecol. Econ. 81:21–32. http://dx.doi.org/10.1016/j. ecolecon.2011.11.012. Galland, G., Harrould-Kolieb, E., Herr, D., 2012. The ocean and climate change policy. Clim. Policy 12:764–771. http://dx.doi.org/10.1080/14693062.2012.692207. Gobler, C.J., Depasquale, E., Griffith, A., Baumann, H., 2014. Hypoxia and acidification have additive and synergistic negative effects on the growth, survival, and metamorphosis of early life stage bivalves. PLoS One 9, e83648. Gruber, N., Hauri, C., Lachkar, Z., Loher, D., Frölicher, T.L., Plattner, G.-K., 2012. Rapid progression of ocean acidification in the California current system. Science 337:220–223. http://dx.doi.org/10.1126/science.1216773. Guerrero, A., Bodin, Ö., McAllister, R., Wilson, K., 2015. Achieving social-ecological fit through bottom-up collaborative governance: an empirical investigation. Ecol. Soc. 20. Harrould-Kolieb, E.R., Herr, D., 2012. Ocean acidification and climate change: synergies and challenges of addressing both under the UNFCCC. Clim. Policy 12:378–389. http://dx.doi.org/10.1080/14693062.2012.620788. IPCC, 2014. Climate Change 2014: Impacts, Adaptation, and Vulnerability; Volume 2: Regional Aspects. Juda, L., Hennessey, T., 2001. Governance profiles and the management of the uses of large marine ecosystems. Ocean Dev. Int. Law 32, 43–69. Kelly, R.P., Caldwell, M., 2012. Ten ways states can combat ocean acidification (and why they should). Harvard Environ. Law Rev. 37, 57–103. Kelly, R.P., Foley, M.M., Fisher, W.S., Feely, R.A., Halpern, B.S., Waldbusser, G.G., Caldwell, M.R., 2011. Oceans. Mitigating local causes of ocean acidification with existing laws. Science 332:1036–1037. http://dx.doi.org/10.1126/science.1203815. Kennedy, C., 2009. An Upwelling Crisis: Ocean Acidification (NOAA Clim). Kerr, R.A., 2010. Ocean acidification unprecedented, unsettling. Science 80 (328), 1500–1501. Kim, R.E., 2012. Is a new multilateral environmental agreement on ocean acidification necessary? Rev. Eur. Commun. Int. Environ. Law 21:243–258. http://dx.doi.org/10. 1111/reel.12000.x. Klinger, T., Newton, J., 2016. Ocean acidification as a problem in systems thinking. Wash. J. Environ. Law Policy 6, 208–212. Kroeker, K.J., Kordas, R.L., Crim, R., Hendriks, I.E., Ramajo, L., Singh, G.S., Duarte, C.M., Gattuso, J.-P., 2013. Impacts of ocean acidification on marine organisms: quantifying sensitivities and interaction with warming. Glob. Chang. Biol. 19:1884–1896. http:// dx.doi.org/10.1111/gcb.12179. Lebel, L., Nikitina, E., Pahl-Wostl, C., Knieper, C., 2013. Institutional fit and river basin governance: a new approach using multiple composite measures. Ecol. Soc. 18.

Mathis, J.T., Cooley, S.R., Lucey, N., Colt, S., Ekstrom, J., Hurst, T., Hauri, C., Evans, W., Cross, J.N., Feely, R.A., 2015. Ocean acidification risk assessment for Alaska’s fishery sector. Prog. Oceanogr. 136:71–91. http://dx.doi.org/10.1016/j.pocean.2014.07.001. Melzner, F., Thomsen, J., Koeve, W., Oschlies, A., Gutowska, M., Bange, H., Hansen, H., Körtzinger, A., 2012. Future ocean acidification will be amplified by hypoxia in coastal habitats. Mar. Biol. 160, 1875–1888. Narita, D., Rehdanz, K., 2016. Economic impact of ocean acidification on shellfish production in Europe. J. Environ. Plan. Manag. 1–19. http://dx.doi.org/10.1080/09640568. 2016.1162705. Narita, D., Rehdanz, K., Tol, R.S.J., 2012. Economic costs of ocean acidification: a look into the impacts on global shellfish production. Clim. Chang. 113:1049–1063. http://dx. doi.org/10.1007/s10584-011-0383-3. Orr, J.C., Fabry, V.J., Aumont, O., Bopp, L., Doney, S.C., Feely, R.A., Gnanadesikan, A., Gruber, N., Ishida, A., Joos, F., Key, R.M., Lindsay, K., Maier-Reimer, E., Matear, R., Monfray, P., Mouchet, A., Raymond, G., Najjar, R.G., Plattner, G.-K., Rodgers, K.B., Sabine, C.L., Sarmiento, J.L., Schlitzer, R., Slater, R.D., Totterdell, I.J., Weirig, M.