Click here to download the entire document as a PDF
  • Return of a damaged site to a functional habitat for plants and animals without necessarily restoring the original species and processes is usually considered to be "rehabilitation," while re-establishment of habitat features, species, and processes that were present prior to disturbance is usually considered to be "restoration." While most North Slope projects are focused on rehabilitation over time spans of ten years and longer, many may eventually (over many decades) evolve into restoration projects as plant communities, ground ice conditions, and other site characteristics mature. On the North Slope, the terms "rehabilitation" and "restoration" are often used interchangeably.
  • The number of rehabilitation and restoration sites has increased dramatically over the past ten years. Currently, industry and agencies monitor well over 100 discrete rehabilitation sites on the North Slope.
  • Many sites have footprints of less than one acre, and large sites are typically no bigger than 5 or 10 acres with the exception of gravel mine sites that can exceed 140 acres.
  • Types of tundra disturbance commonly leading to a need for rehabilitation or restoration include removal of abandoned gravel surfaces (including pads, roads, and airstrips), excavation and burial of reserve and flare pits, oil spill cleanup, trenching to bury cables or pipelines, off-road travel, and gravel mining.
  • Performance standards based on vegetation cover and species richness have become the norm, but the amount of cover, the number and kinds of species, and the time allowed to achieve performance standards vary from site to site. Occasionally, other performance standards must be met, such as stabilization of ground surfaces and absence of flooding during the growing season.
  • No substantial effort has been made to synergistically assess North Slope rehabilitation needs at the landscape level that accounts for water flows, habitat fragmentation, and other factors beyond the site specific scale.
Overview and Management Relevance
To some degree, the terms "rehabilitation" and "restoration" are used interchangeably on the North Slope. However, for clarity in this document "rehabilitation" is the return of a damaged site to a functional habitat for plants and animals without necessarily restoring the original (pre-damage) species and processes, while "restoration" is the re-establishment of habitat features, species, and processes that were present prior to disturbance. Most North Slope projects are focused on rehabilitation over time spans of ten years and longer, but many may eventually (over many decades) evolve into restoration projects as plant communities, ground ice conditions, and other site characteristics mature.
Recovery of disturbed sites in the Arctic depends on the nature and the extent of the disturbance and the type of habitat affected, but even sites suffering from relatively minor damage, such as tire tracks, often require active rehabilitation efforts. Rehabilitation projects are challenged by environmental conditions on the North Slope, such as generally limited precipitation during the growing season, a growing season of less than ten weeks a year, low soil nutrient availability, extreme cold winter temperatures, slow rates of soil development, the presence of shallow permafrost, grazing by geese and caribou, and the scarcity of plants that routinely spread onto disturbed ground by seed (Chapin et al. 1992, Jorgenson and Joyce 1994, UC 2009).
Protecting soil from hydraulic and thermal erosion (themokarst) has been one goal of tundra rehabilitation. Often, this goal has led to a desire for establishment of plants as quickly as possible with the belief that plant cover will restore soil thermal equilibrium and will prevent hydraulic erosion, but neither of these beliefs is universally correct. In the past ten years, rehabilitation efforts emphasizing erosion control have been replaced or supplemented by efforts emphasizing improved aesthetics of disturbed sites (McKendrick 1999a), with performance standards typically focusing on establishment of plant cover likely to lead to a gradual return of plant communities typical of those found in the surrounding tundra (Streever et al. 2001).
Although large scale field abandonment and the subsequent rehabilitation or restoration of the greater than 10,000 acre infrastructure footprint on the North Slope lies in the indefinite future, various activities lead to ongoing needs for rehabilitation and restoration. These activities include intentional removal of abandoned gravel infrastructure (roads, pads, and airstrips), excavation and burial of reserve and flare pits, oil spill cleanup, trenching to bury cables or pipelines, off-road travel, and gravel mining. About five to ten new rehabilitation or restoration projects are initiated on the North Slope each year. Earthworks are typically followed by seeding, fertilization, and other attempts to improve growing conditions, in general accompanied by ten years of monitoring, with each site usually monitored once every two to three years. Monitoring reports are completed by the private sector and submitted to relevant agencies for approval. In addition, an annual joint agency-industry meeting and site visits facilitate ongoing communication about site development.
Questions Provided by the Senior Staff Committee
The NSSI Senior Staff Committee provided the Science Technical Advisory Panel (STAP) with six key questions relevant to restoration and rehabilitation practices on the North Slope. These questions are addressed below:
  1. Under what circumstances is actual restoration to predisturbance conditions reasonably likely to succeed within less than twenty years?

