Evidence of learning processes in EIA systems

Abstract

Learning is recognized as an important component of environmental impact assessment (EIA) in particular, due to its potential to transform individual, social, and organizational values, standards and practices to support a sustainable development. Nevertheless, only a few models developed to describe learning processes in EIA systems seem to be grounded on the combination of both theoretical aspects and empirical evidence of learning, which requires further investigation. This paper is based on the development of a conceptual model to understand learning processes through the practice of EIA, and evidence-gathering of learning in EIA systems focused on the role played by the leading organization. The outcomes were based on literature review and empirical research using two EIA regulatory agencies in Brazil as case studies. Data collection and analysis were supported by participant observation, personal interviews and qualitative content analysis. Although specific to the context, the evidence produced have corroborated the implicit feature of learning in EIA systems and, we believe, reinforced the relevance of the conceptual model as developed in the paper.

Keywords:

• Learning processes
• conceptual model
• organizational learning
• EIA systems

Introduction

The concept of learning is associated to cognitive changes due to knowledge acquisition, promoted by different processes, experienced by an individual, a group, an organization or even by the society in general (Argyris and Schön 1978, 1996; Crossan et al. 1999; Jha-Thakur et al. 2009).

According to Diduck and Mitchell (2003) learning occurs primarily at the individual level, ideally changing people’s perspectives, beliefs, values, and attitudes. Besides, according to Berkes (2009) and Bond and Morrison-Saunders (2013), there’s a linkage of individual and organizational learning levels that may occur throughout participatory or collaborative approaches that enable the exchange of values with different group members.

Due to its potential to transform individual and organizational standards and practices to support a sustainable development (Kidd et al. 2011), learning is also considered an important component of environmental impact assessment (EIA) (Jha-Thakur et al. 2009) related to a wide range of outcomes, which includes the acquisition of knowledge, development of skills, and the promotion of new behaviors and values (Sánchez and Mitchell 2017).

The literature recognizes learning as an intrinsic but complex feature of EIA systems, inherently associated to the uncertainties of impact assessment and the need to adapt over time (Bond and Morrison-Saunders 2013). It is also a core element of EIA effectiveness (Bond et al. 2010) once it can support more evidence-based process (Bond and Morrison-Saunders 2013).

The significant involvement of stakeholders (from community members to proponents, regulators, consultants, and other participants of EIA processes) in EIA creates opportunities of knowledge sharing, which is also referred elsewhere as a process of collaborative learning (Saarikoski 2000). In this sense, according to Sinclair et al. (2008), EIA can be understood as a learning platform.

Therefore, the comprehension of learning processes and their mechanisms in EIA systems can help to increase the awareness about strengths and weaknesses in both the assessment and decision-making processes, and to create opportunities to improve EIA’s effectiveness (Fitzpatrick 2006; Bond et al. 2010; Bond and Pope 2012).

Despite the relevance of learning to EIA theory and practice, empirical evidence of learning through EIA process seems to be scarce in professional literature and in need of more attention (Bond et al. 2010; Kidd et al. 2011; Sánchez and Morrison-Saunders 2011; Sánchez and Mitchell 2017). Moreover, only a few models developed to describe learning processes in EIA systems seem to be grounded on the combination of both theoretical aspects and empirical evidence of learning, which justifies further investigation.

This paper presents a conceptual model developed to understand learning processes through the practice of EIA, focused on the role played by the leading organization within EIA systems – herein represented by the environmental agencies. The model was developed based on both literature review and empirical evidence of learning as found in two environmental/regulatory agencies responsible for the coordination of EIA processes at different levels (state and federal) in Brazil. Data collection was supported by content analysis applied on EIA documentation, participant observation and personal interviews.

