360
john and coworkers. Another serious criticism of
this technique is that it creates an unrealistic vertical temperature profile in the soil, affecting aboveground ecosystem components only indirectly and
deeper soil more than surface soil. This is opposite
to global warming, which involves increased transfer of heat to the soil from above. Buried wires do
not warm foliage and thus do not provide a wholeecosystem warming method. Despite the important
role of soil in ecosystem function, a soil-only
warming experiment is not a sufficient substitute
for a whole-ecosystem manipulation.
Warming Soil Directly: Fluid-Filled Pipes
at the Soil Surface
A method similar to buried electrical wires but with
fewer disadvantages is heating the soil surface with
warming pipes. Hillier et al. (1994) used a system
of fluid-filled, stainless steel pipes at the soil surface
to produce a continuous, constant temperature differential. They used this pipe system in a meadow
in southern England to manipulate the length of the
growing season and to simulate a milder winter.
Each warmed plot was paired with an unheated reference plot of similar vegetation structure and with
a dummy pipe network to cancel out shading effects
introduced by the apparatus. The pipe network was
supported by a metal grid, which also functioned
in the lateral dissipation of heat, creating a more
uniform temperature distribution.
There are several advantages to this type of system. Unlike buried electrical wires, this method
does not create an unrealistic vertical temperature
profile in the soil, nor disturb the soil surface or
low-lying vegetation. And unlike greenhouses,
there is not blocked precipitation, reduced wind, or
light attenuation. The apparatus can be used to produce heating or cooling, enabling a greater variety
of climate manipulation scenarios. For bare soil or
low-lying vegetation, this method is a good one for
manipulating foliage as well as soil temperature.
For elevating temperature in ecosystems with taller
plants, Hillier et al. (1994) mention the use of infrared heaters suspended over the vegetation, to
which we now tum.
Increasing Downward Radiation:
Overhead Heaters
In global warming, the total heating effect is the
sum of the direct infrared (IR) radiative effect of
increased greenhouse gas concentrations and the reKarin P. Shen and John Harte
suIting feedback effects involving the hydrologic
cycle, atmospheric circulation, and, in ways that are
less well understood, the biota, all operating from
local to global spatial scales. Although elevated
greenhouse gas levels result in surface warming because of both enhanced downward IR and because
warmer air convects and conducts more heat to ecosystems, the consequences of both the direct and
feedback effects on surface warming can be simulated by raising the downward IR radiation level
appropriately.
Manipulating radiation can be preferable to other
warming methods for many reasons. Increasing the
downward IR radiation mimics the "top down" nature of heating that actually occurs under an atmosphere with increased levels of greenhouse
gases. Nijs et al. (1996) found that applying heating
from above maintains the natural temperature gradients within the canopy. Infrared lamps transfer
heat noninvasively, and there are few unintended
effects on microclimate. Year-round manipUlations
are possible, although in areas with very deep winter snowpack, the heaters may become covered if
mounted too low, thus temporarily creating an unrealistic warming scenario. Greenhouses, buried
wires, and surface pipes all have greater problems
with application in winter, however, so comparatively, overhead heaters are a better option.
Overhead IR lamps can be thermostated to
achieve a constant increase in soil temperature, but
there is no reason to believe that forcing a continuous, constant temperature increase would realistically simulate the effects of global warming. In
fact, if soils dry under global warming, then the
temperature effect on soil of global warming should
vary diurnally (Harte et al. 1995b). Seasonal variation is also predicted by GCMs (Kattenberg and
Maskell 1995). Use of a constant level of downward IR allows investigation of the temporal variability in soil temperature response, as well as the
subsequent ecosystem responses.
