Review: Groundwater in Alaska (USA)
- First Online:
- Cite this article as:
- Callegary, J.B., Kikuchi, C.P., Koch, J.C. et al. Hydrogeol J (2013) 21: 25. doi:10.1007/s10040-012-0940-5
- 789 Views
Groundwater in the US state of Alaska is critical to both humans and ecosystems. Interactions among physiography, ecology, geology, and current and past climate have largely determined the location and properties of aquifers as well as the timing and magnitude of fluxes to, from, and within the groundwater system. The climate ranges from maritime in the southern portion of the state to continental in the Interior, and arctic on the North Slope. During the Quaternary period, topography and rock type have combined with glacial and periglacial processes to develop the unconsolidated alluvial aquifers of Alaska and have resulted in highly heterogeneous hydrofacies. In addition, the long persistence of frozen ground, whether seasonal or permanent, greatly affects the distribution of aquifer recharge and discharge. Because of high runoff, a high proportion of groundwater use, and highly variable permeability controlled in part by permafrost and seasonally frozen ground, understanding groundwater/surface-water interactions and the effects of climate change is critical for understanding groundwater availability and the movement of natural and anthropogenic contaminants.
KeywordsAlaska (USA)Groundwater/surface-water interactionsPermafrostCold regionsClimate change
Panorama: L’eau souterraine en Alaska (USA)
L’eau souterraine dans l’état américain d’Alaska est essentielle à la fois pour les humains et pour les écosystèmes. Les interactions entre physiographie, écologie, géologie, climat passé et actuel, ont largement déterminé la localisation et les caractéristiques des aquifères comme d’ailleurs le rythme et l’amplitude des flux entrants, sortants et internes au système aquifère. Le climat s’échelonne du maritime dans la partie Sud au continental dans l’intérieur à l’arctique sur le Versant Nord. Durant l’ère quaternaire, topographie et nature des roches se sont combinées avec les mécanismes glaciaires et péri-glaciaires pour former les aquifères alluviaux non consolidés d’Alaska, d’où ont résulté des hydrofaciès extrêmement hétérogènes. De plus, la longue persistance d’un sol gelé, soit saisonnier soit permanent, affecte grandement la distribution de la recharge et de la décharge des aquifères. En raison d’une forte utilisation de l’eau de surface et de l ‘eau souterraine, et d’une perméabilité de la nappe hautement variable, contrôlée en partie par le permafrost et par le gel saisonnier, comprendre les interactions eau souterraine-eau de surface ainsi que les effets du changement climatique est crucial pour l’appréhension de la disponibilité en eau souterraine et du transfert des polluants naturels et anthropiques.
Revisión: El agua subterránea en Alaska (EEUU)
El agua subterránea en el estado de Alaska es crítica para los seres humanos y los ecosistemas. Las interacciones entre la fisiografía, ecología, geología y el clima actual y pasado han determinado en gran parte la ubicación y las propiedades de los flujos, así como el tiempo y magnitud de flujos hacia, desde y dentro del sistema de agua subterránea. El clima varía desde marítimo en la porción sur del estado a continental en el interior, y ártico en la ladera norte. Durante el período Cuaternario, la topografía y el tipo de roca se ha combinado con procesos glaciales y periglaciales para desarrollar los acuíferos aluviales no consolidados de Alaska y ha resultado en hidrofacies altamente heterogéneas. Además, la gran persistencia del terreno congelado, ya sea estacional o permanente, afecta en gran medida la distribución de la recarga y descarga del acuífero. Debido al alto escurrimiento y el uso del agua subterránea, y la permeabilidad altamente variable controlada en parte por el permafrost y estacionalmente por el terreno congelado, la comprensión de la interacción superficial – agua subterránea y los efectos del cambio climático es critico para el conocimiento de la disponibilidad de agua subterránea y los movimientos de contaminantes naturales y antropogénicos.
美国阿拉斯加州的地下水资源对于人类和生态系统都是至关重要的。自然地理条件、生态环境、地质条件和古往今来的气候条件之间的相互作用在很大程度上决定了含水层的位置和特性, 以及流入、流出和存在于地下水系统中的水流的流动时间和规模。州内的气候条件变化很大, 由南向北, 南部为海洋性气候, 中部为大陆性气候, 阿拉斯加北坡为北极气候。在第四纪期间, 地形和岩石的类型与冰期、间冰期过程相结合, 形成了阿拉斯加松散的冲积含水层, 导致了水相的高度非均质化。除此之外, 长期呈冷冻状态的土壤, 无论是季节性的还是永久性的, 都极大地影响了含水层源汇区的分布。由于径流量、地下水使用量很大, 且部分受永久性、季节性冻土控制的渗透性是高度变化的, 弄清地下水-地表水的相互作用和气候变化对其的影响, 对于了解地下水的可用性和自然、人为污染物的运移是非常重要的。
Revisão:Águas subterrâneas no Alasca (EUA)
A água subterrânea no estado norte-americano do Alasca é fundamental para os seres humanos e para os ecossistemas. Interações entre a fisiografia, a ecologia, a geologia e o clima atual e passado determinaram largamente a localização e as propriedades dos aquíferos bem como a temporização e a magnitude dos fluxos de, para e dentro do sistema de águas subterrâneas. O clima varia de marítimo na parcela sul do estado a continental no interior, e a ártico na Encosta Norte. Durante o período Quaternário, a topografia e o tipo de rochas combinaram-se com os processos glaciar e periglacial para desenvolver os aquíferos aluvionares não consolidados do Alasca, resultando em hidrofácies altamente heterogéneas. Além disso, o congelamento persistente do solo, seja sazonal ou permanente, afeta extremamente a distribuição da recarga e da descarga dos aquíferos. Devido ao elevado escoamento superficial e do uso da água subterrânea, e da variabilidade elevada da permeabilidade, controlada em parte pelo permafrost e pelo solo sazonalmente congelado, a compreensão das interações água subterrânea/água superficial e dos efeitos das mudanças climáticas é crítica para o conhecimento da disponibilidade de água subterrânea e do movimento de contaminantes naturais e antropogénicos.