Background

Learning processes constitute a topic of great interest in Impact Assessment/Environmental Assessment (IA/EA) literature. Basically, learning has been explained relying on two complementary approaches: (i) as an intrinsic aspect of IA/EA systems and, therefore, several aspects may be related to the promotion of learning along the EIA process (see, for example, Jha-Thakur et al. 2009; Bond et al. 2010); and (ii) as an outcome of IA/EA processes, which means that learning is obtained through EIA practice (Fitzpatrick 2006; Kidd et al. 2011; Sánchez and André 2013; Sánchez and Mitchell 2017).

Learning outcomes are often described in terms of two perspectives: single-loop and double-loop learning. The first is normally associated to measures adopted along the EIA process to correct (or to improve) the defined actions that are not compatible with the expected results, and the second is illustrated by the adjustment of strategies, behaviors, and cultures that have guided these actions (Armitage et al. 2008; Sánchez and Morrison-Saunders 2011; de Jong et al. 2012).

Also, a single-loop learning outcome refers to improvements in the ability of making and implementing collective decisions, directly affecting EIA procedures, methods and approaches, while a double-loop learning outcome is linked to improvements on strategies through innovative approaches (Pahl-Wostl 2009), thus affecting the objectives of the EIA system. Kidd et al. (2011) suggest that there should be a balance between single and double-loop learning, with an increased support toward the second due to the need of rethinking established rules and practices of IA/EA.

Regarding the factors that may influence learning in EIA systems, time, and financial resources invested, as well as the ability of engagement and communication between stakeholders (individuals and/or organizations), are crucial for better learning outcomes (Fischer et al. 2009; Bond and Morrison-Saunders 2013).

Individual learning influences the system’s learning outcomes, but it is also influenced by both the institutional arrangements and the EIA process itself (Sinclair et al. 2008; Sánchez and Mitchell 2017).

Ideal conditions for learning are linked to an effective participation of stakeholders along the assessment process. This would imply in a context where: (i) a complete and accurate information is shared; (ii) participants are opened to alternative perspectives; (iii) the opportunities to participate and influence on the decisions are balanced; (iv) the different arguments could be objectively assessed; (v) critical thinking is stimulated and developed; and (vi) consensus could be achieved through a rational and collaborative debate (Diduck and Mitchell 2003; Chávez and Bernal 2008).

Kidd et al. (2011) have invested a considerable effort to describe learning from the strategic perspective of IA/EA, developing a conceptual model (Figure 1) that draws our attention to four main questions:

(i)	The role of individuals and organizations (who?) involved in Strategic Environmental Assessment (SEA), which may influence the learning structure of the system and even the learning mechanisms;
(ii)	The influence of the context (where?);
(iii)	The methods and approaches applied in SEA (how?);
(iv)	The existence of different levels of learning that can be verified in SEA, stressing the relevance of learning outcomes to SEA effectiveness (with what effect?).

Figure 1. Conceptual model for learning processes through Strategic Environmental Assessment (Kidd et al. 2011, p. 55)

Plenty of attention has been given to learning factors such as legislative and institutional arrangements and professional training, and a growing interest has been devoted to factors related to cultural, structural, and behavioral conditions within organizations (Kidd et al. 2011). Wherever organizations encourage reflective dialogs between colleagues (which can occur in less formal situations referred to as shadow spaces in Figure 1) and a critical view of their prevailing routines, values, and traditions, more opportunities of learning are expected to arise.

Social learning occurs in EIA systems when actors and organizations work together and share information toward effective and socially accepted strategies to mitigate impacts and identify opportunities to promote the environment in decision-making (O’Faircheallaigh 2010). Social learning is considered an essential aspect of decision-making processes (Berkes 2009), to achieve sustainable results (Sinclair and Diduck 2001; Diduck and Mitchell 2003; Sinclair et al. 2008) and to support double loop learning (Pahl-Wostl 2009), thus creating opportunities for social changes (Webler et al. 1995).

According to Crossan et al. (1999) the different learning levels within an organization are connected through four sub-processes of communication: intuition, interpretation, integration, and institutionalization. The authors have proposed a conceptual model (Figure 2) assuming intuition as a process that occurs at the individual level, while institutionalization occurs at the organizational level. In addition, interpretation mechanisms connect the individual and group levels and, finally, integration mechanisms should secure proper conditions to reach the organizational level from the group level.

