(introduction...)

Cumulative effects refer to the accumulation of changes in environmental systems over time and across space in an additive or interactive manner. Changes may originate from actions that are single or multiple, and similar or different in kind. A unit of environmental change attributable to an individual action may be considered insignificant because of confined spatial and temporal scales. However, environmental changes originating from repeated or multiple human actions can accumulate over time and across space, resulting in cumulative effects deemed significant.

CEA is the process of systematically analysing and assessing cumulative environmental change. The practice of CEA is complex because of the need to consider multiple sources of change, alternative pathways of accumulation, and temporally and spatially variable effects. CEA can be guided by an approach that recognizes the components of sources, pathways, and effects and distinguishes attributes specific to each component. Such guidance is particularly relevant in Canada where enactment of the Canadian Environmental Assessment Act in 1992 has simulated inquiry into the theoretical and methodological bases of CEA.

Some countries have incorporated an explicit requirement to address cumulative environmental effects in their EA legislation, for example, Canada and the United States. The requirements to analyse and assess cumulative effects reflects a broadened perspective on the nature of environmental change. This perspective acknowledges multiple perturbations, complex causation, interactive processes, expanded and permeable spatial boundaries, and extended time horizons and time lags. These attributes characterize cumulative effects, or cumulative environmental change.

CEA literature generally concentrates on pervasive, regional environmental problems. Examples include acid rain, agricultural land loss, and watershed management. Clearly there is a need for regional planning and management initiatives to address such matters, but a CEA perspective can also be incorporated into individual project EIAs. Indeed, it is essential because EIA requirements usually focus on individual projects. Such requirement, and the desirability of placing project EIAs within a broader environmental management perspective, contribute to an urgent need for practical, project-level CEA approaches.

Table 9.1 highlights the major differences between conventional EIA and CEA. The distinctions listed in the table create something of a false dichotomy; in practice, it is more a question of emphasis. Conventional EIA can be applied at the policy and programme levels in ways that mirror CEA characteristics. Similarly, project-level planning can apply many CEA properties. Thus, there is considerable fertile ground within the overlap between these two related fields. A careful attention to this middle ground will both renew EIA and ground the largely conceptual field of CEA.

Table 9.1 Characteristics of conventional EIA and CEA

Aspects	Conventional EIA	CEA
Purpose	Project evaluation	Management of pervasive environmental problems
Proponent	Single proponent	Multiple projects and/or no proponents
Sources	Individual projects with high potential for adverse environmental impacts	Multiple projects and/or activities
Disciplinary perspective	Disciplinary and, to a lesser extent, interdisciplinary	Transdisciplinary and, to a lesser extent, interdisciplinary
Temporal perspective	• Short to medium term • Continuous dispersion over time • Proposed activity	• Medium to long term • Discontinuous dispersion over time (e.g., time lags) • Past, present, and future activities
Spatial perspective	• Site-specific • Focus direct on-site and off-site impacts • Continuous dispersion over space	• Broad spatial patterns • Wide geographic areas (e.g., cross-boundary impacts)
Systems perspective	• Tendency - single ecological system • Tendency - single socio-economic system	• Multiple ecological system • Multiple socio-economic systems
Interactions	• Interactions among project components • Interactions among components of environment • Interactions between project and environment • Primarily major, direct interactions • Assumption that interactions are additive	• Also interactions among projects and other activities • Also interactions among environmental systems • Also interactions between activities and environmental systems • Major and minor, direct and indirect interactions • Expectation that some interactions are non-additive (e.g., synergistic, antagonistic)
Significance of interpretations	• Significance of individual effects interpreted • Assumption that if individual impacts are insignificant, combined impacts are also insignificant	• Significance of multiple activities interpreted • Expectation that combined impacts may be significant even though individual impacts may be insignificant
Organizational level	• Intraorganizational	• Interorganizational
Relationship to planning	• Weak links to comprehensive environmental objectives • Project-level planning • Incremental project evaluation	• Explicit links to comprehensive environmental objectives • Programme and policy-level planning • Middle ground project evaluation and comprehensive planning
Relationship to decision-making	Reactive; after initial decision to initiate activity	Proactive; anticipates future actions
Impact management	Monitoring and management of major, direct impacts	Comprehensive impact monitoring and management system

(introduction...)

