Linkage of Sustainability to Environmental Impact Assessment Using Ecosystem Services Concept; Lessons from Thailand

Ecosystem Services (ES) could support Environmental Assessments (EAs) purpose towards sustainable approaches. This study systematically analyzed the ES contents of Environmental Impact Statements (EISs) to ascertain whether they addressed appropriate data that could be used as a tool for sustainable project implementation. Three levels of EISs including Environmental Health Impact Assessments (EHIA), Environmental Impact Assessments (EIA), and Initial Environmental Examinations (IEE) were analyzed. The results indicated that the quality of EISs which reect the ES depended primarily on the project type. Relationships among ecological, social and economic components, based on land use consideration were crucial to consider the supply of resources and the demands made by the project. However, indirect effects, especially residual and cumulative impacts and alternative evaluations were lacking. Mitigation and monitoring specications were satised, but could not guarantee the eciency of project control due to lacking of mitigation hierarchy. The weakness of ecological impact analysis directly concerned biodiversity compensation, which links to net loss and net gain in ecosystem.The possibility of ES integration in EAs was reected by the limitations and opportunities detailed in the content which was nally developed.


Introduction
Ecosystem services (ES) can be de ned as the bene ts that people obtain from ecosystems (the Millennium Ecosystem Assessment, MEA 2015). This idea is a valuable tool for transforming ecological knowledge into economic information, quanti es natural resource management, and enhances the decision-making process for developers (Potschin et al. 2016;Engel et al. 2017). ES directly support the goals of sustainable development (Schröter et al. 2017) since they include a wide variety of bene ts for people from existing or ecosystems, and they affect ecological sustainability, social equity, and economic e ciency (Geneletti 2013). Since 2007, interest in the theory of ES and its relevance to policy implementation have increased (Costanza 2016), and ES have become particularly important in the eld of international politics (Maron et al. 2017;Ingram et al. 2018). ES are grouped into four categories (MEA 2015): supporting services are natural processes that maintain other ES categories (Landsberg et al. 2011); provision services deliver ecosystem products (Costanza et al. 2016); regulating services control ecosystem processes, for example, through biogeochemical cycles and biophysical structures at different scales (Fürst et al. 2016); and cultural services are the intangible bene ts that people get from nature (Sanna and Eja 2017). These categories reinforce the understanding of both supply and demand in ecosystems and carrying capacity. Potschin et al. (2016) noted that ES are used to support the relationship between environmental assessment, the ecosystem, and monitoring. Such linkages provide the structure in multi-purposes for sustainability which is the dimensions of environment, social and economic (Atun et al. 2018). However, the adaptation of the ES approach is based on a different purpose (Balvanera et al. 2016). Among these is the Environmental Impact Assessment (EIA).
Since its adoption by the National Environmental Policy Act (NEPA) in the United States in 1969, EIA has become an increasingly familiar term in both developed and developing countries. Over the following ve decades, EIA requirements have been adopted in various forms in planning, policy, and higher levels of legal hierarchy (Elvan 2018). The development of EIAs in more than 80 countries is ongoing, in terms of laws, regulations, and implementation (Morrison-Saunders et al. 2014). At the same time, EIA knowledge is diversifying, not just at project level but also at strategic levels and in many disciplines (Geneletti 2013). EIA helps to ensure that environmental and socio-economic issues are identi ed and addressed throughout the planning and implementation phases of projects and the higher levels (Victor and Agamuthu 2014; Elvan 2018). It should provide su cient information and justi cation to enable decisions to be made, based on predictions of the potential effects of the development, and identi cation of ways to reduce and mitigate unacceptable impacts (Wathern 1999). It is understood that the EIA is one of the mechanisms for regulating the environment and all development projects related to sustainable development (Höjer et  Environmental Impact Assessment, as a national instrument, shall be undertaken for proposed activities that are likely to have a signi cant adverse impact on the environment and are subject to a decision of a competent national authority. Although the target of the EIA is to support sustainable development in the project implementation, it remains imperfect (Swangjang 2018). One reason for this is that, at all stages of the EIA, it depends on legal enforcement in each country and region. The ES concept considers the relationship between the supply and the demand of the ecosystems, which are affected by a project's sustainability. Thus, it involves not only the function of the ecosystem but also socio-economic development. in which the ES were fully managed for the impact assessment framework. In the approach of Geneletti (2013), the impact assessment was used as an essential tool to focus on spatial planning policies for future ES. Many ES researches have been conducted recently in Europe and the United States. ES in relation to EIA have also been recognized in Asian regions. The study by Shoyama et al. (2017) examined the approaches used in the evaluation of ES and found that modeling and biophysical indicators were the most commonly used methods, but the impairment of geographical distribution and the practical use of models was still limited. Abcede Jr and Gera (2018) found that the inconsistencies and differences between different legal frameworks were the weak points in promoting ES for mining developments in the Association of Southeast Asian Nations (ASEAN). Many researches on ES in EIA have identi ed some limitations, but it has also led to the improvement and promotion of sustainable development in practice.
In Thailand, an ASEAN country, the concept of ES is new, even though Thailand is geographically located in the tropical zone where the ecosystem is complex and unique. ES were launched in the 11th National . Although the EIA in Thailand are very advanced, the promotion of ES for legal purposes is still at an early stage (Swangjang 2018). In contrast, EIA, as one of the Environmental Assessment (EA) approaches considered at project level, has been strengthened in all authority hierarchies. EIA in Thailand was o cially initiated in 1975 as part of the National Environmental Quality Act (NEQA). It should be noted that EIA is the familiar term for EA at project level. It was prescribed in the Constitution of the Kingdom of Thailand in 1997, and since then has been changed over time until it was nalized in 2018. There are three levels of environmental impact study at project level (Table 1), namely Environmental Health Impact Assessment (EHIA), Environmental Impact Assessment (EIA), and Initial Environmental Examination (IEE). The type and size of development projects required for each study depend on the seriousness of the impacts. The details of the studies are somewhat different from each. EHIA documents are the most signi cance, both in terms of the details and the presentation of the project, while IEE documents give the least detail.  The framework of ES in EAs has been introduced and developed on a case by case basis, dependent on the nature of each region, and the development of knowledge in this discipline. However, most such studies have been conducted in developed countries. Thailand, as developing country, although both legal and organizational progress in EIAs is being made, the outcomes still do not meet the target of sustainable development (Swangjang 2018). To ensure a sustainable target in EA, integration of ES in environmental impact studies is one of the crucial approaches. This led us to investigate the content of Environmental Impact Statements (EISs) in Thailand to ascertain whether they addressed or clari ed appropriate data to support ES for an effective EA that could be used as a tool for sustainable project implementation. The aim was to integrate the approach in order to strengthen ES in environmental impact studies which could be both direct and indirect factors for the EA effectiveness.