Figure 2. Organizational learning as a dynamic process (Crossan et al. 1999, p. 532).

Moreover, Crossan et al. (1999) state that feedback and feedforward processes are responsible for bridging the four communication sub-processes. Feedforward processes provide knowledge transfer from individuals and groups to the organization, which becomes institutionalized through systems, structures, strategies, and procedures. By their turn, feedback processes enable the institutionalized learning to influence individuals and groups. Thus, processes implemented toward organizational learning provide opportunities of both knowledge assimilation and internalization of the lessons learned within organizations.

Methods

In order to develop a conceptual model that describes learning processes in EIA systems, a qualitative and exploratory approach was adopted based on Silverman (2005), Seidman (2006) and Yin (2010).

The similarity approach described by Perdicoúlis (2012) followed herein enables either to identify connected models or to develop a new one. In our case, the model developed in this paper relies on the complementarity between the models proposed by Crossan et al. (1999) and Kidd et al. (2011).

The methodology was established as follows:

(i)	Literature review to provide the theoretic framework regarding the processes of learning through the practice of EIA, particularly focused on the identification of the mechanisms and the respective evidence of learning as reported in literature;
(ii)	Analysis of regulations and EIA documentation, to set the context for empirical research;
(iii)	Design of a conceptual model based on the previous stages, that helped to identify the key topics to be covered by interviews and empirical research;
(iv)	Identification of empirical evidence of learning, as occurring in the context of two regulatory EIA agencies;
(v)	Personal interviews with representatives of the agencies to confirm the collected evidence and to ensure the comprehension of learning mechanisms within each agency.

Finally, the conceptual model to understand learning in EIA systems was grounded on the outcomes of previous stages.

It was assumed that the environmental agencies/regulating bodies occupy a singular position within EIA systems, usually coordinating/moderating the EIA process. This was also assumed in a variety of case studies reported in recent literature to different contexts, such as Pölönen et al. (2011), Kolhoff et al. (2013), Jha-Thakur and Fischer (2016), Suwanteep et al. (2016), including Brazil (Bragagnolo et al. 2017).

In our case, two regulatory agencies in Brazil (state and federal levels) have provided the context to the case studies.

Data collection for the identification of learning evidence within the selected organizations was carried out through participant observation, semi-structured interviews, and the analysis of regulations and EIA documentation, which provided a previous comprehension of the routines and EIA practice in the different contexts.

Elements of organizational learning identified in the literature helped to compose a pre-script prepared for the interviews that were carried out personally and individually with environmental analysts and division managers in each of the selected agencies. A total of 20 interviewees from the State Institute of Environment and Water Resources (IEMA, according to its acronym in Portuguese) and 8 from the General Coordination of Oil and Gas (CGPEG) have participated in this research.

To the purposes of this paper, five categories proposed by Dibella et al. (1996) to describe the organizational learning capability were assumed as being directly related to the context of EIA systems: skills development (in different levels: individual, group, and organizational); knowledge sources (internal and external); documentation mode or knowledge repositories (individual and collective); knowledge dissemination mechanisms (formal and informal); and the learning focus adopted by the organizations (assumed herein as related to the learning loops previously described).

Coupled to the learning categories, three questions have guided the investigation at this stage: what can be learned?; how learning can be achieved?; and with what effect?. Therefore, relying on the mechanisms and learning evidence described by Dibella et al. (1996), Crossan et al. (1999), Sánchez and Morrison-Saunders (2011), Sánchez and André (2013) and Sánchez and Mitchell (2017), a number of guiding elements were defined to help finding evidence of learning in EIA systems as presented in Table 1.

Table 1. Guiding elements to find evidence of learning through EIA.