Several concepts and principles contribute to the development of a conceptual framework of cumulative environmental change.

9.2.1.1 Model of causality

A cause and effect relationship exists between the perturbation and the response of the system. This causal model is fundamental to a framework of cumulative environmental change. The nature of the cause and effect relationship is complex because of multiple causation, feedback mechanisms, and variable system response.

9.2.1.2 Input-process-output model

An input-process-output model provides the elemental structure for a framework of cumulative environmental change. The three elements of input, process, and output are inherent in the notion of environmental systems, and parallel the basic parts of a stress response model. Each component is briefly elaborated on below.

Input refers to a stimulus which acts as the causative agent of change. Inputs may be differentiated by type, magnitude, and frequency. Key considerations for cumulative environmental change include whether inputs are single or multiple, similar or different in kind, continuous or discrete, short or long term, and concentrated (i.e., point source) or dispersed (i.e., non-point source).

Process alludes to the pathway or mechanism followed to transfer a unit of input into a unit of environmental change. It determines a system's ability to resist, absorb, or adapt to perturbation. Processes of accumulation may be additive or interactive. The latter implies feedback mechanisms, a concept to be included in a framework of cumulative environmental change.

Output or response represents a change in system structure (e.g., hierarchy, spatial) or system function (e.g., primary production, nutrient cycling) after perturbation. A typology of cumulative effects should distinguish changes in structure and function.

9.2.1.3 Temporal and spatial accumulation

Time and space are generic to each of the components of the input-process-output model of cumulative environmental change. A temporal perspective recognizes that in an environmental system exposed to continuous or repeated inputs, the interval between each input may be insufficient for system recovery before the next input occurs, resulting in temporal accumulation. Processes with lengthy feedback loops may contribute to time delays. The output or system response may differ over the short and long term, as the response frequently requires critical thresholds to be reached before cumulative effects are apparent.

A spatial dimension is also evident. In environmental systems subjected to multiple inputs, such as those from non-point sources, the spatial proximity between inputs may be too small to disperse each input, resulting in spatial accumulation. The additive and interactive mechanisms of environmental processes at local scales may increase and contribute to regional environmental change. System responses may also involve cross-boundary movement or alter the spatial pattern of a landscape.

9.2.1.4 Control factors

Several factors control the components of input, process, and output. They are not mutually exclusive and may act dependently or independently of each other. The factors listed below are described briefly in terms of their influence on system response to perturbation.

Boundaries. Spatial and temporal dimensions define the perimeter of a system and so distinguish it from the external environment. Boundaries determine whether inputs are foreign or internal to the system, and also establish margins to identify cross-boundary flows between systems. Cumulative effects assessments are generally characterized by broad temporal and spatial boundaries to incorporate the accumulation of environmental changes over long time frames (i.e., decades, centuries) and among spatial scales (i.e., local, regional, global).

Hierarchy. Each level of organization (e.g., individual, population, community, ecosystem) within a system operates with a degree of autonomy by functioning at time and space scales which differ from other levels in the hierarchy. Different levels of organization may be associated with varying types of system response. For example, a perturbation such as cultural eutrophication may eliminate or replace a single fish species (e.g., trout), but the aquatic system as a whole may remain intact. Thus, cumulative effects assessment investigates environmental changes within and among various levels of organization.