    Existing experience suggests that restoration within a reasonable time frame (less than ten years) may be possible where impacts do not damage the vegetative mat and shallow permafrost. For example, restoration may be possible after ice roads, pads, and airstrips melt, after winter seismic trails have been abandoned, after some tundra fires, and, occasionally, after light damage by off-road traffic.

    Ice Road Restoration: Ice roads, pads, and airstrips (for convenience, referred to collectively here as "ice roads") have been built since the mid 1970s and are the preferred method of logistical support for exploration access on the North Slope. Ice roads are also used extensively for the pipeline construction and maintenance. The construction of ice roads may damage tundra vegetation by affecting the plant’s thermal environment at the base of the ice road or through scraping or compression. Experience shows that the greatest impact occurs when ice roads are constructed in low snow areas on dry upland tussock tundra sites, where the ice can damage graminoids, shrubs, forbs, and bryophytes. Wetland areas show little or no impact from ice road construction because these areas are seasonally inundated with water and at the time of ice road construction are already frozen. Studies of an ice road constructed in 1978 showed that twenty four years later species composition and density of the vegetation had fully recovered comparable to unimpacted reference area conditions (Guyer, BLM Tech Rpt 2005). Analysis of soil profiles showed no significant variation in active layer thickness on multiple ice road locations and no evidence of thermokarst formation. Alaska Department of Natural Resources has monitored active layer depths on ice road routes the summer after construction since 2003. In all cases but one they found no significant difference in active layer depths between the ice road and the adjacent undisturbed tundra (Schultz, DNR, pers. comm.). In several cases DNR did see high levels of tussock damage from roads constructed during low snow winters. In all of these instances there was very little exposure of bare ground and no sign of thermokarst formation. Due to the low level of impacts associated with ice roads, vegetation damaged from ice road construction typically requires no restoration treatments and recovers naturally within ten years.

    Seismic Trails Restoration: Research shows that the vegetation impacted by winter seismic trails can recover without intervention within eight years (Emers et al. 1995, Jorgenson unpublished). Critical factors affecting the impact on vegetation caused by seismic trails are depth and character of snow, ground pressure, number of passes by vehicles, vehicle types, terrain conditions, and types of vegetation. In recent history, improved methods for seismic exploration (e.g., low pressure tires for vehicles, use of tracked vibrators, and use of tracked vehicles with less aggressive tracks) have reduced the damage to tundra by reducing the weight, tracks, or number of vehicles used. In the Arctic National Wildlife Refuge, studies showed minor effects and rapid recovery on trails in flat areas of wet tundra. About one percent of seismic trails cause disturbances that require a much longer recovery period or may cause permafrost subsidence and erosion (Emers et al. 1995, Jorgenson unpublished). Current permit stipulations and common practices have reduced tundra damage from seismic activities and in most cases restoration does not require active intervention. In some cases survey areas have been reduced or surveys canceled due to inadequate snow. On those sites where damage does occur, the most appropriate method of restoration is to replace dislodged areas of tundra vegetation and apply fertilizer to the disturbed area.