Summarizing groundwater conditions and the results of groundwater research in an area as large and diverse as Alaska is challenging. Special considerations for understanding groundwater hydrology in Alaska include the influence of extreme variations in topography and climate from one part of the state to another. Groundwater (GW) and surface-water (SW) interactions affect groundwater availability, water quality, and habitat suitability. They vary with climate and topography and depend on such factors as permafrost (ground, soils, ice, and rocks that have remained at a temperature of less than 0 °C for two consecutive years or more), thermokarst terrain (landforms that have slumped or collapsed due to thawing of frozen ground), and the complexity of hydrofacies (Ritzi et al. 1994; Younger 1989; Anderson 1989) and bedrock fracturing (Piotrowski 2007) driven by variations in glacial processes.
It is likely that glacial processes have affected nearly all of Alaska’s aquifers, including portions at depths greater than are affected under the current climate regime (Hanor et al. 2004). Glaciation alters local and regional groundwater flow systems (Lemieux et al. 2008), not only through alteration of sediment transport with consequent subsurface sediment heterogeneity, but also through changes in hydraulic pressure and loading. High subglacial hydraulic pressures cause injection of surface water and recharge of groundwater that would not occur under atmospheric pressure (Lemieux et al. 2008; Piotrowski 2007). Thus, glaciation can increase the depth of recharge and mixing in sedimentary basins and cause dissolution of evaporites (Macintosh et al. 2012; Lemieux and Sudicky 2010). Although this may be apparent for unconsolidated alluvial aquifers, glacial processes can affect bedrock aquifers as well. For example, depending on geology and ice thickness, glaciers can cause decreases in porosity (Neuzil 2012) as well as increases in bedrock permeability due to both subglacial high-hydraulic-pressure fracturing (hydrofracturing) and load-reduction fracturing resulting from reduction of load and consequent isostatic rebound during periods of glacial retreat (Piotrowski 2007). The hydraulic conductivity of glacially and periglacially derived sediments tends to vary abruptly in the subsurface. On steep mountain slopes, glacial outwash and colluvium tend to be gravelly to cobbly in texture (Karlstrom 1964). Fine-grained eolian silts derived from glacial scouring cover much of the lower terrain of Interior Alaska (Muhs et al. 2003). These silts froze upon deposition, creating ice-rich permafrost (termed Yedoma) and low-permeability terrain. Consequently, rivers flowing out of mountain ranges in Alaska often diminish significantly or disappear into permeable glacial colluvium, only to reappear further downslope as increases in fine-grained facies promote perching and subsequent discharge of groundwater.
This paper attempts to give the reader an overview of the issues relevant to understanding groundwater in Alaska. However, this is not an exhaustive review of all groundwater-related research and issues in Alaska. That would take a far longer document. Brief overviews of physiography and climate, water quality and climate change are given, but the bulk of the text is devoted to a discussion of current and past research in the areas of hydrogeology, ice-related phenomena, and groundwater–surface-water (GW–SW) interactions.
Physiography and climate
Geology and hydrogeology
Alaskan geology is dominated by the results of collisions and interactions among oceanic and continental plates that have occurred since the Precambrian (Plafker and Berg 1994). This includes the development and accretion of terranes composed of rocks derived from continental-margin or plate-boundary-associated volcanics, fore-arc basins, and island arcs. There is only a small portion of the state that is considered to be continental in origin. Alaska is still geologically active with over 50 historically active volcanoes and frequent seismic activity (Schaefer and Nye 2008; US Geological Survey 2012b). This has implications for potentially hazardous interactions among volcanism, groundwater, permafrost, and climate change (Abramov et al. 2008; Schneider et al. 2008; Begét et al. 1996), as well as for the potential effects of earthquake activity on groundwater flow and storage (Brodsky et al. 2003; Rojstaczer and Wolf 1994). The southern (Southwest, Southcentral and Southeast) portion of the state is, in particular, a focus of volcanism, earthquakes, and faulting due to plate boundary-subduction, as the North American and Pacific plates grind into and past one another (Winkler et al. 2000; Plafker and Berg 1994). Isostatic rebound is still occurring in southern Alaska as the result of removal and or thinning of glaciers, which can affect water-table elevations, and saltwater intrusion (see section Groundwater–surface-water (GW–SW) interactions).
