4. Atmospheric and Geologic Constraints
59
creased vegetation growth rates, and increased hydrologic N losses from
temperate forest ecosystems (Hedin et al., 1994; Aber et al., 1991; Aber,
Nadelhoffer, Steudler, & Melillo, 1989).
Here we use the chemistry of watershed streams to understand the
importance of external sources of major elements as long-term biogeochemical constraints for temperate rainforests in North and South America.
Because stream-water chemistry is also affected by internal watershed
processes, such as vegetation nutrient demand or organic matter mineralization (e.g ., Hedin & Campos, 1991; Likens et al., 1977), we use
several independent approaches to evaluate the role of atmospheric and
weathering supplies of individual elements . We focus on elements with
significant atmospheric and/or geologic sources (i.e ., Ca2+, Mg
2+
, Na+,
K+, S, Cl" and Si0 2 ) . We have not considered N or P; a detailed
evaluation of patterns of N loss from temperate rainforests in Chile can
be found in Hedin, Armesto, & Johnson (1995).
Study Areas
Temperate rainforests in North and South America are characterized by
high rates of precipitation (>1400 mm yr"": see definition in Alaback,
1991) and stream discharge (see Lawford, this volume), moist soils, a
strong influence of marine air masses, moderate temperatures with cool
summers and mild winters, and the occurrence of well-developed forest
ecosystems with relatively high contents of live and dead biomass (Alaback,
1991).
For North America, we analyzed data from three watershed studies of
temperate rain forest in British Columbia, Canada: Carnation Creek
(CC) (49°N, 125°W ; Scrivener, 1975); the University of British Columbia
Research Forest (UBCRC) (49°N, 122°W; Feller, 1987); and Jamieson
Creek (JC) (49°N, 123°W; Zeman & Slaymaker, 1978). These sites
receive >2500mm/yr precipitation, as shown in Table 4.1, and are
located in the western hemlock biogeoclimatic zone (Krajina, 1969), with
forest communities dominated by western hemlock (Tsuga heterophylla),
western red cedar (Thuja plicata), Douglas fir (Pseudotsuga menziesii) ,
and varying contributions of amabilis fir (Abies amabilis) and Sitka spruce
(Picea sitchensis). CC and JC contain relatively undisturbed old-growth
forests, while UBCRC contains a 70- to 90-year-old second-growth forest
subjected to selective logging and fire. UBCRC and JC are located on
predominantly acid igneous rock (quartz diorite), while CC is located on
metamorphic quartz-dominated substrates, with some basalt and crystalline limestone. We used published annual volume-weighted averages
for each element. We also used data from the Dennis Creek (DC)
watershed study (49°N, 119°W; Hetherington, 1976) (see Table 4.1)
located near the JC and UBCRC watersheds yet not strictly classified as a
temperate rainforest because of lower annual precipitation amounts (ca.
700mm/yr). DC vegetation is dominated by Engelmann spruce (Picea
59
creased vegetation growth rates, and increased hydrologic N losses from
temperate forest ecosystems (Hedin et al., 1994; Aber et al., 1991; Aber,
Nadelhoffer, Steudler, & Melillo, 1989).
Here we use the chemistry of watershed streams to understand the
importance of external sources of major elements as long-term biogeochemical constraints for temperate rainforests in North and South America.
Because stream-water chemistry is also affected by internal watershed
processes, such as vegetation nutrient demand or organic matter mineralization (e.g ., Hedin & Campos, 1991; Likens et al., 1977), we use
several independent approaches to evaluate the role of atmospheric and
weathering supplies of individual elements . We focus on elements with
significant atmospheric and/or geologic sources (i.e ., Ca2+, Mg
2+
, Na+,
K+, S, Cl" and Si0 2 ) . We have not considered N or P; a detailed
evaluation of patterns of N loss from temperate rainforests in Chile can
be found in Hedin, Armesto, & Johnson (1995).
Study Areas
Temperate rainforests in North and South America are characterized by
high rates of precipitation (>1400 mm yr"": see definition in Alaback,
1991) and stream discharge (see Lawford, this volume), moist soils, a
strong influence of marine air masses, moderate temperatures with cool
summers and mild winters, and the occurrence of well-developed forest
ecosystems with relatively high contents of live and dead biomass (Alaback,
1991).
For North America, we analyzed data from three watershed studies of
temperate rain forest in British Columbia, Canada: Carnation Creek
(CC) (49°N, 125°W ; Scrivener, 1975); the University of British Columbia
Research Forest (UBCRC) (49°N, 122°W; Feller, 1987); and Jamieson
Creek (JC) (49°N, 123°W; Zeman & Slaymaker, 1978). These sites
receive >2500mm/yr precipitation, as shown in Table 4.1, and are
located in the western hemlock biogeoclimatic zone (Krajina, 1969), with
forest communities dominated by western hemlock (Tsuga heterophylla),
western red cedar (Thuja plicata), Douglas fir (Pseudotsuga menziesii) ,
and varying contributions of amabilis fir (Abies amabilis) and Sitka spruce
(Picea sitchensis). CC and JC contain relatively undisturbed old-growth
forests, while UBCRC contains a 70- to 90-year-old second-growth forest
subjected to selective logging and fire. UBCRC and JC are located on
predominantly acid igneous rock (quartz diorite), while CC is located on
metamorphic quartz-dominated substrates, with some basalt and crystalline limestone. We used published annual volume-weighted averages
for each element. We also used data from the Dennis Creek (DC)
watershed study (49°N, 119°W; Hetherington, 1976) (see Table 4.1)
located near the JC and UBCRC watersheds yet not strictly classified as a
temperate rainforest because of lower annual precipitation amounts (ca.
700mm/yr). DC vegetation is dominated by Engelmann spruce (Picea