    Burn Site Restoration: Observations of vegetative recovery of the Anaktuvuk River Fire (2007) have shown that virtually no lichen cover remained in the burn area. In some areas extensive thermokarsting developed. Willows and grasses resprouted and by 2009 some willows were more than 25 cm tall. Normally, tundra fires burn the low-lying vegetation and surface of the peaty soils and rarely affect the thick organic mat, encouraging the re-sprouting of plants sometimes within days or weeks after the fire. As the climate on the North Slope warms and if the tundra environment dries, it is anticipated that more severe fires may burn away the surface peat and organic-rich soils. This may require decades for damaged areas to recover and in some cases may permanently alter plant composition. To date, sites burned by tundra fires have been allowed to naturally restore without intervention, but if fire severity increases intervention may be necessary.

    Light-damage by Off-road Traffic: Tundra is occasionally damaged by off-road traffic—for example, by vehicles breaking through ice roads, by inappropriate use of tundra-approved vehicles (such as rolligons), and by vehicles that accidentally leave the approved road system. When damage is light, such as crushed vegetation, restoration can be achieved without intervention or with a combination of seeding and fertilization. Heavier damage can require rehabilitation approaches similar to those used following tundra excavation.

  2. Under what circumstances is rehabilitation likely to succeed?
    Rehabilitation efforts on the North Slope are most often associated with removing unused gravel (abandoned roads, pads, and airstrips), excavation and burial of reserve and flare pits, oil spill cleanup, trenching to bury cables or pipelines, and abandonment of gravel mines.

    Gravel Pad Rehabilitation: Gravel pads left in place generally support little vegetation, probably because of a lack of essential nutrients, low organic matter content, and poor moisture retention. When gravel pads are surrounded by typical North Slope tundra, the absence of a seed source for upland species capable of surviving on gravel may also be a factor. In general, natural revegetation of gravel pads without active intervention usually leads to less than 1% vegetative cover after three or more growing seasons (Kidd et el. 2004). Nevertheless, experiments at BP’s Put 1 gravel pad and growth at BP’s Seed Nursery Pad show that substantial vegetation cover can be achieved on gravel pads under some circumstances if appropriate seeds and fertilizer are introduced.

    Beginning around 2000, removal of all or part of the gravel constituting abandoned pads, roads, and airstrips became a common practice. Although achieving elevations comparable to those of the surrounding tundra remains challenging, this approach offers some promise as a means of achieving rehabilitation objectives. Early concerns about massive ground ice loss and subsequent formation of lakes following gravel removal have not materialized. Gravel removal, followed by some melting of shallow ground ice (especially within ice wedges) leaves behind an uneven surface and leads to a polygonized landscape similar to that of natural tundra in at least some cases. The removal of gravel makes nutrients, organic matter, and soil moisture available for seeds, possibly increasing the number of species that can be planted successfully (Jorgenson, Joyce 1994). Some studies of treatments on gravel removal sites showed a total live vascular cover of 15 percent following seeding of native grass cultivars and three growing seasons. In some cases, grass cultivars die off after 7 to 10 years, allowing replacement by locally abundant species. Since about 2004, seeding of native grass cultivars has only been used in areas where rapid plant cover is desired, in part because grass cultivars may slow the development of a more natural plant community. Instead, fertilizer is applied and seeding occurs by natural recruitment or active seeding of selected native species, including, at a few sites since 2010, sedge seeds. Experience suggests that 10 percent plant cover by native species after ten years is a realistic projection, on average.

    Excavation and Burial of Reserve and Flare Pits: Gravel pads are often accompanied by reserve pits and flare pits, and when gravel pads are removed reserve and flare pits are often also excavated (to remove contaminated material) and then backfilled to slightly above grade. Backfilling to above-grade elevations prevents ponding of surface waters and limits the spread to tundra of any contaminants inadvertently left behind. Backfilling is usually accomplished with a combination of gravel and mineral soils, with about 20 cm of mineral soils placed on top of gravel backfill. Often, grass cultivars are seeded to encourage rapid establishment of plant cover and to prevent erosion, although the actual benefits of seeding with grass cultivars have not been documented.