Questions	Categories	Evidence	References
What can be learned? (learning outcomes)	Learning loops (single and double loop)	Single loop learning	Sánchez and Mitchell (2017)
		Increased scientific knowledge
		Increased capacity to mitigate impacts and enhance benefits
		Preparing better IA documents (ToRs, EIS, environmental and social management plans etc.)
		Adopting more effective communication strategies
		Corrective and incremental actions	Dibella et al. (1996)
		Double loop learning	Sánchez and Mitchell (2017)
		Recognizing the need for negotiating multiple objectives and trade-offs when adopting sustainability as a policy goal
		Developing reasonable alternatives consistent with sustainability objectives
		Addressing gaps in legislation/regulation that hinders EIA effectiveness
		Innovative thinking	Dibella et al. (1996)
How can learning be achieved? (processes to facilitate learning/to deliver learning outcomes)	Knowledge sources	Training and capacity building	Sánchez and Mitchell (2017)
		Internal: knowledge acquisition through internal processes, lessons from experience, internal research	Dibella et al. (1996)
		External: alliances with universities, research centers or benchmark organizations, technical visits, continuous feedback from external organizations
		Formal: investment in formal educational methods, new staff (source of knowledge acquisition)
		Informal: informal dialogs, senior experience
	Knowledge dissemination	Formal: written communication, periodic meetings, feedback systems	Dibella et al. (1996)
		Informal: continuous dialog, personal contacts, early socialization
	Skills development (individual, group, organizational)	Individual learning: individual actions	Crossan et al. (1999)
		Group learning: interactive conversational processes
		Group learning: team skills, investments in group training	Dibella et al. (1996)
		Organizational learning: dialog and joint actions, shared understanding, coordinated actions by members of a workgroup, establishment of formal rules and procedures by the workgroup	Crossan et al. (1999)
	Documentation mode/knowledge repositories	Establishing and maintaining internal structures to facilitate organizational learning: information sharing, information interpretation, organizational memory	Sánchez and Mitchell (2017)
		Manuals and templates	Sánchez and Morrison-Saunders (2011); Sánchez and André (2013)
		Technical sheets
		Key environmental factors
		Guidelines for environmental review documents
		Environmental protection policies
		Guidance statements
		Environmental analysis and follow-up reports
		Central information systems
		Files
		Staff meetings
		Informal talks
		Senior staff
		Documentation center	Dibella et al. (1996)
		Shared repository of lessons learned

The presence of the researchers in the agency’s office in different opportunities allowed to observe the EIA system ‘in action’ and to better understand the different working routines (which included, e.g. formal and informal meetings and discussions, analysis of environmental impact reports, project follow-up meetings, and so on).

Whenever the researcher (observer) perceived an event (which included, for example, a behavior, the subject of a conversation, a meeting, etc.) that could be related to the driving forces, inputs and outcomes of learning processes, his/her personal judgment was noted down and subsequently discussed with the group of authors. Doubts and uncertainties were subsequently clarified directly with the interviewees.

The empirical evidence of learning found in the selected organizations was tested against the theoretical framework and confirmed with the interviewees, thus securing the coherence of the conceptual model as presented in the paper.

Results and discussion

Conceptual aspects of learning in EIA systems

Based on the aforementioned concepts and models, and mainly on the outcomes of literature review, we postulate that a conceptual model for learning in EIA systems should consider that:

•	Learning may transpose diverse levels within an EIA system, from individual to group and to organizations and society. These levels must be connected through feedback and feedforward processes in order to support meeting the goals of the organization (Crossan et al. 1999) and/or the broader EIA system it integrates;
•	Contextual factors may influence learning processes and outcomes, thus affecting the effectiveness of EIA systems. These factors are related to institutional arrangements and professional training and skills (Kidd et al. 2011) as much as to the organization’s culture, structure, and behavior (Argyris and Schön 1978). Learning outcomes (or effects) may vary due to social, cultural, and historical context (Lawrence 2003);
•	Single and double loops of learning can be verified in EIA systems. The first represents changes at the process level, encouraging the individuals to question whether they are doing things right (know how) while the second takes place at the systems level and encourage individuals to reflect about whether they are doing the right things (know why) (Argyris and Schön 1978, 1996; Kidd et al. 2011);
•	Collaborative and social learning are essential to promote solutions through a normative consensus (considering both legal and social responsibilities) and may be facilitated by public participation mechanisms (Webler et al. 1995; Fitzpatrick 2006; Bond and Morrison-Saunders 2013; Glucker et al. 2013; Sánchez and Mitchell 2017).