Ice-related features that influence groundwater hydrology in Alaska
There are a large range of impacts that ground-ice-related features have on geomorphology, surface- and groundwater flow, and infrastructure, and there are complex feedback loops among these factors and air temperature, vegetation, fire, and climate that are not completely understood. Permafrost refers to any ground, soils, ice, or rocks that have remained at a temperature of less than 0 °C for two consecutive years or more (van Everdingen 1998). Permafrost covers about 60 % of Alaska, although its extent is decreasing in concert with trends in permafrost distribution worldwide (Fig. 4; Grosse et al. 2011; Jorgenson et al. 2008). Regional permafrost may be described as continuous (>90 % lateral extent), discontinuous (50–90 %), sporadic (10–50 %), or isolated (0–1 %; Jorgenson et al. 2008; Ritter et al. 1995). Continuous permafrost tends to be thick (reaching depths greater 600 m) and spread evenly over large areas of the North Slope, and Brooks and Alaska ranges (Jorgenson et al. 2008). Discontinuous permafrost is commonly thinner (reaching depths of over 100 m) and includes areas of unfrozen ground that pierce the permafrost connecting the surface to underlying unfrozen material. Sporadic permafrost refers to areas in which zones of permafrost are surrounded by zones of permanently or seasonally unfrozen ground.
With regard to groundwater, ice-rich permafrost stores water (as ice) and acts as a geologic facies that inhibits groundwater flow. Permafrost affects flow paths, and such processes as infiltration and recharge (Woo et al. 2008). In regions of continuous permafrost, artesian conditions can lead to the formation of mounds of ice-rich sediment domes called pingos (Jones et al. 2012). These mounds can be up to tens of meters tall, and form in valley bottoms or drained lake basins. The presence of a pingo can be used to infer groundwater movement. Taliks, areas of unfrozen ground surrounded by permafrost, may also be present in areas of continuous permafrost. These thawed zones can occur in many situations. Where the thawed region beneath a lake erodes the permafrost below it, then groundwater flow can take place between the lake and deeper groundwater systems (Yoshikawa and Hinzman 2003).
Thermokarst terrain describes areas containing landforms that have slumped or collapsed due to thawing of frozen ground (van Everdingen 1998). Thermokarsting is often associated with increasing air temperatures and/or snow cover, and rapid water and heat flow, leading to soil degradation through mechanical and thermal erosion (Osterkamp et al. 2009; Yoshikawa and Hinzman 2003). The resultant complex flowpaths and surface and subsurface networks are similar to those developed in classic karst systems that develop when limestone is eroded by water to form caves. These networks of flowpaths have significant implications for alteration of recharge, GW–SW interactions, and contaminant transport. Soil pipes are sometimes present in thermokarst environments. They are elongated subsurface voids that orient subparallel to the dip of a slope and allow rapid runoff of water, sediments, and solutes. Whereas soil pipes exist in many temperate catchments (Jones 2010), they are often also present at high latitudes, possibly related to the weakened structure of cryoturbated soils and the combined thermo-mechanical erosive potential of water on frozen ground (Koch et al. 2013, this issue), and/or large transitions in hydraulic conductivity that occur at frozen/unfrozen or organic/mineral soil boundaries (Carey and Woo 2000, 2002). Understanding the geometry, extent and temporal variability of frozen ground and subsurface ice is critical to assessing both chemical transport and groundwater availability for humans and ecosystems.
Groundwater movement can also affect surface processes and even cause flooding via the formation of aufeis. Aufeis is formed when the flowpath of water underground or under ice is interrupted by a barrier or fracture causing a buildup of ice by sequential overtopping of streambanks and river ice (Wanty et al. 2010; Schohl and Ettema 1986, 1990). It forms in and along the banks of stream channels to thicknesses in excess of 3 m, covering areas as large as 30 km2 (Yoshikawa et al. 2007). Aufeis distribution in Alaska is greatest in the Arctic, especially along the northeast slope of the Brooks Range. It generally declines in density south- and westward and appears to be controlled by a number of factors including geology and elevation (Yoshikawa et al. 2007; Dean 1984). By impeding surface- and groundwater flow, aufeis can cause overbank flooding and backwater effects that can threaten housing, industry, and infrastructure. Research findings vary as to the relative contribution of groundwater versus surface water, and the geologic setting necessary to promote aufeis formation (Malenchak et al. 2011; Wanty et al. 2010; Yoshikawa et al. 2007). Aufeis has been documented as arising from a variety of sources including suprapermafrost water, ice dams in flowing rivers, or discharge from springs issuing from terminal moraines, limestone, or fractured rock. It is likely, however, that the relative contribution of surface water or groundwater is dependent on local conditions and that it can vary both spatially and temporally.
