Biological Projects

Climate Change on the Kenai Peninsula
Version 1.22

Ed Berg, Ph.D., Ecologist

edward_berg@fws.gov

USFWS: Kenai National Wildlife Refuge June 21, 1999

The following paragraphs summarize recent observations about climate change on the Kenai Peninsula, and with special attention to the Kenai National Wildlife Refuge.

(1) Rapid glacial retreat. Wiles and Calkin (1994) developed a 2000 year chronology of glacial advance and retreat on the Kenai Peninsula, and found that glacier front positions on the western side of the Kenai Mountains are controlled primarily by summer temperatures, whereas glacier fronts on the Prince William Sound side are controlled by winter snowfall. They showed that glacier fronts have generally been receding since the end of the Little Ice Age in the 1860's. For example, they dated the outermost terminal moraine of Grewingk Glacier in Kachemak Bay at 1858, and showed that the ice has pulled steadily back more than 4 kms since that time.

Rice (1987) examined aerial photographs of the Harding Icefield in the Kenai Mountains and found 5% loss of ice area between 1950 and 1985. A recent study by Adalgeirsdottir (1997) from the UAF Geophysical Institute reported a 70' reduction in the thickness of the Harding Icefield between the early 1950's and mid-1990's.

(2) Rising treeline. Sitka and white spruce on the flanks of the Kenai Mountains show a strong upslope gradient to younger trees. We have found that ring-widths of these trees generally do not show a strong correlation with temperature records of local meteorological stations. This indicates that the trees are not stressed for temperature and that they could grow at still higher elevation. Physiological tree line thus appears to be advancing so rapidly that the trees have not kept up with it. Local residents in Kachemak Bay say that treeline has visibly risen at least several hundred feet since the 1940's (Yule Kilcher, pers. obs., 1997).

Furthermore, this process appears to be unidirectional, because one does not see old dead trees at treeline that might suggest that treeline has temporarily receded at some point in the past. This unidirectional character of all climate-driven processes on the Kenai is quite striking, and suggests that this climate change is a long-term trend and not an oscillating process.

(3) Wetland drying. This takes several forms on the Kenai Peninsula:

(A.) Kettle pond disappearance. The hilly moraine areas of the KNWR have many kettle holes, left by foundered blocks of ice during retreat of the glaciers about 13,000y ago. The 1950 USGS quadrangle maps and 1950 aerial photos show these kettles as water-filled ponds, but today many are grassy pans with varying degrees of spruce and hardwood invasion. They do not appear to have been water-filled in recent years, and they would no longer be mapped as wetlands. Horse packers who in the past depended on these ponds report increasingly difficulty in finding water holes for their horses during fall moose hunts (Lou Albrant of Sterling, pers. obs., 1998).

One could ask if 1950 was such a wet year that ponds had an unusually high (but transient) water table; on the contrary, the winter-summer of 1949-50 had very low total precipitation (Kenai reported 12.8" for Sept-July as opposed to a mean of 16.5", SD " 3.4"), so if these ponds were ever to be dry, they should have been dry when photographed in August 1950.

Many small ephemeral ponds used by wood frogs on the Refuge have either gone dry or their levels have dropped drastically between the first wood frog survey in 1991 and the most recent survey in 1998 (Ted Bailey, KNWR, pers. obs., 1998) .

Spruce invasion of the 'Island' soil series has been noted at least since the 1960's (Rieger et al. 1962). This soil series consists of dark silt loam andisols, usually found in small, open bowl-shaped depressions in forest uplands. Prominent hummocks provide a thick insulating sod that keeps soil temperatures low and has effectively repelled trees in the past. Comparison of these depressions with the 1950 aerial photography, however, shows rapid forest encroachment (Scott Stewart, Mike Gracz, Homer NRCS, pers. obs., 1998).

Rieger et al. (1962) reported that in the Kenai-Soldotna area many of these depressions are completely forested. They still have the hummocky surface characteristic of the Island soils, but the soils have taken on most of the properties of the surrounding 'Soldotna' soil series, which are more acidic. We thus see a continuum from water-filled kettle ponds to grassy hummock depressions (with Island series soils) to forested depressions with forest-influenced soils of the Soldotna series. We expect that a careful look at tree ages and aerial photos will show that there is again a unidirectional process here, that the process initiated within the last 100-150 years and that it has greatly accelerated since the 1950's.

(B.) Spruce invasion of wetlands. The Kenai lowlands have tens of thousands of acres of shallow lakes and marshes. Many of the marsh edges show invasion of stunted black spruce trees that appear to be living at the limit of their tolerance to water-logged soils. In some cases distinct halos of small black spruce can be seen around wetlands; in most cases the invasion is more diffuse and has no distinct boundary. We sampled black spruce at two marsh edges and found that trees 1' - 2' tall were as much as 30-40y old, with fairly even recruitment beginning in the 1950's. Like treeline, we observed no visible mortality (dead stems) in the stunted trees, which would have indicated a temporary rise of water level above what the trees could tolerate (EB, KNWR, pers. obs, 1996). This recruitment also acts like a unidirectional process.