In Table 1, the learning loops represent possible learning outcomes and are described in terms of actions to improve performance within existing processes (single-loop), such as the adoption of corrective and incremental actions; or new behaviors and strategies that affects the system’s objectives (double-loop), for example stimulating innovative thinking or addressing gaps in legislation/regulation that hinders the effectiveness of IA (Dibella et al. 1996; Sánchez and Mitchell 2017).

Regarding the processes that may facilitate learning within EIA systems, it is important to comprehend the dynamic within the organizations involved, as well as the individual’s behavior. The elements/categories detailed below have contributed to understand how learning can be achieved in EIA systems.

Knowledge sources, according to Dibella et al. (1996), represent the different sources that may be used by an organization to gather knowledge and build learning results. Sources may be described as internal or external to organizations, which means that learning outcomes may derive from internal operations results or ideas from external sources (Dibella et al. 1996). Also, formal and informal mechanisms may be used to acquire knowledge, like formal educational processes (also suggested by Sánchez and Mitchell 2017), or even a dialog or a discussion group (Crossan et al. 1999).

The learning capacity of an organization is influenced by knowledge dissemination. This element is characterized by formal and informal mechanisms as learning may occur through written communication, institutional methods, and experiences sharing among individuals or group/teams, or either through continuous and informal dialogs (Dibella et al. 1996).

The levels of skills development and their interconnection are presented by Crossan et al. (1999) as feedback and feedforward mechanisms within an organization, from the individual level to group and organizational levels. An example of group learning can be seen through the adoption of an interactive conversational process, which reaches the organizational level when it is detected a shared understanding and coordinated actions by members of a working group (Crossan et al. 1999).

Knowledge repositories contribute to the learning process and are directly related to the documentation within the organizations and mechanisms used to keep the organizational memory. This is an element that may benefit from individual or collective approaches within organizations.

Sánchez and Morrison-Saunders (2011) and Sánchez and André (2013) highlight that learning can be lost if not managed in order to be renewed, strengthened and transferred. If managed only at the individual level and not raised to the collective level by structuring mechanisms devoted to organizational memory, learning may leave the organization sphere (Dibella et al. 1996).

Thus, mechanisms to access and preserve knowledge, also known as knowledge repositories or organizational memory, are essential for learning outcomes (Fitzpatrick 2006).

The conceptual model developed in this paper is presented in Figure 3. Based on the outcomes from previous stages of the research, the model reinforces the fact that, in a given EIA system, learning processes may occur at the stakeholders’ level (who?), which includes individuals, organizations and society involved in the EIA process. Primarily occurring at the individual level, learning exchange through feedback and feedforward processes can eventually reach also group and organizational levels, advancing to the social level through stakeholder’s engagement and participation.

Figure 3. Conceptual model for learning through EIA.

Moreover, the model highlights the relevance of contextual factors (where?), procedures, methods and approaches to EIA (how?) and its objectives (what?), assumed here as driving forces of learning outcomes (with what effect?) that, by their turn, might have a strong influence on EIA effectiveness.

The learning outcomes affect the whole EIA system and are related to the learning levels reached by the stakeholders based on their behaviors and ambitions. They may be described in terms of two different cycles, single, and double-loop, which are facilitated by the organizational culture and openness to innovation and changes.

Evidence of learning processes, mechanisms, and outcomes

The Brazilian regulating agencies studied in this paper are responsible for the EIA process at federal and state levels.