Groundwater–surface-water (GW–SW) interactions
Groundwater in Alaska and its relationship to humans and ecosystems cannot be understood without an understanding of GW–SW interactions. There are multiple connections and feedback loops between humans, climate, ecosystem processes, and the movement and physical state of water (ice-rich permafrost, aufeis) and contaminants (sorbed, in solution, precipitate). For example, construction and infrastructure can alter plant cover and soil temperature, causing thawing of permafrost, and subsequent development or coalescence of thermokarst lakes (Jorgenson et al. 2010). This may also establish or strengthen connections with subpermafrost groundwater, creating conduits for contamination and even lake drainage with potential economic consequences for fisheries and health consequences for humans and aquatic ecosystems. Because of spatially and temporally complex variations in hydrofacies, precipitation, and temperature, the timing and location of recharge, contaminant transport, and gaining and losing reaches in rivers and streams can be difficult to predict. In Alaska, these interactions can be quite complex, and, as in other regions, are tied in not always well understood ways to geologic, topographic, biological, and climatic controls (Oasis Environmental 2010; Sophocleous 2002; Tóth 1999; Winter 1999). Regional differences in GW–SW interactions may be distinguished on the basis of these as well as anthropogenic factors such as groundwater pumping and construction.
GW–SW interactions: Southeast Alaska
In Southeast Alaska, an extensive karst system has developed in marine limestone and marble, with many karst features located along fault and fracture sets (Hendrickson and Casey 2007; Aley et al. 1993). This area receives abundant precipitation, and is vegetated by conifer forests and peatlands. Both vegetation types are associated with acidic throughflow, with pH ranging from 2.4–6.4. Carbonate-rock dissolution rates as high as 1.7 mm yr−1 (Allred 2004) have been observed in peatland settings. Dye-injection tracer tests performed at Prince of Wales Island, Southeast Alaska, showed the movement of groundwater through the karst system in connection with caves, springs and streams (Aley et al. 1993).
Much of Southeast Alaska is undergoing rapid deglaciation, resulting in the highest observed rates of modern-day isostatic rebound in the world, with peak uplift rates as high as 3.2 cm yr−1 (Larsen et al. 2005). Rising land-surface elevations have the effect of lowering water tables, and consequently baseflow in streams. In bedrock-dominated environments with little fracturing, this may be only a minor effect. In shallow alluvial-fill or fractured-bedrock systems, however, there may be significant changes over time in losing (recharge to groundwater) and gaining (discharge from groundwater) reaches in streams and river systems. The saltwater–freshwater interface in coastal aquifers may also experience changes over time as a result of isostatic rebound. Deglaciation can leave behind proglacial landforms and sediments that alter GW–SW interactions through changes in geomorphology, and sediment availability and heterogeneity. For example, newly formed proglacial lakes can capture sediment leading to a cascade of effects, including downstream channel incision, long-term declines in water-table elevation, and reductions in baseflow, as were observed in the Mendenhall River and its associated aquifer near Juneau, Alaska (Neal 2009). The combination of rapid uplift rates and landscape evolution associated with recent deglaciation provide a unique opportunity to study annual to decadal-scale hydrogeologic change in a glacial aquifer system.
GW–SW interactions: Southcentral Alaska
Groundwater issues of concern to Southcentral Alaskans are dominated by those of the Cook Inlet area. In this most populous region of the state, it is important to understand the way in which groundwater pumping affects streamflow, contaminant movement, and storage. Patrick et al. (1989) used a groundwater flow model to simulate hydrologic effects of groundwater withdrawals in the Anchorage area, including effects on flow rates between different aquifers, and between groundwater and surface water. Steady-state simulations indicated that groundwater withdrawals in lower confined aquifers would lead to leakage from upper unconfined aquifers, ultimately inducing stream leakage and decreasing groundwater discharge to streams.
Groundwater in Southcentral Alaska is important not only for human uses, but also for ecological function. Fisheries are economically and culturally important in Alaska, and groundwater discharge to streams and rivers creates upwelling zones that provide thermal refugia for fish (Maclean 2003). As a result, several studies have investigated patterns in groundwater discharge to salmon stream habitat. On the Kenai Peninsula, Bellino (2009) used temperature and natural chemical tracers to quantify groundwater contribution to headwater tributaries in different geomorphic settings in four drainages on the southern portion of the Kenai Peninsula. Groundwater contribution to the stream was found to be higher in steep valleys with shallow high-gradient streams than in broad flat low-gradient streams; however, chemical tracers indicated that the subsurface residence time for groundwater discharging to the stream was longer at low- rather than high-gradient streams. GW–SW interaction in water and energy budgets of streams is dependent not only upon landscape position as in these studies, but upon seasonal cycles as well, such as in arctic or high-elevation environments. Depending on the geographic location and annual variability in climate, rainfall-runoff contributions to streams may end in October and not begin until snowmelt occurs during the following March or April.