(C.) Spruce invasion of muskegs. There are extensive glacial lake beds of Naptowne age (~16,000y) south of the Kenai River toward Kasilof and in the Anchor Point area (Reger and Pinney 1997). These are very flat with only an occasional channel for drainage. They are dominated by sedges, Sphagnum moss, ericaceous shrubs, and cloudberry (Rubus chamaemorus) on the wet end and grade into grass (Calamagrostis canadensis) on the dry end. We see every stage of spruce invasion on these lake beds, from open treeless areas to scattered stunted black spruce to closed canopy black spruce thickets. We have not aged any of these trees but would expect that most of this recruitment has taken place within the last 100-150 years, as water levels have slowly declined. Again, one does not see stands of dead trees on these lake beds, and we infer that the water table has declined unidirectional.

(D.) Falling lake levels. On the Kenai lowland there are many examples of lakes whose water levels have fallen several feet in recent years. Residential boat docks can be seen which no longer reach the water (e.g., Bernice Lake, EB, KNWR, pers. obs, 1998). In some cases we see willow, cottonwood, or alder recruitment on the newly exposed shores, but in other cases we see only herbaceous weeds which favor exposed mineral soil. These patterns suggest that lake levels have fallen within the last five years or so.

Closed basin lakes are probably the best candidates to show water table changes, because they are fed exclusively by the water table and slope runoff. Nevertheless, changes have also been observed in open basin lakes; for example, a chain of several lakes (below Upper Jean Lake) has dried to the point that there is no longer a stream flowing from lake to lake, and water level at one these lakes has fallen at least 4' below its former outfall. Abundant cottonwood shoots on the exposed shore indicate that the lake level has been down for several years but not longer. There are no flooded stems which would indicate that lake level had once been lower; so this again indicates a unidirectional process (EB, KNWR, pers. obs., 1998). Water levels have also declined significantly in Picnic, Browse, and Campsite Lakes over the past 5 years, changing the characteristics of these lakes, i.e., with increased submergent vegetation and algae (Ted Bailey, KNWR, pers. obs., 1998).

It is worth remarking that various long-term hydrological changes occurred on the western Kenai Peninsula as a result of the March 27, 1964 earthquake. These should not be confused with climate change effects. Some lake levels fell after the earthquake, such as that of Coyote, Birch, and Buteo Lakes at the end of Swan Lake Road. Coyote Lake, for example, has a broad shoreline shelf with birch regeneration dating to the late 1960's and an old shoreline ~5-6' above the present water level . Other lakes dropped for a few months until the following summer (1965), because they were seasonally perched above the wintertime low of their regional water tables, and drained when their substrates were fractured (Waller 1966). In the Snow River floodplain at the head of Kenai Lake the water table rose and killed many trees, some of which are still visible as standing snags (pers. obs., Dominique Collet of Sterling, and Dona Walker, a lifelong resident of Seward, 1998). The elevated water table was presumably caused by an eastward tilting of the Kenai Lake basin (Waller 1966) as well as compaction of sand and gravel underlying the floodplain. The numerous "ghost forests" along the eastern shores of Cook Inlet were caused by salt water intrusion into soils following tectonic subsidence of the bedrock and compaction of unconsolidated shoreline sediments (Plafker 1969).

(4) Strongly increasing temperatures at the Kenai and Homer meteorological stations. Kenai records a 2.9^oF/50y increase in mean annual temperature since the mid-1940's, and Homer records a 3.9^oF/50y increase in the same period. Summer degree-days (>60^oF) likewise increased 56 deg-day/50y in Kenai and 86 deg-day/50y in Homer. Much of this increase occurs in warmer Decembers (~9^oF/50y) and Januarys (~7^oF/50y), but summer temperatures are up 2.5^oF/50y in Kenai and 4.1^oF/50y in Homer. These are extremely strong gradients. (Data are from monthly NOAA Climatological Data Reports.)

Annual precipitation varies considerably on the Kenai Peninsula, with Kenai annual precipitation ranging from 11" to 27" with mean of 19.2 "3.7"(SD) (N=53yr), and Homer annual precipitation ranging from 13" to 38" with mean of 24.7" "5.6" (N=65yr). In spite of great year-to-year variation at both stations there is no apparent long-term trend toward lower or higher precipitation values, such as we see in the temperatures.

If precipitation is more-or-less constant and temperatures are rising, this suggests that increased evapotranspiration is the source of the declining water tables described above.