Method
The selection of the EISs for review was based on a subjective view. EA studies at project level in Thailand comprised three types (IEE, EIA, and EHIA), based on legal applications (see Table 1). The selection criteria depended on judgments regarding a representative of each document type. For EIA, activities related to different types of projects were used, while for IEE, the consulting rms or approved projects were used. Therefore, condominium and housing projects, studied by different consultant companies, were selected as an example for IEE. An exploration and oil production project was selected for EIA because it had a signi cant impact, and its project activities were different from condominium project. For EHIA, the assessment year was used because the project activities for all EHIA were classi ed as high signi cant impacts. Different generations of EHIA over time illustrated the development of EHIA studies. The selected EISs are shown in Table 2. Note As to con dential reason, alphabet symbols were represented as the name of consultant companies.
A qualitative analysis of ES in environmental impact studies was conducted by content analysis. The review of EIS contents were originally conducted by Lee and Colley (1990 The criteria for the main aspects of ES in the environmental impact studies were: existing environment; impact assessment; and mitigation measures, and monitoring measures. The concept of audit criteria was adopted and included three main contents, namely; the su cient data of EIA, the linkage data to ES, the data which directly support ES outcome (Table 3) Table 4). The total score for each category was calculated as an average, to re ect the content of EISs that supported ES.

Ecosystem services in environmental impact studies
The criteria re ecting the quality of the environmental impact study in support of the ES were divided into three parts: the baseline description (project description and existing environment), impact assessment, and mitigation and monitoring measures, as following.