Organizational complexity. The degree of organizational complexity also influences the capacity of a system to respond to varying amounts of stress. Mature and complex organizations tend to resist cumulative effects characterized by small and short-term stress, but are more likely to succumb to severe and prolonged stress. Immature stages of organization are more likely to absorb or adapt to extreme events by rapid rebuilding of system structure and function. The response to cumulative effects differs between mature and rudimentary systems because as organizational complexity increases, the degree of specialization and connectivity among elements usually increases, and dynamic variability of system processes generally decreases.

Assimilative capacity. All environmental systems possess an assimilative capacity that regulates the amount of input a system can receive without degradation to a system component or process. A stress of high intensity and short duration may quickly exhaust the assimilative capacity of a system, resulting in sudden system response. A stress of low magnitude and frequent repetition may deplete assimilative capacity at a gradual rate. Low levels of repeated stress may result in increments of environmental change which accumulate over time and delay the system's response (i.e., time delay).

Thresholds. Accumulation of environmental change can result in a critical threshold. This is the point at which the intensity or duration of an input is sufficient to result in system response. Thresholds control the response function by defining the level at which a system can no longer resist or absorb inputs. Perturbations that exceed the critical threshold result in adaptation or breakdown. Analogous to assimilative capacity, critical thresholds may be reached quickly under high level of stress or incrementally under low stress levels.

Dynamic variability. Perturbation may force an environmental system or a system variable to function outside its normal operating range. Dynamic variability is a measure of the amplitude or degree of fluctuation beyond this range. It determines the capacity of a system to adapt to stress, whether extreme events or the accumulation of incremental environmental changes. Systems with high dynamic variability generally have a greater capacity to adapt to severe stress than systems with low variability.

Stability and resilience. Closely related to dynamic variability, the factors of stability and resilience also influence the manner in which systems respond to perturbation. Stability is characterized by a low degree of fluctuation around an equilibrium state and a rapid return to this state following stress. Resilience is distinguished by high variability and the ability of a system to maintain its structure and function. A system with low stability and high resilience is more likely to persist in the face of extreme stress than a system with high stability and low resilience. The latter can absorb incremental cumulative effects, but is vulnerable to change when thresholds are reached.

The influence of the control factors varies among the three components of input, process, and output so that some govern inputs (e.g., boundaries, assimilative capacity), others regulate processes (e.g., thresholds, dynamic variability), and still others determine the type of output response (e.g., hierarchy, organizational complexity, stability, and resilience).

(introduction...)

Building on the above concepts and principles, a conceptual framework of cumulative environmental change can be developed. The framework is based on an input-process-output model.

• Input is represented by sources of cumulative environmental change (i.e., human actions). Sources are characterized by time, space, and the nature of the perturbation.

• Process is manifested in pathways of cumulative environmental change which are distinguished as additive or interactive.

• Output is exemplified by the resulting cumulative effects, broadly differentiated as structural or functional.

There are three components of the conceptual framework, that is the source, pathways, and effects. They are interconnected in as much as there is a cause and effect relationship between components, together with feedback mechanisms. The various pathways may stimulate other sources of cumulative environmental change and an effect itself may become a source, or activate other pathways, of cumulative environmental change. The following discussion briefly elaborates on each of the three components of the conceptual framework.

Table 9.2 A typology describing the source of cumulative environmental changes applied to three examples of human-environment interactions

Attribute			Examples
			Construction of Hydroelectric dam	Forestry clear-cutting	CO2 emissions (from fossil fuels)
Temporal
	Scale	short	¼
		long		¼	¼
	Frequency	discontinuous	¼
		continuous		¼	¼
Spatial
	Scale	local	¼
		regional		¼
		global			¼
	Density	clustered	¼
		dispersed		¼	¼
	Configuration point	linear	¼
		areal		¼	¼
Perturbation
	Type	similar	¼	¼
		different			¼
	Quantity	single	¼
		multiple		¼	¼