-F., Yamanaka, Y., Yool, A., 2005. Anthropogenic ocean acidification over the twenty-first century and its impact on calcifying organisms. Nature 437. Ostrom, E., 2010. Nested externalities and polycentric institutions: must we wait for global solutions to climate change before taking actions at other scales? Econ. Theory 49: 353–369. http://dx.doi.org/10.1007/s00199-010-0558-6. Pahl-Wostl, C., 2009. A conceptual framework for analysing adaptive capacity and multilevel learning processes in resource governance regimes. Glob. Environ. Chang. 19, 354–365. Raven, J., Caldiera, K., Elderfield, H., Hoegh-Guldberg, O., Liss, P., Riebesell, U., Shepherd, J., Turley, C., Watson, A., 2005. Ocean Acidification Due to Increasing Atmospheric Carbon Dioxide. The Royal Society Policy Document, London. Salisbury, J., Green, M., Hunt, C., Campbell, J., 2008. Coastal acidification by rivers: a threat to shellfish? Eos. Trans. Am. Geophys. Union 89:513. http://dx.doi.org/10.1029/ 2008EO500001. Strong, A.L., Kroeker, K.J., Teneva, L.T., Mease, L.A., Kelly, R.P., 2014. Ocean acidification 2.0: managing our changing coastal ocean chemistry. Bioscience 64:581–592. http://dx. doi.org/10.1093/biosci/biu072. Sunda, W.G., Cai, W.-J., 2012. Eutrophication induced CO₂-acidification of subsurface coastal waters: interactive effects of temperature, salinity, and atmospheric PCO₂. Environ. Sci. Technol. 46:10651–10659. http://dx.doi.org/10.1021/es300626f. Talmage, S.C., Gobler, C.J., 2010. Effects of past, present, and future ocean carbon dioxide concentrations on the growth and survival of larval shellfish. Proc. Natl. Acad. Sci. USA 107 (17) (246–17 251). Treml, E.A., Fidelman, P.I., Kininmonth, S., Ekstrom, J.A., Bodin, Ö., 2015. Analyzing the (mis)fit between the institutional and ecological networks of the Indo-West Pacific. Glob. Environ. Chang. 31:263–271. http://dx.doi.org/10.1016/j.gloenvcha.2015.01. 012. Vatn, A., Vedeld, P., 2012. Fit, interplay, and scale: a diagnosis. Ecol. Soc. 17, 12. Waldbusser, G.G., Salisbury, J.E., 2014. Ocean acidification in the coastal zone from an organism's perspective: multiple system parameters, frequency domains, and habitats. Annu. Rev. Mar. Sci. 6:221–247. http://dx.doi.org/10.1146/annurev-marine-121211172238. Waldbusser, G.G., Brunner, E.L., Haley, B.A., Hales, B., Langdon, C.J., Prahl, F.G., 2013. A developmental and energetic basis linking larval oyster shell formation to acidification sensitivity. Geophys. Res. Lett. 40:2171–2176. http://dx.doi.org/10.1002/grl.50449. Waldbusser, G.G., Hales, B., Langdon, C.J., Haley, B.A., Schrader, P., Brunner, E.L., Gray, M.W., Miller, C.A., Gimenez, I., Hutchinson, G., 2015. Ocean acidification has multiple modes of action on bivalve larvae. PLoS One 10, e0128376. http://dx.doi.org/10.1371/ journal.pone.0128376. Wallace, R.B., Baumann, H., Grear, J.S., Aller, R.C., Gobler, C.J., 2014. Coastal ocean acidification: the other eutrophication problem. Est. Coast. Shelf Sci. 148, 1–13. Young, O.R., 2002a. The Institutional Dimensions of Environmental Change: Fit, Interplay, and Scale. MIT Press, Cambridge. Young, O.R., 2002b. Institutional interplay: the environmental consequences of crossscale interactions. In: Ostrom, E., Dolsak, N., Stern, P.C., Stonich, S., Weber, E.U. (Eds.), Drama of the Commons. National Academies Press, Washington DC, pp. 263–292. Young, O.R., 2008. Building regimes for socioecological systems. In: King, S. (Ed.), Young. Institutions and Environmental Change. MIT Press, Cambridge. Young, O., Osherenko, G., Ekstrom, J.A., Crowder, L., Ogden, J., Wilson, J., Day, J., Douvere, F., Ehler, C., McLeod, K., Halpern, B., Peach, R., 2007. Solving the crisis in ocean governance: place-based management of marine ecosystems. Environment 49, 8–19. Young, O.R., Schroeder, H., King, L., 2008. Institutions and Environmental Change. MIT Press, Cambridge. Zelli, F., van Asselt, H., 2013. The institutional fragmentation of global environmental governance: causes, consequences, and responses. Glob. Environ. Polit. 13, 1–13.