CGPEG is responsible for the environmental licensing processes of offshore oil and gas exploitation and production projects in Brazil. Established in 1998, CGPEG currently works with about 80 environmental analysts (Mendonça 2015). From 2010 to August 2016, 130 projects were submitted to the EIA process managed by CGPEG, including 60 seismic researches, 34 projects of drilling wells, and 36 projects of oil and gas production.¹

By its turn, IEMA is responsible for the coordination of environmental policies in Espirito Santo State (southeast of Brazil), including monitoring, surveillance, research, environmental assessment, and environmental licensing process of development projects from different types (IEMA 2016). Established in 2002, it has currently 13 environmental analysts dedicated to the EIA process. From 2010 to August 2016, 29 large projects were submitted to the licensing process managed by IEMA, including 11 different project types.

External knowledge sources have shown to be common to both agencies, basically related to technical visits to similar projects in other contexts. The acquisition of knowledge from internal sources is usually related to the documentation/information flowing through the EIA process, including internal memorandums exchanged by statutory authorities and other stakeholders.

Evidence of formal knowledge acquisition mechanisms found in both organizations are mainly related to individuals’ initiatives, e. g. the participation of the analysts in training/capacity building courses and, also, to their involvement with the professional community (e.g. attending to scientific conferences) and collaborative work (e.g. being involved in research projects). It was found that CGPEG fosters the formal capacitation through a paid leave grant during the period of, for example, postgraduate studies. Also, the Brazilian Institute of Environment and Renewable Resources (IBAMA), from which CGPEG is subordinated, offers training courses to the analysts through its Environmental Licensing Board (although the interviewees may find the courses not suitable to the organization’s needs justifying that the EIA practice is further more complex than the theory covered by them).

Informal dialogs and the experience of senior analysts are important sources of knowledge access at these regulators, which is similar to the findings of Sánchez and André (2013) in Canada. Bond et al. (2010) also reinforces the importance of informal mechanisms for knowledge acquisition toward EIA effectivity, since they deal with issues beyond the formal process like group work, leadership, and collaboration.

Knowledge dissemination within these organizations was found to be based on both formal and informal mechanisms. Informal dialogs and chats characterizes the informal mechanisms for knowledge dissemination, while formal mechanisms were verified through internal meetings and knowledge sharing within working groups (WG). At IEMA, the creation of these groups is formally demanded by the environmental agency, whenever necessary, as part of a project’s follow-up. Acting temporarily, it was perceived that the working groups created at IEMA apparently do not retain lessons from previous works, which may be related to this particular aspect.

At CGPEG, there are permanent thematic WGs created to discuss critical environmental issues in order to develop advanced guidelines and improve the licensing process. Although the creation of the WGs by the analysts had happened informally and voluntarily, they are currently part of the organizations’ routine and fully acknowledged by the analysts and managers. These groups involve analysts from all the working teams and divisions, distributed by their themes of interest, thus creating opportunities for continued dialogs toward changes in strategies and values to improve the EIA system.

Similar to the outcomes reported by Gazzola et al. (2011), communication and team working skills were improved by bringing together different experts to work at CGPEG. The WGs meetings promote an interdisciplinary engagement of the environmental analysts, contributing for learning development at different levels, especially toward the organization goals.

Concerning the levels of skills developed at CGPEG, there is a clear feature to the organizational learning level, which is also strengthened by the elaboration of technical notes (TN) (mainly created within the WG discussions), whose goals are to guide the environmental licensing process through the implementation of improvements on the process. According to one interviewee, the TNs reduces discretion at the decisions, creates a record of the developed knowledge and makes it accessible to the entire organization (CGPEG, interviewee 4).