Human-induced alterations to streamflow regimes such as dams, may influence GW–SW interactions. In response to concern about the effect of streamflow regulation by the Bradley Dam upon downstream spawning gravels, Rickman (1993) coupled synoptic discharge measurements with observations of temperature and specific conductance to investigate GW–SW interaction on the Bradley River near Homer (Fig. 5). Streamflow was observed to either remain constant or to increase downstream of the dam, suggesting that groundwater inflows compensated for flow reductions from the dam. In the Matanuska-Susitna Valley, Kikuchi et al. (2012) used hydrometric and tracer-based measurements to quantify GW–SW fluxes at multiple spatial scales along salmon-bearing Lucile Creek, near Wasilla (Fig. 5). Differential discharge measurements indicated a strongly gaining reach along the creek; the longitudinal evolution of streamwater isotopic composition and specific conductance indicated that a leaky regional confined aquifer was the source of the upwelling groundwater. In these studies, temperature and specific conductance were used as environmental tracers and indicators of groundwater discharge. As in other studies of GW–SW interaction, under the right circumstances, these relatively inexpensive measurements provide valuable screening tools for areas of potential groundwater upwelling. These studies were performed on clearwater streams, but glacial rivers also serve as pathways for salmon migration. Curran et al. (2011) identified spring-fed clearwater side channels along the turbid, glacially dominated Matanuska River, and calculated the accumulated thermal units in spawning gravels. The accumulated thermal units were found to be sufficient to provide spawning habitat for salmon, indicating that warm subsurface inflow is important in glacial rivers.
Daily tidal cycles in Southcentral Alaska influence both water quality and subsurface flow rates in coastal areas. Tidal influence on GW–SW interactions is particularly important for pink salmon, which commonly spawn in coastal streams. The Exxon-Valdez oil spill in Prince William Sound, Southcentral Alaska, significantly affected coastal ecosystems, including spawning areas for pink salmon. Carls et al. (2003) used injected artificial tracers, salinity measurements, and groundwater level monitoring to study tidal effects on solute transport to a coastal stream. Daily reversal of groundwater hydraulic gradients associated with tides was found to transport an injected tracer from a beach area into a nearby stream. This particular transport mechanism is likely responsible for oil deposition in salmon gravels upstream of the maximum tidal extent at the beach surface.
Much of Cook Inlet Basin, and indeed, much of Alaska, is covered by wetlands. Extensive wetlands delineation and mapping efforts in the Matanuska-Susitna Valley and on the Kenai Peninsula (Gracz 2011) are elucidating the role of wetlands as hydrologic features embedded in regional and local groundwater flow systems. Reeve and Gracz (2008) simulated groundwater flow through two-dimensional (2D) cross-sections in wetlands and underlying aquifers of the Kenai Peninsula. Analysis of model results indicated that simulated flow patterns were more sensitive to hydraulic properties of mineral soil and peat deposits than to recharge, and that surface runoff is likely an important component of wetland water budgets.
GW–SW interactions: Northwest Alaska
Northwest Alaska is mostly covered by continuous or discontinuous permafrost that contains both areally extensive regions of unconsolidated coarse-textured sediments as well as fine-textured sediments in some coastal and deltaic areas. GW–SW interactions occur at locations along rivers where taliks have formed and groundwater sustains baseflow in winter (Sloan et al. 1986). They also occur in bedrock. For instance, Munter et al. (1991) found that discharge at Moonlight Springs near Nome on the Seward Peninsula was closely connected with spring thaw and probably formed part of a flow system linked to the occurrence of dissolution features in marble formations as well as regional faulting. In areas of ice-rich permafrost, the potential for GW–SW interchange is limited; however, thawing can result in loss of integrity of soils and sediments. For example, many lakes in Northwest Alaska are associated with thermokarst features, and with rising temperatures, they are changing dynamically under thermomechanical erosion of nearby seasonally frozen sediments (Jones et al. 2011). Lack of GW–SW interactions can be problematic for people. In coastal and lowland communities in Northwest Alaska, permafrost and fine-textured soils limit flow and porosity, so that there is little water available for wells and little opportunity for flushing of saline or contaminated water. Thus people tend to use surface-water supplies for drinking water (Alaska Dept. of Commerce, Community, and Economic Development 2012; Clarus Technologies 2006; Dorava and Brekken 1995).
GW–SW interactions: Southwest Alaska
Permafrost is a major factor at some locations in the northern portion of Southwest Alaska, but further south where permafrost is absent, GW–SW interactions will depend primarily on geology, topography and climate. Various efforts to study GW–SW interactions are underway or have been recently completed. For example, the Southwest Alaska Inventory and Monitoring Network has initiated hydrologic and water-quality monitoring in five national park units in Southwest Alaska to provide baseline data for this hydrologic region. River basins in Southwest Alaska provide spawning habitat for salmon in Bristol Bay, the world’s largest commercial salmon fishery. GW–SW interactions in these basins, including hyporheic exchange are critical to the sustainability of this fishery. This occurs via the transfer and overwinter storage of stream nutrients in groundwater until they reenter the stream ecosystem the following spring (O’Keefe and Edwards 2002). This area is relatively sparsely populated, but gold mining has been and probably will continue to be important with several mines proposed for development in this area, including the Donlin Creek, Nixon Fork and Pebble Mines (Alaska Department of Natural Resources 2012a, b, c). Significant local and regional environmental, geological, and hydrological assessments have resulted from feasibility studies for these and other mines. At the Nixon Fork site, for example, three types of aquifers have been encountered: in loess and weathered bedrock deposits on hillslopes, in the active layers of creek beds above permafrost, and in bedrock about 240 m below the surface (Alaska Department of Natural Resources 2012b).
