24. Ecosystem Climate Manipulations
transport during the day and impeding dew formation at night. In some situations it may be possible to avoid unwanted microclimate effects. In the
CLIMEX facility (Beerling and Woodward 1994),
for example, ambient precipitation is gathered and
distributed inside the greenhouse, and Chapin et al.
(1995) argued that uneven microtopography allowed air circulation beneath their tundra greenhouses, reducing blockage of gas exchange. In
general, though, greenhouses significantly and unavoidably modify microclimate, and usually not in
ways that conform to global warming predictions.
Greenhouses also have undesired impacts on the
contained vegetation. The wind protection they
provide reduces mechanical damage caused by ice
crystal abrasion, flailing, and wind pruning, as well
as surface heat loss and evaporation (Kennedy
1995b). A greenhouse can be a barrier to pollinators, dispersal agents, grazers, and pathogens.
Greenhouse-induced changes in light and snow
cover affect photosynthesis, canopy morphology,
and phenology (NSFESP 1991). These effects, as
well as unintentional microclimate effects, increase
the difficulty of attributing observed ecosystem
changes to temperature elevation alone, or to the
combination of microclimate changes expected under global warming.
Despite their shortcomings, greenhouses, if
carefully planned and designed, can be useful in
studying effects of warming. There are frequently
situations in which there are no other alternatives,
such as in remote or powerless areas (Chapin et
al. 1995; Kennedy 1994). Improving greenhouse
applications requires careful microclimate control
and monitoring, with microclimate modification
achieved through appropriate selection of design
and materials. Open-topped greenhouse designs
may be one of the best. Marion et al. (1997) note
that open-top greenhouses are able to achieve a
significant temperature effect, while minimizing
most of the unwanted ecological effects.
Warming Soil Directly: Buried Wires in Soil
An alternative to greenhouses is electrical resistance wire buried in the soil. This technique was
first attempted by Rykbost et al. (1975), who buried
wires 92 cm deep in cropland; they were not attempting to simulate climate warming, but rather
were testing the feasibility of using waste heat wa359
ter from power plants to boost crop production. The
applications to global warming manipUlations were
apparent, and the technique has since been used by
others for that purpose. Van Cleve et al. (1983;
1990) used buried electrical cables to heat 400 m 2
of Alaskan black spruce forest soil to approximately 9°C above ambient. Peterjohn et al. designed (1993) and applied (1994) a system ofheating cables buried 10 cm deep and 20 cm apart in a
temperate deciduous forest soil. Using thermostats,
they maintained a nearly steady soil temperature
elevation of 5°C above ambient over a variety of
environmental conditions. To recreate their design,
Peterjohn et al. (1993) suggest a power density of
at least 67 W m - 2, and nesting experimental plots
within a larger heated area. They also recommend
creating two types of control plots, one with nonheating wires to mimic the disturbance caused by
the installation and presence of the wires, and one
completely undisturbed.
Buried wires can, if desired, achieve a constant,
controlled temperature elevation above ambient
with suitable electronic control. They do not cause
as many unintended microclimate effects as greenhouses, such as blocking precipitation and wind.
Although they require electricity, and therefore cannot be used in powerless areas, they are not too
complicated or costly, and can be applied in many
kinds of systems. Buried wires are one of the best
options for soil warming manipulations in forest
ecosystems. While this would not constitute a
ecosystem-level manipulation, Peterjohn et al.
(1993) contend that warming entire ecosystems
may be unnecessary if (and this is a big if) changes
in soil processes are the dominant ecological response to warming.
The use of buried wires has drawbacks for ecosystem warming, such as disturbance of the turf and
soil to install the wires. Although Peterjohn et al.
(1994) found that disturbing the soil seemed to
have no effects on daily average temperatures,
other soil and ecosystem processes, such as gas diffusion, water transport, mesofaunal activity, and
rooting may still be affected. Comparison with appropriate control sites may help to cancel out these
effects. Year-round simulations are not realistic, as
midwinter applications of heat via buried soil cables under a snowpack are obviously inappropriate
(NSFESP 1991). Also, the apparatus may cause
fires, as has happened in the experiments of Peter-
transport during the day and impeding dew formation at night. In some situations it may be possible to avoid unwanted microclimate effects. In the
CLIMEX facility (Beerling and Woodward 1994),
for example, ambient precipitation is gathered and
distributed inside the greenhouse, and Chapin et al.