At IEMA, the development of learning at the individual level overrides the organizational level since, in practice, organizational learning was found to be highly dependent on the existence of feedback and feedforward processes. This is a clear limitation for learning development at the organizational level, and could be noted in different moments of the interviews:

We still need a better definition of competences, responsibilities and more clarity related to our duty as an environmental agency in order to diminish the complexity of our task, and this must come from the State and Federal level. (IEMA, interviewee 1);

I believe the organization has deviated from its final role, it gives me the impression that the client is solely the companies. Currently, the bureaucratic side of EIA prevails over its principles and fundaments, and our job is much more focused on the procedural aspects of EIA and licensing than properly on the assessment of environmental impacts. Acceptability is normally based on monetary compensation and not on environmental benefits. (IEMA, interviewee 2);

We (environmental analysts) need more feedback related to our actions and decisions, we have to learn how the actual impact occurred in order to verify whether we did a good prediction or not, and use this knowledge to improve the next assessments. (IEMA, interviewee 6);

Central organization (IBAMA) is clearly trying to assume the control over final decisions, thus leaving to CGPEG the role to inform the decisions to be taken. This is, obviously, leading to a conflicting environment within the institution, reinforced by the various changes and modifications of decisions that we took considering the environmental assessment in first place and now are taken clearly based on political convenience to IBAMA and federal government. (CGPEG, interviewee 3).

It is important to note that the structure of CGPEG seems to ease internal organizational learning, what may be related to the influence of contextual factors. This can be explained by its organizational framework, which reveals a greater proximity between EA analysts and the board of directors/coordinator, reinforced by initiatives such as the creation of permanent thematic working groups as previously mentioned. Therefore, it is reasonable to conclude that the ‘individuals’ are sufficiently close to the ‘organization’ so that the feedback is effective. On another hand the organizational environment found at IEMA has demonstrated severe limitations to promote enough feedback and feedforward processes, possibly related to the variety of teams and working groups organized by types of projects and other aspects that contribute to dissipate the knowledge that otherwise could be retained from EIA practice.

The main contextual difference between both organizations is related to their jurisdiction (federal and state levels) and legal competences. IEMA’s attributions observes a wider scope compared to CGPEG’s, including the coordination of the whole environmental licensing process in Espirito Santo State and dealing with a much diverse type of projects, which may add complexity to the EIA process and, therefore, may challenge the learning development. Nevertheless, the empirical evidence found in this organization corroborates the implicit feature of learning in EIA systems according to Bond et al. (2010) and Fitzpatrick (2006).

Knowledge repositories are essentially the same, meaning that both organizations take advantage of document templates, technical notes and regulatory instructions that retain the lessons learned, archives of environmental licensing processes/documents, environmental studies, and meetings memoir, digital files shared via computer networks and the experience/memory of the analysts. As a common deficiency, physical, and digital files recovery proved to be difficult, thus suggesting a need for more efficient mechanisms to track the documents. This is similar to the repositories (templates, guidelines, information systems or documentation center, staff meeting, and informal talks) found in the context of Canada (Sánchez and André 2013) and Western Australia (Sánchez and Morrison-Saunders 2011).

Learning evidence associated to single and double looping processes were found in both organizations, thus suggesting they may be characterized as flexible and adaptive organizations according to Morrison-Saunders and Arts (2004). The elaboration of technical notes, technical instructions, and other procedures for normalization, optimization, and improvement of the decision-making process are actions related to the concept of single loop learning and are present at IEMA and CGPEG.

Public participation mechanisms were also noted in both organizations: at IEMA, public consultation during the scoping stage of EIA also can be taken as an evidence of single loop learning, although not systematically implemented, which reflects somehow the knowledge acquired by the organization. According to one interviewee, the socioeconomic area has a great support from the managers and our organization is acknowledged as the state government agency that interacts the most with the public (IEMA, Interviewee 4). At CGPEG, a public consultation of TNs is done before their approval, which legitimizes the decisions and procedures elaboration (CGPEG, Interviewee 8). Another evidence of single loop learning is the regional approach adopted by CGPEG at the licensing process, which is a good initiative for the integration of environmental management programmes required for projects approval, thus contributing to enlarge the vision of analysts and decision makers during the EIA process, with an evident reflex on the consideration of cumulative impacts.

