Results and Preliminary Conclusions
The below represent our results and preliminary key conclusions thus far, as of the end of the 2001 field season.

Total Aboveground Phytomass (TAP)
We found that total above ground phytomass (TAP) increases with summer temperature on both acidic (MAT) and nonacidic (MNT) parent materials, but the regression lines diverge as SWI increases (Figure 16). At the northern coast, the phytomass is similar for both MAT and MNT, but by the southern end of the gradient there is a 225% increase in biomass on MAT sites, and a 50% increase on MNT sites. This increase in phytomass is reflective of the increasing SWI as we move south, with more than a 3-fold increase in SWI from our furthest north site at Barrow (9¾C) to our furthest south site at Council (34¾C). A 5¾ increase in the SWI from Barrow to Quartz Creek (65¾N, SWI: 32¾C) correlates with about a 115 g m-2 increase in the aboveground phytomass for zonal vegetation on acidic sites and about 60 g m-2 on nonacidic sites. The only anomaly is the lower phytomass at the Atqasuk site (vs Barrow), but this is likely due to the sandy, leached, nutrient-poor soils at Atqasuk, which lies within a late Pleistocene-age sand sea (Carter, 1981).
Between all sites, shrubs account for most of the aboveground phytomass increase on acidic substrates, whereas mosses account for most of the increase on nonacidic soils. The phytomass of shrubs increased 12-fold in MAT as SWI increased, but MNT shrub phytomass showed no correlation with temperature (Figure 17). The reason for the discrepancy in the shrub response are major differences in the species composition of acidic and nonacidic tundras. The dominant shrubs in MAT are Salix pulchra and Betula nana. These shrubs exhibit large changes in growth-form with increased temperature. In MNT, Dryas integrifolia, Salix reticulata and Salix arctica are the dominant shrubs, and all are very short or prostrate, and show little change in height in response to temperature. The dominant MNT erect shrubs are Salix richardsonii and S. glauca, but these are usually scattered and do not form a major component of the MNT plant canopies, even at the southern end of the temperature gradient.
The large response of MNT mosses to increased summer warmth was an unexpected result and one of the most interesting observations (Figure 18). The phytomass of mosses increased by 250% on the MNT sites as SWI increased. Previous studies have noted that mosses greatly affect the thermal, hydrologic, and nutrient properties of the soils, and are one of the main factors that control the transitions of zonal vegetation from MNT to MAT near the southern boundary of Subzone 4 (Walker et al, 1998). It is possible that soil surface temperature could be greatly affected by this increase in mosses, and decrease the activity of frost boils, which play an important role in nutrient availability and a variety of other ecosystem properties that maintain the nonacidic ecosystems. This strong increase in MNT moss phytomass has not been previously observed and needs to be confirmed with further studies. The phytomass of the other plant functional types generally increased with warmer temperatures, except for the sedges in nonacidic tundra, and the lichens in acidic tundra. As we moved from the Arctic coast southward to the MNT/MAT boundary, it was very interesting to observe the changes in the dominant growth forms (Figure 20). The dominant growth forms near the coast (in order of dominance) for MAT was mosses, graminoids, lichens, then shrubs, versus on MNT it was graminoids, mosses, shrubs, then lichens. By the MNT/MAT boundary, the dominant growth forms for MAT had shifted to shrubs, mosses, graminoids, and lichens, and for MNT it had shifted to mosses, graminoids, shrubs and lichens.

 

 

Leaf Area Index (LAI)
LAI is positively correlated with SWI on acidic sites (Figure 19). There is approximately a 150% increase in MAT LAI across the more than 3-fold increase in SWI. On the nonacidic sites the LAI/SWI relationship is unclear. Since the phytomass increase on MNT sites is due primarily to the increase in mosses, the LAI of these sites could not be adequately measured as the LICOR LAI-2000 sensor used for measurements is too large to insert into the moss layer. While this instrument offers a means to quickly obtain large quantities of LAI information for the vascular-plant component of Low-Arctic vegetation, we have concluded that it is of limited use in high moss areas as well as for extremely low-growing prostrate vegetation that is common on wind-blown sites and in the High Arctic.