    Tundra Rehabilitation after Spill Cleanup: The tundra rehabilitation methods used at spill sites vary depending on the nature of the spill and the approach to spill cleanup. In the most extreme cases, spill cleanup requires excavation of tundra to remove fluids that have migrated into the soil. The excavated area can be backfilled with gravel topped with overburden, and the overburden can be seeded with grass cultivars or allowed to revegetate naturally. Since 2006, tundra sodding has been attempted on several spill sites. Sodding involves harvesting native tundra from an area slated for destruction (such as a site where mining is planned), transporting sod to the spill site, and placing the sod, usually with fertilizer. Based on very limited experience, this approach seems to lead to rapid rehabilitation and possibly even restoration of communities similar to those of the surrounding tundra within a few years. However, it is labor intensive and expensive, and only a limited quantity of sod is available from donor sites slated for destruction (e.g., mine sites). In at least two instances, lemming grazing led to extensive but temporary damage. This method would work equally well on many excavated and backfilled sites, but to date has only been used on spill sites. Only one site—the ~0.15 ha L3 spill site—has been entirely sodded. Pipeline and Power Cable Trenches: Pipelines and power cables are occasionally buried in trenches cut into tundra. Trenching, under some circumstances, leads to thermokarsting one to several seasons after ground disturbance, leaving behind linear depressions that can trap water and alter surface hydrology. Repeated backfilling of entire trenches or portions of trenches, sometimes supplemented by seeding with native grass cultivars or placement of sod in selected areas, can arrest thermokarsting and stop water flow, eventually allowing recovery of vegetation. Although trench restoration sites do not typically cover large areas, they can stretch for long distances across remote tundra that is difficult to access. Also, thermokarsting can be insidious, occurring from several to many years after initial trenching and even recurring after stabilization appears to be complete, thus requiring ongoing monitoring. On the other hand, narrow trenches that are not substantially affected by thermokarst can support communities similar to those of the surrounding tundra within a few years.

    Mine Site Rehabilitation: The opening of the North Slope to petroleum development led to a need for large quantities of gravel for the construction of gravel roads and infrastructure. In the early 1970s, most gravel was collected from shallow scrapes in river floodplains that presumably recovered as new gravel moved along river beds. However, concerns over floodplain impacts led to a state policy in support of gravel pit mines. Developing a gravel pit mine changes the hydrology and soil properties of the mined area so drastically that the original vegetation cannot be restored. Some gravel pits were allowed to flood after all usable gravel had been removed. It was discovered that many of the deep flooded gravel pit sites that do not freeze to the bottom can provide overwintering habitat for fish. Stream channel access, flooding and shoreline modification creating littoral zones have been used to enhance aquatic ecosystems in the Arctic to create fish and migratory bird habitat in rehabilitated gravel pits. Shallow areas in flooded mine pits, as well as the surrounding disturbed tundra, can be rehabilitated using methods similar to those used for gravel pad rehabilitation.

  3. What are appropriate monitoring methods (how often and to what degree)? What measuring techniques should be used?

    Monitoring of North Slope rehabilitation sites is undertaken almost exclusively to track progress toward performance standards stipulated in plans and permits. In general, North Slope rehabilitation performance standards, and therefore monitoring, focus on plant cover, plant species richness, and ground surface stability.

    Most plant cover monitoring relies on variations of the walking point method, in which a trained botanist records the number and species of "hits" along a transect or transects. The point may be the tip of a cane or, as has become common over the past decade, the tip of a laser light held in an easily moved frame; the frame projects the light downward, the point is read, and the frame is moved to the next point. For most sites 500 points are sampled, usually at roughly 1-m intervals along 5 or more transects. In some cases, only the first hit (the "canopy hit") is counted, but in other cases multiple hits are counted, making it possible to achieve more than 100% cover. In comparing plant cover among sites, it is important to understand the methods used and, especially, whether or not multiple hits or only canopy hits are counted. Species richness counts are often done using data from walking point transects, but they can also be done using random searches (or "drunken walk" searches). In some cases, only those plant species occurring with some pre-established frequency—such as 0.2%—are included in species richness counts, so that "trace" species do not contribute to the overall species richness on restoration and rehabilitation sites.