9.2.2.1 Sources of cumulative environmental change

A typology to describe and classify various sources of cumulative environmental change is shown in Table 9.2. The typology differentiates sources according to temporal, spatial, and perturbation attributes. Three examples illustrate various ways in which the typology can be applied to different sources. Construction of a hydroelectric dam is typically viewed as a single, discontinuous event at the local scale. However, dam construction is a potential source of cumulative effects when characterized by multiple perturbations of the same type (e.g., several hydroelectric projects) or different type (e.g., access roads, transmission corridors), expanded spatial scales (e.g., loss of upstream terrestrial habitat as the impoundment fills, altered downstream flow), and extended temporal bounds (e.g., eventual release of methyl mercury and the gradual deposition of sediment in the reservoir).

It could also be argued that clear-cutting of a forest is a single, discontinuous event at the local scale. However, once a particular stand of trees is cut, the operators move elsewhere. Clear-cutting is a potential source of cumulative effects because the action is repeated over time and across space (see Table 9.2).

Finally, CO₂ emissions are also representative of cumulative effects. The increasing accumulation of CO₂ in the atmosphere has occurred over the long term (i.e., since the pre-industrial era) and at a global scale. CO₂ emissions come from diverse and multiple sources (e.g., thermal power plants, transportation, heating) (see Table 9.2).

This typology broadens the consideration of sources (i.e., human actions) beyond the bounded projects typically appraised by environmental assessments to include activities which are repeated over time and dispersed across space.

9.2.2.2 Pathways of cumulative environmental change

Environmental changes accumulate through different processes or pathways. As with sources of change, these pathways vary by number, type, and temporal and spatial attributes. A perturbation may follow single or multiple pathways and involve additive or interactive processes. Additive pathways are summative because one unit of environmental change can be added or subtracted from a previous unit of environmental change. Interactive pathways are multiplicative, or synergistic, in that the nett accumulation is more or less than the sum of all environmental changes. Temporally, pathways may be characterized by instantaneous processes or involve time lags. From a spatial perspective, pathways may function at local, regional, or global scales, and involve cross-boundary movement among systems at the same scale.

Cumulative environmental change generally involves processes that are characterized by a series of incremental changes in environmental components or relationships. These incremental changes are categorized into four types by Sonnatag et al.

1 Linear additive changes are distinguished by a series of small, incremental additions to, or removals of, energy or materials from a fixed large storage (e.g., a lake). Each addition or removal has the same effect as the previous increment. A linear dose-effect relationship between a contaminant and a fish species exemplifies this category of change.

2 Amplifying or exponential changes involve a series of incremental additions to or removal from a seemingly limitless storage (e.g., atmosphere). Each increment of change has a greater effect than the previous one so that system response increases over time. An example is the steady release of CO₂ into the atmosphere and the associated change in global temperature.

3 Discontinuous changes involve incremental additions or removals that are assimilated until a threshold is reached. Each increment of change that exceeds the threshold results in a response. An example is the addition of nutrients into a lake which triggers algae blooms once critical concentrations are attained.

4 Structural surprises refer to a process whereby increments of local and slow environmental changes gradually accumulate so that spatial scales are increased (i.e., local to regional to global) and temporal scales are intensified (i.e., slow to rapid rates). The result is a collection of various effects on system structures. These effects are measured by spatial homogenization of key system variables and a loss of major system functions. The spatial and temporal accumulation of wetland loss, and the subsequent change in wetland functions (e.g., groundwater recharge, biotic diversity, floodwater storage), is an example of structural surprise.

This categorization describes various mechanisms involved in the accumulation of incremental environmental changes. The categories increase in complexity as the mechanisms shift from linearity to nonlinearity, continuity to discontinuity, and uniform spatial and temporal scales to hierarchical scales. They also question a common premise that an incremental change in an environmental component or process is similar to the previous unit of change. The above categories recognize that non-linear processes, trigger mechanisms, and surprise events can intensify and amplify the effect of each successive increment. Controlling factors such as assimilative capacity, thresholds, and dynamic variability regulate the accumulation of incremental environmental changes.