(1995) argued that uneven microtopography allowed air circulation beneath their tundra greenhouses, reducing blockage of gas exchange. In
general, though, greenhouses significantly and unavoidably modify microclimate, and usually not in
ways that conform to global warming predictions.
Greenhouses also have undesired impacts on the
contained vegetation. The wind protection they
provide reduces mechanical damage caused by ice
crystal abrasion, flailing, and wind pruning, as well
as surface heat loss and evaporation (Kennedy
1995b). A greenhouse can be a barrier to pollinators, dispersal agents, grazers, and pathogens.
Greenhouse-induced changes in light and snow
cover affect photosynthesis, canopy morphology,
and phenology (NSFESP 1991). These effects, as
well as unintentional microclimate effects, increase
the difficulty of attributing observed ecosystem
changes to temperature elevation alone, or to the
combination of microclimate changes expected under global warming.
Despite their shortcomings, greenhouses, if
carefully planned and designed, can be useful in
studying effects of warming. There are frequently
situations in which there are no other alternatives,
such as in remote or powerless areas (Chapin et
al. 1995; Kennedy 1994). Improving greenhouse
applications requires careful microclimate control
and monitoring, with microclimate modification
achieved through appropriate selection of design
and materials. Open-topped greenhouse designs
may be one of the best. Marion et al. (1997) note
that open-top greenhouses are able to achieve a
significant temperature effect, while minimizing
most of the unwanted ecological effects.
Warming Soil Directly: Buried Wires in Soil
An alternative to greenhouses is electrical resistance wire buried in the soil. This technique was
first attempted by Rykbost et al. (1975), who buried
wires 92 cm deep in cropland; they were not attempting to simulate climate warming, but rather
were testing the feasibility of using waste heat wa359
ter from power plants to boost crop production. The
applications to global warming manipUlations were
apparent, and the technique has since been used by
others for that purpose. Van Cleve et al. (1983;
1990) used buried electrical cables to heat 400 m 2
of Alaskan black spruce forest soil to approximately 9°C above ambient. Peterjohn et al. designed (1993) and applied (1994) a system ofheating cables buried 10 cm deep and 20 cm apart in a
temperate deciduous forest soil. Using thermostats,
they maintained a nearly steady soil temperature
elevation of 5°C above ambient over a variety of
environmental conditions. To recreate their design,
Peterjohn et al. (1993) suggest a power density of
at least 67 W m - 2, and nesting experimental plots
within a larger heated area. They also recommend
creating two types of control plots, one with nonheating wires to mimic the disturbance caused by
the installation and presence of the wires, and one
completely undisturbed.
Buried wires can, if desired, achieve a constant,
controlled temperature elevation above ambient
with suitable electronic control. They do not cause
as many unintended microclimate effects as greenhouses, such as blocking precipitation and wind.
Although they require electricity, and therefore cannot be used in powerless areas, they are not too
complicated or costly, and can be applied in many
kinds of systems. Buried wires are one of the best
options for soil warming manipulations in forest
ecosystems. While this would not constitute a
ecosystem-level manipulation, Peterjohn et al.
(1993) contend that warming entire ecosystems
may be unnecessary if (and this is a big if) changes
in soil processes are the dominant ecological response to warming.
The use of buried wires has drawbacks for ecosystem warming, such as disturbance of the turf and
soil to install the wires. Although Peterjohn et al.
(1994) found that disturbing the soil seemed to
have no effects on daily average temperatures,
other soil and ecosystem processes, such as gas diffusion, water transport, mesofaunal activity, and
rooting may still be affected. Comparison with appropriate control sites may help to cancel out these
effects. Year-round simulations are not realistic, as
midwinter applications of heat via buried soil cables under a snowpack are obviously inappropriate
(NSFESP 1991). Also, the apparatus may cause
fires, as has happened in the experiments of Peter-