Baseline study
Considering the baseline description as the initial stage (Table 5), the quality of the information in the EHIAs was higher than in the IEEs, while the type of project affected the baseline data in the EIAs. For all EISs, the data presentation was scienti cally well de ned, but the area of the study was not exible. A de ned distance of either 1 or 5 km from the project location was frequently mentioned in the EISs. The balanced among environmental and socio-economic issues were based more on formal guidelines than the characteristics of the speci c area. Alternative identi cation, land use, and urban planning had satisfactory scores. The integration of land use and the ecosystem was linked to social, economic, and environmental issues. Ecological data is important as it can provide the necessary framework, not only to obtain an effective EA (Dunster 1992) but also to provide spatially explicit data for ES (Mandle and Tallis 2016). The presentation of the ecological baseline in terms of an ecological index could justify the category of supporting services in ES. The linkage between the ecological, social, and economic data based on land use consideration was an advantage which were su cient to consider the balance between the supply of resources and the demands made by the project on the ES, especially for EIA 1 (score 3) and EHIA 2 (score 4). The best presentation of the regulatory support was found in EIA 1 (score 4) in which both national legislation and the international agreement were provided. This directly supported the mechanism of regulating services during the baseline stage. As a result, basic information to support ES values in the EISs were most evident for EIA 1 (score 3) and EHIA 2 (score 4).

Impact Assessment
In the next phase, the impact assessment (Table 6), the components presented in the baseline data (project description and existing environment sections) could be used to evaluate impacts or be combined with other environmental components in impact assessments of a speci c environmental component. For example, at the EHIA level, air quality, terrain, and land use were combined to evaluate impacts on air quality. In contrast, many components were presented only as basic data and not used further for impact assessment. According to the contents, the highest average score (2.63) was for EIA 1.
In this EIS, the consistency of the causes of the impacts that affected the sustainable components was considered based on biodiversity within the proposed area. However, for the other EISs, the correspondence of many criteria revealed the inadequacy of using the baseline data in evaluations of project impacts. Although some mathematical models were used for some components, in particular, air quality, quantitative details were used only the values of speci c parameters, without connecting the effects with the supply and demand in the ecosystem. Due to the failure to incorporate indirect effects in the impact assessment, the connection to the ecosystem based was at a low level. These details were, for example, "species of organisms are so common hence a low impact is predicted" or "... the project's wastewater was collected in the combined wastewater treatment system, so the "impact on biological resources is negligible." According to Tallis et al. (2015), ES in EAs should estimate the impact on their value and be included in ES delivery. The assessment of impacts on a single component did not accurately re ect the bene t of supply and demand in ecosystem. The incorporation of the 'no net loss' and 'net gain' concepts in biodiversity, together with the other biodiversity criteria, was insu cient due to the failure to conduct further assessment of indirect impacts, especially residual and cumulative impacts.
Furthermore, a lack of alternative evaluations of the project was the weakest point for all EISs, even if these criteria had satisfactory assessment scores in the baseline phase. In contrast, the best average score for the impact assessment was for the consideration of the project life cycle and the members of public who were affected (score 3.00). Surprisingly, for EHIA 2, the assessment of impact was de cient, with a score of 2.00, although the quality of the baseline were satisfactory. This is one of the problems in EA studies, which use more of the budget for the presentation of data, without making the necessary budget available for the subsequent phases.

Mitigation and Monitoring Measures
With regard to mitigation and monitoring measures (Table 7), the best average score (3.55) was once again for EIA 1, which was a well-de ned requirement for project control activities. This issue was signi cant because the impact assessment output was a tool for project implementation. Mitigation and monitoring identi cation in the EHIAs were better than in the IEEs. Although basic details of mitigation and monitoring implementations were satisfactory, they were not guarantee the e ciency of project control. This was particularly the case for mitigation hierarchy, which can connect project control to biodiversity offsets. Consequently, biodiversity offsets, which are directly related to supply in the ecosystem, were lacking for IEEs and unclear for EHIAs. The best score was EIA 1, in which the consideration of biodiversity loss and the programs to control it were speci ed and covered the project lifecycle. An alternative aspect was lacking for all EISs under review. However, the identi cation of the chance to enhance or change mitigation measures in case of the future nding of unpredictable impacts was crucial for mitigation development. The scores for the quality of the baseline information in the different types of EISs were in the range 2.00-3.75. Surprisingly, the quality of the information for impact assessment, which is a crucial stage, was found to have the lowest average score (1.83). For mitigation and monitoring, the average score was 2.24. The project type in uenced the quality of the mitigation and monitoring, as the scores for EHIAs were higher than those for IEEs (Fig. 1).