9.2.2.3 Cumulative effects

Cumulative effects have been categorized in various ways. Lane et al. characterized four types of cumulative effects by their primary driving force (cause) and their basic spatial pattern (effect).

1 Type A effects are proponent-driven, large, single projects that induce environmental change over a large region (e.g., NATO low-level aircraft flights over Labrador).

2 Type B effects are proponent-driven, multiple projects (related or unrelated) that interact, resulting in spatially diffuse and complex environmental change (e.g., shoreline development along the Great Lakes).

3 Type C effects are ecosystem-driven (no identifiable proponent), catastrophic, or sudden events (natural or anthropogenic origin) with abrupt environmental changes (e.g., eruption of Mount Pinatubo in the Philippines, pollution from oil well fires in Kuwait during Operation Desert Storm).

4 Type D effects are incremental and widespread ecosystem-driven (no identifiable proponent) environmental changes attributed to diverse temporal and spatial processes (e.g., increase in the atmospheric concentration of carbon dioxide).

Table 9.3 A typology of cumulative effects

Type	Main characteristics	Example
Time crowding	Frequent and repetitive impacts on an environmental system	Forest harvesting rate exceeds regrowth
Time lags	Delayed effects	Exposure to carcinogens
Space crowding	High spatial density of impacts on an environmental system	Pesticides in streams from non-point sources
Cross-boundary movement	Impacts occur away from the source	Acid rain deposition
Fragmentation	Change in landscape pattern	Fragmentation of wetlands
Compounding effects	Effects arising from multiple sources or pathways	Synergism among pesticides
Indirect effects	Secondary impacts	Release of methyl mercury in reservoirs
Triggers and thresholds	Fundamental changes in system behaviour or structure	Global climate change

The emphasis here is on the spatial pattern and minimizes the recognition of temporal attributes. Also, effects are distinguished according to the source of the cumulative environmental change rather than characteristics inherent to different types of cumulative effects. Nevertheless, it makes a useful contribution by discriminating cumulative environmental change according to their cause-effect relationships.

A typology of cumulative effects which incorporates temporal and spatial attributes more explicitly is shown in Table 9.3. These effects can be broadly grouped into two categories. Functional effects refer primarily to the accumulation of time-dependent cumulative environmental changes. Temporal accumulation occurs when the interval between perturbations is less than the time required for an environmental system to recover after each perturbation. An example is harvesting renewable resources, such as forests and fish, at rates that exceed those of replacement. Time lags are exemplified in the continuous exposure to toxins in a food chain which may contribute to intergenerational genetic abnormalities.

Structural effects are primarily spatially oriented (e.g., space crowding, cross-boundary movement, fragmentation). Spatial accumulation is where the spatial proximity between each perturbation is smaller than the distance required to remove or disperse each perturbation. Space crowding is evident in the systematic collection of pesticide residues by farm drainage systems and its movement at a higher concentration into an aquatic system. This movement also demonstrates cross-boundary flow from one environmental system to another (i.e., agroecosystem to aquatic system). Spatial fragmentation is manifested in changes in size, shape, and contiguity of forests and wetlands in intensely farmed rural landscapes.

Other types of cumulative effects such as compounding, indirect effects, and triggers and thresholds are indicative of the manner of accumulation (Table 9.4). These types generally contribute to or manifest themselves as functional or structural effects, or both.

An attempt has been made to relate the types of cumulative effects to specific pathways of accumulation. Although all cumulative effects are potentially associated with each pathway to a certain degree, some effects are potentially more related than others. For example, time crowding, space crowding, and fragmentation are usually dominated by additive pathways. Time lags and cross-boundary movements are likely to also involve interactive processes. Compounding by definition is interactive.