GW–SW interactions: Interior Alaska
Portions of the Interior fall within the zone of discontinuous permafrost, and are therefore expected to witness significant changes in GW–SW interactions related to climate change. However, monitoring the extent and stability of permafrost is a significant challenge. Regional-scale efforts to quantify GW–SW interactions in the interior have focused on large river hydrographs and chemistry, which integrate the signature of upstream landscape drainage. Changes in the ratio of summer versus winter discharge have been used as an indicator of increased flow through deep or thawing flowpaths (Walvoord and Striegl 2007). Similarly, decreased organic and increased inorganic carbon export may indicate greater mineralization in the basin, which suggests that water is flowing through longer or deeper flowpaths (Striegl et al. 2005). Uranium isotopes are an emerging tool that may improve our knowledge of these shifting flowpaths by providing evidence of the contact between water and sediments. Hydrograph separations based on uranium isotopes in the Yukon River Basin suggest the influence of multiple shallow and deep aquifers on river chemistry (Kraemer and Brabets 2012).
Research in Interior Alaskan lowlands has improved our ability to observe hydrologic changes and ascribe them to GW–SW processes. Recent work has attempted to identify the most likely physical mechanisms to explain lake loss. Talik development followed by exchange between lakes and deeper groundwater may explain recent trends in a pair of lakes in the Yukon Flats (Jepsen et al. 2012 (an article in this issue of Hydrogeology Journal)), while a larger regional study on paired lakes ascribes lake loss to thermokarsting, which increases the flow season and potential for drying of waterbodies (Roach et al. 2011). In addition, electromagnetic data collected in the Yukon Flats to map permafrost in the subsurface, provides evidence of talik formation beneath lakes in conjunction with movement of the Yukon River across its floodplain (Minsley et al. 2012). Variations in subsurface ice extent and the effect of permafrost on catchment drainage in Interior Alaska have been considered by monitoring hydrograph recessions in long-term streamflow records (Jones and Rinehart 2010). Hydrograph recessions have also been combined with uranium isotopes to observe the effect of active layer thaw on runoff and erosion of frozen ground (Koch et al. 2013, this issue). Fire may affect active layer depth by melting ice and decreasing surface albedo, which may lead to wholesale loss of shallow permafrost in coarse soils and subsequent infiltration and subsurface flow through taliks (Jorgenson et al. 2010; Yoshikawa et al. 2003). Many studies have considered the consequences of spatial variability in permafrost extent, active layer thickness, and runoff variability on transport and cycling of critical biogeochemical elements such as carbon and nitrogen (Frey and McClelland 2009). Permafrost prevents deep infiltration, and areas with high permafrost extent have been found to correspond to carbon and nitrogen export (Koch et al. 2010; Maclean et al. 1999; Petrone et al. 2006).
GW–SW interactions: Arctic
The North Slope of Alaska is underlain by continuous permafrost that can be greater than several hundred meters thick (Osterkamp and Payne 1981). This limits GW–SW interactions to shallow, suprapermafrost water except in the vicinity of the Brooks Range where faulting and limestone karst promote both groundwater recharge and discharge, and warming temperatures promote thermokarst formation and expansion (Yoshikawa et al. 2007; Bowden et al. 2008). In the foothills of the Brooks Range, GW–SW interactions are limited, but increasing as the landscape undergoes substantial change related to climate warming. Groundwater flows through certain portions of the limestone in the Brooks Range, issuing from springs and contributing to river baseflow and winter aufeis (Yoshikawa et al. 2007). Thawed regions form beneath the north-flowing rivers and grow during the summer with greater thaw beneath high energy streams (Brosten et al. 2006; Zarnetske et al. 2007), implicating hyporheic exchange as an important source of heat for melting. GW–SW interactions also arise from thermokarsting, which has increased in recent decades in parts of the Brooks Range foothills. Thawing of permafrost acts to decrease soil resistance to erosion, as ice cementation decreases and soil hydrostatic pressure increases (Gatto 1995). This reduces internal cohesion and friction, and increases soil disruption, leading to erosion and increased stream sediment and persistent nutrient loads (Bowden et al. 2008; Gatto 1995). Given that the North Slope is severely nutrient limited, thermokarst features have the potential to alter downstream primary productivity and plant growth, which are likely to have significant effects on future albedo, active layer depth, and water availability with implications and possible feedbacks through vegetation colonization and succession (Bowden et al. 2008).
