Fig 1.
The 5 National Park Service (NPS) units in northern Alaska make up the Arctic Inventory and Monitoring Network (ARCN). They are Bering Land Bridge National Preserve (BELA), Cape Krusenstern National Monument (CAKR), Gates of the Arctic National Park and Preserve (GAAR), Kobuk Valley National Park (KOVA), and Noatak National Preserve (NOAT). Sample nodes (n = 24, marked with black dots) are sample locations, each with approximately 20 sample plots.
Fig 2.
Vegetation units are generalized from 44 types mapped by classification of satellite imagery [24].
Fig 3.
Shrub Thickets Along Tundra Drainageways.
Elongated thickets of Salix pulchra (yellow) and alder (Alnus viridis, green areas) grow along wet drainageways in the Noatak National Preserve (27 August 2009). Areas between the drainageways support Betula nana (darker reddish areas mostly in the upper half of the photo), or tussock tundra with sparse shrub cover (brownish areas in the foreground).
Fig 4.
Thickets of alder (Alnus viridis) dominate on well-drained soils just above treeline over parts of the south slope of the Brooks Range. Trees are white spruce (Picea glauca). Photo from Kobuk Valley National Park, 17 June 2009.
Fig 5.
Mean July Temperature 1971–2000 in ARCN.
Modeled values by PRISM Climate Group [25], rounded to the nearest integer.
Fig 6.
July Mean Temperatures at Kotzebue (1935–2015; circles) and Bettles (1951–2015; triangles), Alaska.
The straight trendlines are linear regressions where x is the calendar year and y is mean July temperatures in °C: y = 0.01719x − 21.74, r2 = 0.06, P = 0.02 for Kotzebue (lower) and y = 0.01651X − 17.69, r2 = 0.05, P = 0.08 (upper) for Bettles. The regression slopes suggests an increase in July temperature of about 1.7°C per century at both stations. The irregular lines are 11-year moving averages for Kotzebue (lower) and Bettles (upper). Data from the Western Regional Climate Center [27].
Table 1.
Point-Intercept Height Classes
Fig 7.
Distribution of weighted average rooting zone soil pH values.
The count of plots is given for classes, corresponding to pH rounded to the nearest 0.5 units.
Fig 8.
Normal Date of Snow Cover Loss on the Sample Plots.
Snow loss ordinal dates are from Macander and Swingley's[36] analysis of Landsat data for the period 1985–2011.
Table 2.
Correspondence of Soil Drainage and Soil Thaw Depth Classesa.
Table 3.
Correspondence of Soil Drainage Classes and Floodinga.
Table 4.
Correspondence of Soil Thaw Depth Classes and Floodinga.
Table 5.
Coefficients of Determination (R-squared) Matrix for the Environmental Variablesa.
Table 6.
Summary Statistics of Common Potential Tall Shrub Species.
Fig 9.
Scatter Plots of Shrub Canopy Volume vs. Environmental Factors for All Potential Tall Shrub Species together.
The solid lines are curves fitted to the 95th quantile. The dashed lines mark a canopy volume of 0.5 m.
Fig 10.
Scatter Plots of Shrub Canopy Volume vs. Soil Thaw Depth Class.
The solid lines connect the 95th percentile value in each class. Note that vertical scale varies between species; the dashed lines mark a canopy volume of 0.5 m.
Fig 11.
Scatter Plots of Shrub Canopy Volume vs. Soil Drainage Class.
The solid lines connect the 95th percentile value in each class. Note that vertical scale varies between species; the dashed lines mark a canopy volume of 0.5 m.
Fig 12.
Scatter Plots of Shrub Canopy Volume vs. Soil pH.
The solid lines are normal curves fitted to the 95th quantile. Note that vertical scale varies between species; the dashed lines mark a canopy volume of 0.5 m.
Fig 13.
Scatter Plots of Shrub Canopy Volume vs. Date of Snow Cover Loss.
The solid lines are normal curves fitted to the 95th quantile. Note that vertical scale varies between species; the dashed lines mark a canopy volume of 0.5 m.
Fig 14.
Scatter Plots of Shrub Canopy Volume vs. July Mean Air Temperature.
The solid lines are sigmoid curves fitted to the 95th quantile. Note that vertical scale varies between species; the dashed lines mark a canopy volume of 0.5 m.
Fig 15.
Proportion of ARCN Land Area with Tall Shrub or Forest Vegetation, by July Mean Temperature Class.
Temperatures are rounded to the nearest °C to form classes. For example, of the land area in the 12°C class (temperature 11.5°C to 12.5°C [25]), 3.3% was in forest and 7.5% in tall shrub vegetation [24].
Table 7.
Canopy volume of potential tall shrubs in relation to flooding.
Fig 16.
Area of ARCN with Various July Mean Temperatures and Soils Favorable to Shrub Expansion.
July mean temperatures are from the PRISM Climate Group [25] modeled raster, rounded to the nearest °C. “Favorable soils” are the area within each July temperature class with soils favorable to the growth of potential tall shrub species, as mapped by Jorgenson et al. [24] and discussed in the text. Areas currently with tall shrub or forest vegetation were excluded.
Fig 17.
Susceptibility Map for Shrub Expansion on Tundra in ARCN.
Susceptibility refers to the proportion of the landscape with soils favorable to the formation of high-volume shrub canopies, and ranges from 1.0 (“High”, green) to 0 (“Low”, red). The maps portray the proportion of soils with high susceptibility in areas of present-day mean July temperature of 12.5 to 13.5°C (top), 11.5 to 13.5°C (middle), and 10.5 to 13.5°C (bottom). These simulate the areas favorable to shrub expansion at the present time and with one and two degrees of additional summer warming ("present", "+1°C ", and "+2°C ").