Linking types of cumulative effects to pathways of accumulation enhances the understanding of system response to perturbation in two ways. First, it provides an indicator of potential cumulative effects in the future when unwanted changes in pathways occur. For example, incremental increases (additive processes) in the area and density of agricultural land drainage within a region might signify future cumulative effects such as fragmentation of wetlands. Similarly, drainage can result in space crowding by the systematic gathering of leached residue from widely dispersed farm inputs (e.g., fertilizers, pesticides), and cross-boundary movements by transporting contaminants downstream. Thus, linkage provides the basis for a predictive tool of analysis. Second, when a cumulative effect is observed and the cause is largely unknown (e.g., change in waterfowl migration, loss of ecosystem functions in a regional landscape), the association among effects and pathways can be used to trace and identify sources of cumulative environmental change. In this case, the association provides the basis for a form of hindsight analysis.

9.2.3 Conclusion

Cumulative environmental problems can be desegregated by distinguishing their sources, pathways, and effects. Application of the typology specific to each component provides a basis for identifying and analysing perturbations, mechanisms of accumulation, and temporally and spatially differentiated effects. The framework can provide guidance to conventional CEA practice characterized by multiple projects bound in time and space, as well as to innovative CEAs of temporally repetitive and spatially dispersed human actions of a non-project nature. Such CEAs can generate new information and insights about environmental changes too frequently deemed insignificant.

Table 9.4 Characterizing cumulative effects

Characteristics		Examples
Sources
	Action quantity	Single, multiple, global, cause unknown
	Action type	Similar or different, common or uncommon, human or natural, additions to or removal from environment
	Temporal characteristics	Historical, existing, or future; short, medium, or long term; low, moderate, or high frequency; continuity of actions over time
	Spatial characteristics	Local, regional or global; small or large scale; continuity of actions over space
	Proponents	Single or multiple; public or private
	Source connections	Connected, unconnected, uncertain connections
Pathways to the environment
	Environmental media	Groundwater, surface water, air, energy
	Degree of concentration	Concentrated or dispersed over time and space
	Degree of continuity	Continuous or discontinuous over time or space (e.g., time or space lags)
	Pathway connections	Connected pathways, unconnected pathways, uncertain connections across pathways
Environment
	System type	Number, type, components, structure, and function of ecological, social, economic, institutional and political systems
	Resources	Number, type and significance
	Significance	Number and type of valuable ecosystem and other ecological components
	State of environment	Healthy, impaired, or collapsed; stable or unstable; resilient or not resilient
	Environmental connections	Connected components, unconnected components, uncertain connections
Interactions
	Connection to sources	Connected, not connected, connections uncertain
	Strength of connections	Strongly connected, weakly connected
	Direction of connection	Direct, indirect, feedback
	Temporal distribution	Concentrated or dispersed; continuous or discontinuous distribution of effects
	Special distribution	Concentrated or dispersed; continuous or discontinuous distribution of effects
	Nature of connections	Additive or interactive, reversible or irreversible
	Significance of connections	Significant, insignificant, uncertain significance

There remain challenges to conduct a CEA focused on sources, pathways, and effects. Sources of cumulative environmental change that are non-project in nature are likely to involve numerous "proponents'', if they can even be identified.

The least understood of the three components of the framework is pathways of accumulation. The complexity of these pathways is evident in multiple routes, feedback loops, and processes that are interactive, synergistic, or involve compounding. Theoretical understanding and tools to identify, monitor, and analyse these pathways are not readily available.

Finally, while cumulative effects can be analysed using available information sources, empirical evidence is often scanty, and quantitative analyses of effects are hindered by insufficient data. CEA requires a temporal scan of long duration and geographic representation at various scales. The limited time span and local focus of many existing databases impede analyses at broader temporal and spatial scales. Rigorous analysis of cumulative effects requires building up the empirical base.

The field of CEA is still in its infancy, so there are few cases where CEAs for major projects have been completed. In general, thus, there is more agreement on the concept of CEA than there is on practical methodologies and techniques.