At the midrange of GCM forecasts, the net surface warming due to both the direct effect of an
equivalent doubling of CO 2 plus the contribution
of major feedbacks is simulated by enhancing
downward IR radiation by 20 to 30 W m -2 (Harte
et al. 1995b). To achieve daily-averaged soil temperature increases as high as 6°C at 10 cm soil
depth, such as is sometimes attained in greenhouse
studies, fluxes of up to 60 to 80 W m - 2 would be
john and coworkers. Another serious criticism of
this technique is that it creates an unrealistic vertical temperature profile in the soil, affecting aboveground ecosystem components only indirectly and
deeper soil more than surface soil. This is opposite
to global warming, which involves increased transfer of heat to the soil from above. Buried wires do
not warm foliage and thus do not provide a wholeecosystem warming method. Despite the important
role of soil in ecosystem function, a soil-only
warming experiment is not a sufficient substitute
for a whole-ecosystem manipulation.
Warming Soil Directly: Fluid-Filled Pipes
at the Soil Surface
A method similar to buried electrical wires but with
fewer disadvantages is heating the soil surface with
warming pipes. Hillier et al. (1994) used a system
of fluid-filled, stainless steel pipes at the soil surface
to produce a continuous, constant temperature differential. They used this pipe system in a meadow
in southern England to manipulate the length of the
growing season and to simulate a milder winter.
Each warmed plot was paired with an unheated reference plot of similar vegetation structure and with
a dummy pipe network to cancel out shading effects
introduced by the apparatus. The pipe network was
supported by a metal grid, which also functioned
in the lateral dissipation of heat, creating a more
uniform temperature distribution.
There are several advantages to this type of system. Unlike buried electrical wires, this method
does not create an unrealistic vertical temperature
profile in the soil, nor disturb the soil surface or
low-lying vegetation. And unlike greenhouses,
there is not blocked precipitation, reduced wind, or
light attenuation. The apparatus can be used to produce heating or cooling, enabling a greater variety
of climate manipulation scenarios. For bare soil or
low-lying vegetation, this method is a good one for
manipulating foliage as well as soil temperature.
For elevating temperature in ecosystems with taller
plants, Hillier et al. (1994) mention the use of infrared heaters suspended over the vegetation, to
which we now tum.
Increasing Downward Radiation:
Overhead Heaters
In global warming, the total heating effect is the
sum of the direct infrared (IR) radiative effect of
increased greenhouse gas concentrations and the reKarin P. Shen and John Harte
suIting feedback effects involving the hydrologic
cycle, atmospheric circulation, and, in ways that are
less well understood, the biota, all operating from
local to global spatial scales. Although elevated
greenhouse gas levels result in surface warming because of both enhanced downward IR and because
warmer air convects and conducts more heat to ecosystems, the consequences of both the direct and
feedback effects on surface warming can be simulated by raising the downward IR radiation level
appropriately.
Manipulating radiation can be preferable to other
warming methods for many reasons. Increasing the
downward IR radiation mimics the "top down" nature of heating that actually occurs under an atmosphere with increased levels of greenhouse
gases. Nijs et al. (1996) found that applying heating
from above maintains the natural temperature gradients within the canopy. Infrared lamps transfer
heat noninvasively, and there are few unintended
effects on microclimate. Year-round manipUlations
are possible, although in areas with very deep winter snowpack, the heaters may become covered if
mounted too low, thus temporarily creating an unrealistic warming scenario. Greenhouses, buried
wires, and surface pipes all have greater problems
with application in winter, however, so comparatively, overhead heaters are a better option.
Overhead IR lamps can be thermostated to
achieve a constant increase in soil temperature, but
there is no reason to believe that forcing a continuous, constant temperature increase would realistically simulate the effects of global warming. In
fact, if soils dry under global warming, then the
temperature effect on soil of global warming should
vary diurnally (Harte et al. 1995b). Seasonal variation is also predicted by GCMs (Kattenberg and
Maskell 1995). Use of a constant level of downward IR allows investigation of the temporal variability in soil temperature response, as well as the
subsequent ecosystem responses.
At the midrange of GCM forecasts, the net surface warming due to both the direct effect of an
equivalent doubling of CO 2 plus the contribution
of major feedbacks is simulated by enhancing
downward IR radiation by 20 to 30 W m -2 (Harte
et al. 1995b). To achieve daily-averaged soil temperature increases as high as 6°C at 10 cm soil
depth, such as is sometimes attained in greenhouse
studies, fluxes of up to 60 to 80 W m - 2 would be