The Arctic Coastal Plain is the region that slopes slightly from the Brooks Range Foothills down to the Arctic Ocean. Hussey and Michelson (1966) estimated that 50–75 % of the landscape is covered by large, shallow lakes or marshes. More recently, Arp and Jones (2009) determined that lakes greater than 1 ha occupy about 17 % of the surface area of the Arctic Coastal Plain. The suprapermafrost active layer is often less than a half meter deep, but varies with topography, vegetation, and soil properties (Nelson et al. 1997). Below this, the permafrost functions as an aquitard, promoting flooding during snowmelt and wetland persistence across the short summer. In many locations, small thermokarst ponds form in landscape depressions and grow as incoming radiation heats the water and thaws the surrounding terrain (Jorgenson and Shur 2007). Taliks likely comprise the majority of suprapermafrost-subsurface-liquid water storage and may extend tens of meters below the large lakes as shown through thermal simulations (Ling and Zhang 2003; West and Plug 2008). The large lakes are connected to varying extents by small streams that are often bounded by frost polygons. Frost polygon topography can promote water storage and decrease runoff (low-centered polygons) or it can promote drainage of the upland landscapes, delivering water, heat, organic matter, nutrients, and sediments to the lake basins via high-centered polygons and polygon troughs (Liljedahl et al. 2012; Brown et al. 1980). Increased subsurface flow may also release carbon stored in carbon-rich sediments (Kanevskiy et al. 2011; Kessler et al. 2012). While subsurface flow is limited on the North Slope, piping and seepage on steep slopes and transport through the polygon network may be an important precursor to catastrophic lake drainage and play an increasingly important role as the Arctic warms.
There are a number of factors known to increase groundwater’s vulnerability to contamination (Nolan et al. 2002; Aller et al. 1987). Several such factors are present or even widespread in Alaska including locally high hydraulic conductivity and precipitation, and shallow water tables that seasonally rise above the ground surface. Potentially exacerbating factors include abundant rivers and lakes, seasonally high runoff, and thawing permafrost. Groundwater is known to be contaminated at over 2,800 sites statewide including seven Superfund sites (ADEC 2008, 2007). Mercury contamination is a concern at some former mining sites. Natural contamination is most commonly caused by high concentrations of iron, arsenic, and manganese; however, the most common problems are related to petroleum hydrocarbons (about 80 % of sites) and wastewater.
Most petroleum contaminated sites are located near populated areas and industrial sites (Munter and Maynard 1987). Sources and causes of petroleum contamination include leaking pipes and storage tanks, fuel spills, and improper handling and disposal—Alaska Department of Environmental Conservation (ADEC; 2008). Typical examples include petroleum hydrocarbons found in groundwater near the airport in Fairbanks (Ray and Vohden 1992) and leaking above-ground storage tanks in Noorvik located in Northwest Alaska on the Kobuk River delta (Clarus Technologies 2006). Other organic compounds of concern at contaminated sites in Alaska include PCBs, herbicides, and dioxin (ADEC 2007), of which PCBs are typically associated with military facilities.
Inorganic contaminants in Alaska come primarily from natural sources or from mining. Iron, manganese, and arsenic are the most common naturally occurring inorganic contaminants and are associated with particular kinds of rocks, usually igneous or hydrothermally altered (Mueller et al. 2001). To move into ambient groundwater at high concentrations, conducive geochemical conditions of temperature, pH, and oxidation-reduction potential must be present and coupled with a hydrogeologic environment that promotes transport (i.e., high hydraulic conductivity and minerals with low sorption or reaction potential). Presence of arsenic and other elements of concern in groundwater is often related to weathering and hydrothermal alteration of sulfide minerals. Of 165 arsenic samples reported from around Alaska, more than 25 % were greater than 10 µg L–1, the current US Environmental Protection Agency (EPA) drinking-water standard (Welch et al. 2000). These samples cluster around mineral deposits and populated areas, which is typical of many types of groundwater information. As a result, they are not necessarily representative of average groundwater conditions across the state. In the Fairbanks area, arsenic and antimony in groundwater exceed primary drinking-water standards, and iron exceeds secondary drinking-water standards. Concentrations are highly variable and depend on complex bedrock geology related to intrusion of Cretaceous granite into older metamorphic Precambrian and Paleozoic quartzite, schist, and slate (Mueller et al. 2001). In Cook Inlet, of 220 samples of groundwater taken between 1969 and 1999, 65 wells contained water with concentrations of arsenic greater than 10 µg L–1 (Glass and Frenzel 2001).
The mercury mineral belt of Southwest Alaska is located primarily in the region of the Kuskokwim River delta (Gray et al. 2000). The abandoned Red Devil Mine, managed by the US Bureau of Land Management, is one of the more contaminated sites (Fig. 1). It includes multiple contaminant types in soil and groundwater. Soil samples at Red Devil Mine encountered maximum concentrations of arsenic (7,190 mg kg–1), antimony (6,100 mg kg–1), diesel-range organics (13,600 mg kg–1), and mercury (73,000 mg kg–1), requiring that this site be managed under Comprehensive Environmental Response, Compensation, and Liabilities Act and Resource Conservation and Recovery Act guidelines (Bureau of Land Management 2012; ADEC 2009). Soil borings encountered free-phase mercury at some locations, and groundwater samples ranged up to 515 µg L–1 arsenic, 1,250 µg L–1 antimony, and 49 µg L–1 mercury. The mercury concentration is well above the EPA’s maximum level of 2 µg L–1. Some site remediation has occurred, but ADEC considers it insufficient for a number of reasons. Mercury contamination in Alaska may originate from other sources such as other types of mines, refining processes (formerly used to concentrate gold from ore), some industries, and releases from coal-fired power plants both in Alaska and in Asia (Durnford et al. 2010; Jaeglé 2010). Permafrost also contains stores of mercury associated with volcanic eruptions in the geologic past as indicated by cores from northern Alaska (Schuster et al. 2008). High concentrations of particulate mercury have been observed in the Yukon River during the last decade and are derived from both thawing permafrost (Schuster et al. 2011) and deposition of aerosols emitted from industrial sources primarily in Asia (Durnford et al. 2010; Jaeglé 2010). Given high organic matter concentrations in these same waters, there is significant concern that methylation could create highly toxic methyl-mercury.