(5) Treeline chronologies. The instrumental meteorological record on the western Kenai Peninsula begins in 1932 in Homer and 1944 in Kenai. It is possible to reconstruct pre-instrumental temperatures from treeline tree rings. At tree line the trees should be stressed for temperature (and not precipitation), so a warm year should produce a wide ring and a cold year should produce a narrow ring. Such temperature-sensitive trees are good recording thermometers, and their ring-widths can be used to estimate past temperatures. KNWR Grad student Andy DeVolder recently prepared a 290 year chronology from hemlock trees growing on a north-facing slope at tree line on the Skyline Trail. He found that the hemlock ring-widths correlated best with growing season temperatures (May-July), and that growing season temps at this site have increased from a low of ~47EF in the 1810's to the present ~50EF. Like the stock market, this chronology has many local ups and downs, but the long-term trend at this treeline site is clearly upward, with ~3EF in 200 years.

(6) Drought stressed trees and spruce bark beetles. Many of the larger white/Lutz/Sitka spruce trees in mature stands show substantial narrowing of the annual rings in recent decades. Slow growing spruce trees are especially vulnerable to bark beetle attack (Hard 1985, 1987). Part of this narrowing is due to increased canopy competition as the stands have matured. Part of it, however, may be due to drought stress, which is a potentially greater problem for large trees than small trees.

Spruce bark beetle outbreaks have followed two recent periods of multi-year warm weather drought stress (the central Peninsula in 1968-69 and the southern Peninsula in 1989-1997). We have substantial tree-ring evidence of regional beetle outbreaks in the 1820's and especially in the 1880's (Berg, 1998; Fastie et al., in preparation). Andy DeVolder's temperature chronology (see above) shows a major cool period in the 1810's and a very dramatic cooling in 1876-78, presumably caused by three high latitude volcanic eruptions in 1875 and 1876. In these two cases the beetle outbreaks occur after a cool period, rather than during a warm period such as 1968-69 or 1989-1997. Probably, the key variable here is drought rather than temperature. Warm summers can certainly create drought stressed trees, but low annual precipitation can also create drought stress. We are hoping to study this problem by preparing a chronology of stable carbon isotopes (C-13/C-12) in tree rings, which should be a better measure of drought stress than ring widths.

Conclusion: climate change on the Kenai Peninsula differs in some dramatic respects from the Interior, because the Peninsula has virtually no permafrost. Melting permafrost in the Interior is producing visibly striking thermokarst on a landscape scale and there is abundant water on the poorly drained land surface. The Kenai Peninsula presumably went through this phase at some point in the last 10,000y since deglaciation, and it is now in a much drier mode. Wetland drying and falling lake levels may thus be the most visible expressions of future climate warming on the Peninsula.

Literature Cited

Adalgeirsdottir, G. 1997. Elevation and volume changes on the Harding Icefield, Southcentral Alaska. University of Alaska-Fairbanks. M.S. Thesis, 128p.

Berg, E. E. 1998. Spruce bark beetle history studies, Kenai Peninsula, interim report 1994-1997. Soldotna AK: Kenai National Wildlife Refuge. 33pp, plus figures.

Berg, E. E. 1999. Spruce Bark Beetle History in China Poot - Peterson Bays and the Kenai Peninsula, Alaska. Soldotna AK: Kenai National Wildlife Refuge. 8pp, plus figures.

Fastie, C. L., Berg, E. E., and T. W. Swetnam. 1999. The response of boreal forests to lethal outbreaks of spruce bark beetles on the Kenai Peninsula, Alaska. (in review).

Hard, J.S. 1985. Spruce beetles attack slowly growing spruce. Forest Science 31:839-850.

Hard, J.S. 1987. Vulnerability of white spruce with slowly expanding lower boles on dry, cold sites to early seasonal attack by spruce beetle in south central Alaska. Canadian Journal of Forest Research 17:428-435.

Plafker, G. 1969. Tectonics of the March 27, 1964 Alaska earthquake. USGS Professional Paper 543-I, 174.

Reger, R.D., and D.S. Pinney. 1997. Last major glaciation of Kenai Lowland. In S.M. Karl, N.R. Vaughn, and T.J. Ryherd (eds.), 1997 Guide to the Geology of the Kenai Peninsula, Alaska. Anchorage: Alaska Geological Society, pp54-67.

Rice, Bud. 1987. Changes in the Harding Icefield, Kenai Peninsula, Alaska. University of Alaska - Fairbanks, School of Agriculture and Land Resources Management, M.S. Thesis, 116p + appendices.

Rieger, S., Allen, G.W., Backer, A.D., Link, E.G., and B.B. Lovell. 1962. Soil Survey of Kenai-Kasilof Area, Alaska. USDA Soil Conservation Service, Series 1958, No. 20.

Waller, R.M. 1966. Effects of the March 1964 Alaska earthquake on the hydrology of south-central Alaska. USGS Professional Paper 544-A, 28p.

Wiles, G.C., and P.E. Calkin. 1994. Late Holocene, high-resolution glacial chronologies and climate, Kenai Mountains, Alaska. Geological Society of America Bulletin 106:281-303.

Version Number

This memo is assigned a "A version number", similar to computer software, with the expectation that the memo will be upgraded periodically to record new observations. Version 1.0 was the first version.

Last updated: September 11, 2008