PARENT SESSION
Contributed Oral Session 140: Forest Photosynthesis and Water Relations
Thursday, August 11, 1:30 PM - 5:00 PM, Meeting Room 520 B, Level 5, Palais des congrès de Montréal

Water use by forests dominated by eastern hemlock (Tsuga canadensis) versus deciduous trees: Relevance to soil water and streamflow after replacement of hemlock forests killed by hemlock woolly adelgid.

Hadley, Julian^*,1, Kuzeja, Paul¹, Mulcahy, Thomas¹, ¹ Harvard University, Petersham, MA, USA

ABSTRACT- Eastern hemlock (Tsuga canadensis) is a late-successional tree species of the eastern U.S. and Canada that creates an unusual forest with a dark, cool, moist understory. The hemlock woolly adelgid (HWA; Adelges tsugae Annand), an Asiatic insect introduced to Virginia in the 1950s, has recently eliminated eastern hemlock from some areas in the eastern U.S., and it is spreading. HWA-infested hemlocks now occur from Georgia to Maine. HWA typically kills hemlocks in about 10 years but this period is very variable. Hemlocks in southern New England have died more quickly in warmer and drier locations. Young trees beneath the dead and dying hemlocks in New England are primarily deciduous, mostly black birch (Betula lenta). To understand how this species replacement will affect forest water use, in 2004 we measured evapotranspiration (ET) by the eddy covariance method in hemlock-dominated and deciduous forests less than 1 km apart in central Massachusetts. The hemlock forest is 100-200 years old and the deciduous forest, dominated by red oak (Quercus rubra), is 45 to about 125 years old. Average leaf area index (LAI) is near 4 in both forests in summer. In June and early July, the hemlock forest lost about 2 mm of water per day to the atmosphere, half of ET by deciduous forest. During this period, ET from both hemlock and deciduous forests was closely linked to incident photosynthetically active radiation (PAR) with r² values of 0.7 to 0.8 and a slope of 2.66 units ET per unit PAR for hemlock forest, compared to 5.41 for deciduous forest. However, ET from deciduous forest declined more rapidly over the summer than ET from hemlock forest. During September hemlock forest ET was about two-thirds of deciduous forest ET, which fell to near zero after leaf fall in mid-October. From mid-June to mid-September, deciduous forest ET nearly equaled total precipitation (25.1 cm) but hemlock forest ET was 10 cm less. Differences in ET clearly affected soil water content. During August and September, there was 23% (1.5 cm) less water in the top 20 cm of soil in hemlock versus deciduous forest. These differences in ET and soil water indicate that small streams draining hemlock forests could disappear in summer after deciduous trees with similar total leaf area replace the hemlocks.

Key words: hemlock, evapotranspiration, soil water, streamflow