Evidence of double loop learning found at IEMA includes the contribution of environmental analysts to the definition of changes in the legislation that set the procedures to be followed during EIA processes in the state. According to the interviewees, the main contributions coming from IEMA to the legislation, as found in the State Decree 1777/2007, are related to the introduction of consultation mechanisms as part of the EIA process (particularly during the scoping stage).

By its turn, CGPEG has played a leading role during the elaboration of the Interministerial Ordinance 198/2012, which has defined a new instrument to support EIA and decision-making, named Environmental Assessment of Sedimentary Areas (AAAS). The AAAS is a multidisciplinary and strategic planning tool for environmental assessments to support the selection of potential offshore areas for the exploitation and production of oil and gas. According to the interviewees, there is a clear expectation that the outcomes of these studies (and, mainly, the lessons to be learnt) may contribute to the decision-making process conducted by CGPEG, fostering the improvement of the EIA system: the AAAS is still an ongoing process, there is a lot of discussions on how to make it work, but this is a slow process (CGPEG, Interviewee 3).

Although Silva et al. (2013) presents a skeptical perspective for the promotion of organizational learning in Brazil through IA/EA due to the lack of a proper system of organizational memory and public communication, the evidence of organizational learning reported in this paper indicates that the existence of individuals’ engagement within organizations can broaden this perspective and best results may be found through experience sharing in a multidisciplinary way (Fischer 2014; Gachechiladze-Bozhesku 2014).

The evidence of learning from the case studies demonstrated that strategic changes in the EIA systems may take place even when higher learning levels are not fostered or supported by the mechanisms adopted by the organization (as found in the context of IEMA), what leads us to the following assumption: higher learning levels may facilitate higher learning loops, though the first is not a must to the second.

Besides, as part of the contribution of the case studies to the conceptual model, the learning evidence corroborated the importance of stakeholder’s engagement at the organizational sphere to reach higher learning levels and the influence of the context on EIA procedures, objectives and learning outcomes.

Final remarks and conclusion

This paper aimed to understand learning in EIA systems in order to develop a conceptual model based on literature review and empirical evidence of learning as found in two EIA systems in Brazil, considering the perspective of regulatory agencies.

The evidence collected throughout this research sheds light to the levels and loops of learning that may be reached within EIA systems and reinforces the importance of the contextual factors to the learning outcomes. These aspects were reflected in the conceptual model presented in this paper, which we believe can be adapted to different contexts.

Although specific to the context, the evidence produced have corroborated the implicit feature of learning in EIA systems, thus increasing the capacity of the conceptual model to represent relevant aspects of learning that may occur in different contexts.

With that said, the model can be applied by scholars and EIA practitioners to comprehend how the different aspects of learning, as described by the literature, can be revealed throughout the EIA process. There is a promising research agenda to be experimented, which we believe can be facilitated by the conceptual model developed in this paper.

Specifically to the context of the Brazilian EIA system, our findings have endorsed the influence that contextual factors can have on the learning process. It also corroborated the key role of individuals’ engagement within organizations to provide concrete learning outcomes, as pointed out by Sánchez and Mitchell (2017).

As a limitation, we recognize the need to explore other aspects of learning in EIA systems in order to identify the evidence that may possibly contribute to enhance EIA effectiveness.

Disclosure statement

No potential conflict of interest was reported by the authors.

Funding

This work was supported by the Foundation for the Coordination and the Improvement of Higher Level or Education Personnel (CAPES); and the National Counsel of Technological and Scientific Development (CNPq) [grant number 142053/2014-7].

Acknowledgments

We acknowledge CAPES and CNPq for the financial support to this research, the contributions of CGPEG and IEMA and the insightful comments of the anonymous reviewers to this paper.

Notes

1. Information collected at the System of Federal Environmental Licensing (SILAF). Available at http://www.ibama.gov.br/licenciamento/