The integration of ecosystem services in environmental impact studies
It is important for the administration of ES to be included in environmental impact studies. The role of the ES can improve the understanding of the ecosystem mechanisms resulting from the project activities.
According to Karjalainen et al. (2013), ES may be considered in the early stage of an EA study through mitigation and monitoring. The possibility of ES integration in EAs was re ected by the limitations and opportunities detailed in the contents, see Table 8.
The ndings of this study are initiated how ES could link in environmental impact study, as follows.
Firstly, in the scoping phase, the selection of appropriate sustainability indicators should be focused and assessed throughout the EA study phase. However, the adequacy of the biodiversity baseline and its link with the other components is important in providing satisfactory information for the services demanded mostly indirect impacts resulting from physical impacts, such as air or water impacts. Moreover, existing quantitative studies, which consider changes at both the temporal and spatial scales of ecosystems, are critical for successful integration of ES and the environmental impact study (Sirami et al. 2007). The temporal scale, which affects both the baseline data and the impact assessment, should include the characteristics of species that occur and an interpretation of the impact predictions (Gontier 2007).
Speci cation of the condition of the fauna and ora within the ecosystem is one approach that could support temporal coverage. This is vital in subsequent stages of the impact assessment. This lack of temporal data is problematic, especially for the small projects characterizing the IEE group, in which baseline data were only collected on a single date.
For the impact assessment, the project phases should include the project lifecycle, at least in the construction and operational phases, and determine the resource demands. Loss and gain of biodiversity resulting from project activities should be the rst priority for any impact assessment. According to Geneletti (2016), an impact assessment should depend on ES as the priority, but more legal guidance is needed that could indirectly assert the assessors. Baseline details should be further assessed for their impacts. Project alternatives, such as project sizes, locations or processes, presented in the baseline data should assess the impacts and be considered further for mitigation and monitoring. The weakness found by our study was the lack of alternative considerations throughout the stages of environmental impact studies. Remarkably, the impacts on ecosystems, from varying levels of human disturbance, were addressed in descriptive terms. This raised some uncertainty regarding the overall impact assessment, which had been done without consideration of biodiversity concepts such as species loss, project effects on the natural habitat, and community and ecosystem components. Ecological impact identi cation and evaluation and contributions to environmental change can be negative and/or positive, residual and/or cumulative, and signi cant and/or magnitude impacts. Impact projections affect subsequent project activities. For example, negative impacts should be a priority in effectively managing the reduction of adverse ecosystem effects.
Finally, the mitigation and monitoring speci cations should consider the ways to maintain supply and demand in ecosystem with reasonable costs and bene ts. Options for mitigation, through the mitigation hierarchy, can improve and maintain the well-being of affected bene ciaries from ES (Landsberg et al. 2011). Mitigation hierarchy is fundamental to EA practice for biodiversity offsets by the consideration of alternatives in program identi cation (Brownlie and Treweek 2016). Project compensation in practice should adhere to the mitigation sequence of avoid, minimize, rectify, reduce, and compensation or offsets (Villarroya and Puig 2010). These can compensate for the loss of ecosystem resulting from the project demands, and it is a crucial approach for incorporating ES in EAs.
The bene t of mitigation and monitoring was the inclusion of public opinion, in accordance with legal enforcement. Those response should re ect the actual requirement of the local public (Chaisomphob et al. 2004). For better implementation, the members of the public identi ed should be representative of the public concerned, and poverty in both income and well-being should be considered. According to the UN sustainable development goals adopted in 2015, poverty (Goal 1) also covers the poor in situations such as climate-related events and other shocks and disasters. Hence, all people should be considered equally.
Program achievements are a vital element of the program performance, and they are exhibited by the project control agencies during project implementation. In this regard, essential monitoring directly affects the likely implementation of EA in practice. The cost of monitoring implementation is a fundamental principle that touches upon all monitoring-related activities. The ways to improve performance relate to these aspects. Firstly, exible programs are required by optimizing the design.