    While mosses, liverworts, and lichens may contribute substantially to ground cover at some sites, and while they contribute to soil development over time, in general only vascular plants are included in plant cover and species richness assessments. This convention was established some time around 2001.

    Monitoring of surface stability can be done subjectively—for example, by looking for obvious signs of thermokarsting—or by undertaking surveys of transects or grids across entire sites. Some surveys are conducted using laser levels while others are undertaken by specialized survey crews using survey-grade differential GPS to establish actual elevations relative to a known stable benchmark.

    A repeated sampling effort undertaken over two summers showed that plant cover estimates vary seasonally and among sampling crews. Although no repeated sampling method has been attempted for elevation monitoring, frost heave and differences in the approaches used by different crews almost certainly lead to inconsistencies. Although plant cover, plant species richness, and surface elevation assessment methods are not entirely consistent, they can provide a reasonable assessment of site progress provided that the possibility of inconsistency is understood by field crews, regulators, and managers.

    Prior to about 2000, monitoring was often required for as little as three years, but early experience showed that three years of monitoring did not adequately document site performance, especially for gravel removal sites. For some trenching sites, monitoring may be required for less than five years. On gravel removal sites and other sites involving excavation and backfilling, monitoring is commonly required for ten full growing seasons, especially on gravel removal sites. However, requirements do not usually call for annual monitoring, but rather for monitoring every three years.

  4. Is one performance standard appropriate, or are different standards needed for different vegetation types?

    In general, North Slope restoration and rehabilitation performance standards focus on plant cover, plant species richness, and ground surface stability. The predominant parameter for gauging the success of plant establishment in the Arctic has been total live vascular cover (TLVC). Over the past twenty years performance standards or the expected result of TLVC has varied based on the goals of each rehabilitation/revegetation project, but in general there has been a transition from a focus on rapid revegetation without regard to species composition to a focus on slower revegetation moving toward plant communities that are similar in species composition and cover to undisturbed sites, with the recognition that many decades may be required for plant cover to even approach that of undisturbed sites.

    The document North Slope Plant Establishment Guidelines Table, developed by industry with input from numerous sources, recommends treatments based on both "wet" and "dry" conditions. On all treatments except for ponds and pond margins the guidelines recommend a predicted outcome of not more than 10% TLVC after ten years. While the long-term goal may call for greater cover, reaching the goal of 10% TLVC offers reasonable assurance that the vegetation will survive and spread over time. The 10% TLVC recommendation can be traced to work undertaken by the Corps of Engineers in the 1990s, in which all available data on North Slope restoration and revegetation cover data were regressed using multiple regression methods, with time (years since project inception) as a primary factor. Plant species richness is typically set at 5 species with at least 0.2% cover each. This standard, which has become to some degree a convention, appears to be entirely arbitrary and likely resulted from discussions between industry and regulatory agencies during the establishment of rehabilitation and restoration plans between 2000 and 2004.

    Standards for surface stability are often subjective, suggesting, for example, that the ground surface should not show evidence of extensive thermokarsting. In some trenching projects, more quantitative standards have been put in place. For example, some trenching projects have standards that prohibit subsidence and flooding during the growing season along trench lines greater than ten meters long. This standard, which has become to some degree a convention for trenching sites, again appears to be arbitrary and likely resulted from discussions between industry and regulatory agencies during the planning of gas pipeline trenching trials at two North Slope sites (MS3 and Put 27) and one site about 40 kilometers north of Fairbanks (the Washington Creek site).

    Performance standards should be consistent with management goals (which should be clearly stated in rehabilitation and restoration plans), appropriate for the ecology of the site, and realistic based on the specific techniques administered. For example, one project’s goal was to establish 35% TLVC after three years from the date plants were broadcast. The evaluation three years later showed TLVC was only 15% and ten years from the date had only achieved a TLVC of 18%.