Effects of climate change
In Alaska, the effects of climate change are already apparent from changes in a number of factors, including temperature, precipitation, permafrost, landcover, rates of coastal erosion, and forest-fire frequency and extent (Barrett et al. 2011; Hinzman et al. 2005). Temperatures in the Arctic region of Alaska have increased more than in the rest of the state and the Arctic Coastal Plain is expected to experience increased temperatures and wintertime precipitation in the coming decades (IPCC 2007). Satellite-derived estimates of groundwater storage (Muskett and Romanovsky 2011) suggest storage increased in the Arctic and declined in the Yukon River Basin from 1999 to 2009. In the Yukon, these changes may be associated with increased connectivity with subpermafrost groundwater and consequent draining of lakes and wetlands. Precipitation has increased on average 10 % around the state, but this is not uniform (Shulski and Wendler 2007). For the Arctic, all global climate scenarios and models predict warming with the largest changes predicted for winter (Chapman and Walsh 2007). Decrease in sea-ice extent for the Arctic Ocean and lower atmospheric pressures in the Bering Sea are predicted to affect storm tracks and autumn temperatures affecting recharge and snow accumulation (Chapman and Walsh 2007; Shulski and Wendler 2007), and thereby permafrost distribution which is predicted to continue to decrease in extent (Grosse et al. 2011). This could have significant consequences for all groundwater budget components in Alaska, as well as substantially alter surface-water storage and runoff. Modeling and measurements to predict changes in permafrost extent by Jorgenson et al. (2010) found complex feedbacks and interactions among topography, soil, vegetation, water, and snow depth or seasonality allowing permafrost to degrade at −20 °C MAAT in some situations and to persist at MAATs as high as +2 °C. Clilverd et al. (2011) modeled water balance and climate changes in the Nome area (Fig. 1), and found that although evapotranspiration (ET) was simulated to increase by up to 40 %, precipitation was predicted to experience a similar increase, with essentially no net changes in recharge rates by the year 2099. In the Arctic, a study on ET potential concluded that additional heat energy is likely to go towards increased ground ice melt rather than increased ET (Liljedahl et al. 2011). Climate change-groundwater interactions are also affecting Alaska Native communities. Since 2003, over 30 villages have been assessed as imminently threatened by erosion and flooding (GAO 2009). This is believed to be due largely to thawing of permafrost and consequent loss of integrity of coastal and riverine sediments. Problems caused by thawing of permafrost are predicted to increase as average temperatures rise with consequent impacts to housing, industry, and infrastructure (Nelson et al. 2002).
Groundwater in Alaska is of great importance to both humans and ecosystems, but both face challenges in the face of the potential effects of climate change, resource development, and population growth on groundwater. Availability and quality vary considerably around the state as the result of natural and anthropogenic factors, as does the degree of knowledge concerning them. Thus, studies of GW–SW interactions are critical for understanding groundwater availability in Alaska. Alaska is diverse in terms of tectonic history, geology, physiography, and climate. This translates to regional spatial differences in hydrogeologic facies and groundwater hydrology with glaciers having played a critical and continuing role in the evolution of unconsolidated aquifers and groundwater recharge. Because of the seasonal and long-term variability of sediment transport and deposition associated with Quaternary climate change, alluvial aquifers tend to be highly heterogeneous. High precipitation in some areas, low cold-season ET, substantial sources of freshwater, and low-permeability permafrost in some areas make for typically high water tables or shallow accumulations of seasonal or permanent ground ice. The thickness and extent of permafrost is controlled by variations in climate, slope, aspect, soil type, and human activity. The long persistence of frozen ground, whether seasonal or permanent, controls the distribution recharge and discharge in aquifers and streams. Climate change is altering the hydrologic landscape in Alaska from changing precipitation and temperature regimes to shifting GW–SW interactions. Permafrost is thawing and, consequently, coastal sediments are eroding at higher rates, and some lakes are draining. In some cases, the change appears to be irreversible given current and predicted climate scenarios. As human population and resource development increase in Alaska, problems of contamination will increase and affect both people and wildlife, but much remains unknown with regard to the effects of these changes.
This manuscript benefited greatly from the constructive and detailed comments provided by US Geological Survey and Hydrogeology Journal peer reviewers. The project was supported by the US Department of the Interior WaterSMART Initiative and the USGS Groundwater Resources Program, Glacial Aquifer System Groundwater Availability Study.