Normalized Difference Vegetation Index (NDVI)
The NDVI is positively correlated with SWI on both acidic and nonacidic soils, but on nonacidic parent material the NDVI is consistently lower than that of the acidic substrates (Figure 20). Mean MaxNDVI on acidic substrates varies from 0.36 to Atqasuk to 0.52 at Quartz Creek, while that on the nonacidic substrates varies from 0.27 at West Dock to 0.48 at Sagwon. We were able to gain a clearer picture of the effect of substrate (by removing climate as a variable) on NDVI at Oumalik and Sagwon, where large areas of both acidic and non-acidic tundra exist at each site. At Oumalik, the mean MaxNDVI is 0.51 on acidic soils compared to 0.37 on nonacidic soils, and at Sagwon the MaxNDVI is 0.53 on the acidic sites compared to 0.46 on the nonacidic sites. This higher NDVI in MAT vs MNT is not unexpected, and is consistent with earlier studies (Shippert et al 1995; Walker et al 1995, 1998). The regression of NDVI vs TAP provides a basis for predicting the possible effects of summer warming in much of the Low Arctic. It is a reasonable assumption that temporal changes to NDVI resulting from climate change might replicate the differences that occur along spatial climate gradients because increased warming is likely to cause increased shrub growth (Chapin et al 1996, Sturm et al 2001). The amount of shrubs is strongly correlated with NDVI and summer warmth, particularly in acidic tundra. A 5¾C increase in the SWI is approximately equivalent to a 1-2¾C increase in the mean July temperature. Since a 5¾C increase in the SWI results in about a 115 g m-2 increase in above-ground phytomass for MAT and a 60 g m-2 increase for MNT, a similar change could conceivably occur with a 1-2¾C change in the mean July temperature caused by global warming.

 

The sandy substrates at Atqasuk had the lowest productivity and NDVI of all the mesic sites, despite relatively warm temperatures compared to the coastal sites. Low nutrient availability accounted for low productivity, and relatively high lichen cover, which has low spectral reflectance in the near-infrared channels, accounted for the low NDVI values. Council presents a special situation in that with only a 2.2¾C increase in SWI above Quartz Creek (just 50 miles NW of Council), the zonal representative vegetation shifts from tussock tundra to shrub-tundra. While it is technically still within Subzone 5, its proximity to the southern coast of the Seward Peninsula and tree line suggests its classification as a special maritime variant of this subzone. Total aboveground biomass at the representative zonal site (Council C3) is more than twice as high as at Quartz Creek (Figure 16), and more than 6 times higher than Barrow. It is 2-fold higher than any of the other MAT sites, and 2.5 times higher than any of the MNT sites. This large increase in biomass above the other sites is due primarily to the significant increase in shrub biomass. Mean shrub biomass for this site was 2135 g m-2, versus that of the other 2 highest sites (Quartz Creek=427 g m-2; Oumalik=465 g m-2). Leaf area index was 2.3 (versus 0.75 at Barrow). NDVI was 4% greater than that of Quartz Creek, and 25% greater than Barrow. While total aboveground biomass was much higher at Council than at any of the other MAT or MNT sites, LAI and NDVI did not show the same dramatic increase (Figures 19 & 20). In summary, Council is a warm maritime climate that is representative of dense shrub tundras such as those in the Alaska Yukon River drainage, as well as in Russia within the Anadyr River drainage, and European Russia (west of the Urals).

The Quartz Creek site, on the other hand, presents a more realistic picture of how a system much like that of the Arctic Foothills in northern Alaska might respond to warming. Shrub biomass in the water tracks is much higher, and the tussock tundra systems display greater tussock height and more sedge biomass. While it does not support the high-biomass shrub-tundra plant communities to the extent that Council does, these communities are certainly present. Based on our analyses, it appears that climate warming in the Arctic will likely result in increased phytomass, LAI, and NDVI on zonal sites. It is also likely that acidic areas supporting abundant shrub phytomass will demonstrate the greatest changes.

 
Key Conclusions
  • Total plant phytomass, LAI, and MaxNDVI increase with temperature along the climate gradient in northern Alaska. The exception is the Atqasuk site, where the sandy substrate inhibits productivity.
  • The phytomass and NDVI are lower on MNT than on MAT. The principal cause of the different response to temperature is the different species composition of the two tundra types. The shrubs in MAT increase in height and biomass with greater summer warmth; whereas most shrub species in MNT are prostrate dwarf shrubs that show little change in stature with increased summer warmth.
  • Mosses show the greatest response (increase) to warming in MNT. This has not been previously observed and requires confirmation with further studies.
  • The dominant growth forms for MAT and MNT shift from north to south. In MAT, the dominant growth form shifts from mosses to shrubs. In MNT, it shifts from graminoids to mosses.