  5. Is fertilizing effective and under what conditions?

    Even though nitrogen and phosphorous are naturally occurring in arctic soils they are not in organic forms in high concentrations and they are not readily available to seedlings. To compensate for the lack of available nutrients, fertilizer has been used to help reestablish vegetation. Fertilization applied to grass cultivars on gravel rehabilitation sites has been shown to increase plant cover for up to several years, possibly facilitating soil development and enhancing the moisture regime. Important factors affecting the success of fertilization are soil properties, amount and timing of fertilization, presence of contaminants, and density of seed application, seed quality, grazing, and weather. In almost all cases fertilization has a beneficial effect on seeding projects of native cultivars or on projects where indigenous plants are established through natural colonization. Experience and some studies suggest that seeding grass cultivars followed by fertilization on gravel pads may also inhibit the natural colonization of other indigenous plant species.

  6. Seeding: What, when, where, and how much?

    Seeding of restoration and rehabilitation sites with native seed and cultivars is a common practice. The seed used depends on a number of factors, including seed availability, site conditions, and project goals and performance standards.

    Since the late 1990s it has been apparent that planting grasses on dry gravel sites is not self-sustaining. On sites at tundra grade with mineral or organic soils, seeding native grass cultivars often promotes rapid establishment of plant cover but probably delays development of plant communities typically found in natural tundra. Native grass cultivar seeds are commercially available and are commonly applied to sites where rapid plant establishment is considered desirable, such as sites where potentially contaminated soils have been capped, leaving above-grade elevations.

    Seeding of indigenous forbs that naturally colonize gravel along arctic streams has been found to be successful in establishing plant cover on drier gravel sites. In most cases North Slope native forbs are not commercially available and must be collected from donor sites. At least one donor site has been intentionally cultivated in the Prudhoe Bay oilfields—that is, one abandoned pad near the Prudhoe Bay Operations Center, known as BP’s Seed Nursery Pad, has been heavily seeded with Oxytropis.

    Projects where gravel has been removed must allow time (up to two years) for the soil thermal regime to stabilize before successful planting can occur. Removing all of the gravel accelerates thaw subsidence, which has the desirable affect of leading to a broken and natural tundra appearance while also providing differing dry and hydric soil conditions. Determining the "dry" and "wet" conditions of the soil is critical in seed selection and planting strategies.

    Numerous site preparation techniques to improve soil moisture and nutrient availability, and therefore seeding success, are being investigated. Of all site preparation treatments, the addition of organic soil shows the greatest promise, at least on restoration and rehabilitation sites with only a gravel substrate or a mineral soil substrate. In some instances organic-rich soil has been used to provide a bed for reseeding. The addition of organic soil improves soil moisture retention and increases nutrient availability. Even without fertilization the amount of available nitrogen remains high in areas treated with organic topsoil.

    Seeding density depends on the species being planted, the conditions in which they are planted, and site goals and objectives. Industry, with input from numerous sources, has provided written guidelines on seeding rates under different circumstances.
  7. Recommendations
    1. Guidelines that address rehabilitation of gravel roads and pads and rehabilitation of gravel mine sites have been written by the state of Alaska and the oil industry. These guidelines, the North Slope Plant Establishment Guidelines Table (BP and ConocoPhillips, 2004) and the North Slope Gravel Pit Performance Guidelines (ADF&G 1993) should be used when planning and implementing rehabilitation projects. These guidelines should be reviewed at ten-year intervals to incorporate new information based on monitoring results.
    2. Industry, state, federal and local agencies currently meet annually to discuss progress on existing rehabilitation projects and plans for future restoration projects. In addition, managers and regulators jointly visit restoration and rehabilitation sites annually. These annual meetings and site visits should continue.
    3. Over the past ten years, restoration and rehabilitation methods have evolved rapidly, but no single report or paper has been generated to summarize these advances. A summary report or paper should be prepared to summarize the status of the science, possibly using this issue paper as a starting point but providing more detailed information and data analyses from monitoring reports on numerous different sites of different ages.