Secondly, the period over which parameters should be monitored should be included in the programs. Thirdly, the e ciency of resource use in monitoring should be a focus. The consequences of giving appropriate attention to these factors would be a reduction in the costs of monitoring unimportant parameters that create wasted effort. These savings could lead to better environmental management.
The success of mitigation and monitoring is compliance. The effectiveness of mitigation and monitoring programs, all of which are de ned in o cial documents, is not a guarantee that the programs will be implemented. This is con rmed by the results of this study in the real estate projects. The mitigation and monitoring actions may have made it di cult to measure the accuracy of impact prediction since the results of the impact assessment were not the key element in identifying the programs. Consequently, the performance during project implementation was questionable.
Roe and Geneletti (2016) indicated that Biodiversity underpins the delivery of essential ES on which the whole of humanity is dependent. However, it depends on the nature of the project and the environment where the project is located. To better incorporate ES, through eco-based objectives, in environmental impact studies, connections between biodiversity content and the different stages of an EA are strongly recommended. These points can help EA to support the sustainable development goals. -The presentation had a scienti c basis.
-All projects complied with the enforcement of urban planning.
-Project alternative and reasonable choices were identi ed.
-Ecological, economic, and social linkages were considered through the characteristics of land use, especially for EIA 1 and EHIA 2.
-The information linking to the ES was urban planning, land use features, and the relationship between project activities and their ecosystem.
-The boundary of the study was xed, lacking the exibility of the surroundings.
-The determining component depended on the o cial guidance rather than project activity and the nature of area.
-The presentation of ecological information was lacking, only super cial information, especially for IEE projects.
-Survey and data collection were only done once.
-Project activities and biodiversity components based on land use conditions are the main supply and demand considered for ecosystem capacity.
-The environmental, social, and economic components should be considered equally depending on the nature of the proposed area.
-Spatial and temporal coverage should be speci ed -Biodiversity context is based on the quality of ecological study in the EA.
Impact Assessment -Impact assessment covered project life cycle.
-The considerations of the affected public were satis ed.
-Project alternatives were not further assessed for their impacts.
-The weaknesses of ecological impact assessment were remarkable, especially the linkage to sustainable development.
-Impact assessment of indirect effects, together with residual and cumulative impacts failed.
-The impact assessment was ambiguous, except for EIA1.
-Baseline data should provide the main issue to determine the assessment of impacts on service delivery and values of ecosystems.
-Criteria for impact identi cation and evaluation should focus on biodiversity values and cover both direct and indirect impacts, together with residual and cumulative impacts.
-The spatial extent of impact assessment should be considered as the biodiversity and associated consequent mechanism depending on the ES of the area. -Quality of mitigation and monitoring were based on project levels related to the enforcement by competent agencies.
-Mitigation implementation was questionable, especially for IEE.
-Mitigation hierarchy is not well de ned.
-Biodiversity compensation of net loss is not clear.
-Alternative mitigation and monitoring were lacking -Time span of monitoring and future modi cation was lacking.
-Mitigation hierarchy should be highlighted. Avoidance of impacts is initiated, with subsequent hierarchy through biodiversity offsets.
-Residual and cumulative impacts are recognized for future mitigation enhancement.
-Appropriate biodiversity offsets should be designed and depend on the landscape of the area.
-Monitoring focuses on high e ciency for project implementation.

Conclusion
The ndings provide some insight regarding the integration of ES in EA. The quality of EISs which re ect the ES depended primarily on the project type. In this regard, the legislation of the respective agency is supreme while the fundamental constraint has been found in real estate projects. Furthermore, the weakness of ecological impact analysis, from species level to biodiversity, directly concerns biodiversity compensation, which links the approach of environmental analysis to net loss and net gain in ecosystem.
The warning of the ecosystem stock could provide the goods and services for the proposed project.

Consent for publication
The author declares that this manuscript or any data contained is received the consent for publication.

Availability of supporting data
All data generated or analyzed in this manuscript are available from the author on reasonable request.

Competing interests
The author declare they have no competing interests.

Funding
Page 20/25 The O ce of the National Research Council of Thailand no. SURDI 590155

Authors' contributions
The author has contribute to study the research, to write and analyse this manuscript.