Cotton W.R., Pielke R.A. Human Impacts on Weather and Climate

Подождите немного. Документ загружается.

Global perspective 229

–0.5

–1

–0.5

0.5

0° 60° E

60°

30°

60°

0°

60°

W0°120° E 120° W180°

–0.5

–1

Figure 11.6 Convective precipitation differences (current minus natural, con-

tour by 05 mm day

−1

) using a nine-point spatial filter for easier visibility. Light

shading represents the 90% significance level for a one-sided t-test. Dark shad-

ing represents the 95% significance level. Reproduced from Chase et al. (2000)

with kind permission of Springer Science and Business Media.

0° 60° E 120° E 180° 120° W60° W0°

60° N

30° N

0°

30° S

° S

Figure 11.7 Height difference of 200 hPa pressure (current minus natural land-

scapes). Contour interval is 20 m. Reproduced from Chase et al. (2000) with

kind permission of Springer Science and Business Media.

230 Global effects of vegetation dynamics

0° 60° E

60° S

30°

60°

0°

60°

W0°120° E 120° W180°

0.5

0.5 0.5

0.5

–0.5

0.5

1.5

–0.5

–3

0.5

1.5

–0.5

–1.5

–1

–1.5

–0.5

–1.5

–0.5

–1

Figure 11.8 Difference in near-surface air temperature (current minus natural

landscape) using a nine-point spatial filter for easier visibility. Contour intervals

are 0.5, 1.0, 1.5, and 30



C. Light shading represents the 90% significance

level for a one-sided t-test. Dark shading represents the 95% significance level.

Reproduced from Chase et al. (2000) with kind permission of Springer Science

and Business Media.

concluded that interannual variations of soil wetness are large enough to influence

GCM climate simulations.

Deliberate land-use change (afforestation or reforestation) has been accepted as

a mechanism to remove carbon dioxide from the atmosphere and sequester carbon

in trees and soils. However, as discussed by Betts (2000) and Pielke (2001b), this

activity may have unintended consequences with respect to radiative forcing in

the atmosphere which again illustrates the complexity of the climate system. For

example, in regions subject to significant snow cover, afforestation would result

in a lower surface albedo and hence a positive radiative forcing, resulting in a

net warming effect despite the removal of carbon dioxide from the atmosphere

(Fig. 11.9). Similarly, increases in the surface fluxes of water vapor could result

in positive radiative forcing.

The biogeochemical effect of enhanced carbon dioxide and trace gas concen-

tration, and of aerosol deposition (such as nitrogen), on landscape dynamics has

also not been adequately considered. For example, Jenkinson et al. (1991) demon-

strated a significant positive feedback where soils released carbon to the atmo-

sphere under warming conditions. Lenton (2000), using a simple box model, and

Cox et al. (2000) using GCM sensitivity experiments, showed that biogeochemical

Global perspective 231

Table 11.2 A summary list of global land-use/land-cover assessments: effect on

weather and climate

Alpert et al. (2005), Avissar and Werth (2005), Baidya Roy and Avissar (2002), Betts

(1999, 2001),

Betts et al. (1997), Bonan et al. (1995), Boucher et al. (2004), Bounoua

et al. (2000, 2002),

Broström et al. (1998), Brovkin et al. (1999, 2002, 2004, 2005),

Brubaker and Entekhabi (1995), Burke et al. (2000), Chase et al. (1996, 2000, 2001),

Claussen (1995, 1998, 2001a,b), Claussen et al. (1998, 1999, 2001, 2002, 2004, 2005),

Collatz et al. (1991), Costa and Foley (2000), Couzin (1999), Cox et al. (2000), DeFries

et al. (2002), deNoblet-Ducoudre et al. (2000), Dirmeyer (1994, 1998), Dirmeyer and

Shukla (1996), Eltahir (1996), Entekhabi and Brubaker (1995), Entekhabi et al. (1992),

Feddema et al. (2005a,b), Ferranti and Molteni (1999), Foley et al. (1994, 1998, 2003,

2005), Fraedrich et al. (1999), Franchito and Rao (1992),

Friedlingstein et al. (2001),

Ganopolski et al. (1998), Gedney and Valdes (2000), Govindasamy et al. (2001), Guillevic

et al. (2002), Hoffmann and Jackson (2000), Idso et al. (1975), Keith et al. (2004),

Kleidon et al. (2000), Koster and Suarez (2004),

Koster et al. (2000, 2004),

Kubatzki

and Claussen (1998), Lamptey et al. (2005), Levis et al. (1999, 2000), Liu et al. (2006),

Lofgren (1995a,b), Loveland et al. (2000), Lyons et al. (1996), Matthews et al. (2004),

McGuffie et al. (1995), Milly and Dunne (1994), Myhre and Myhre (2003), Pielke Sr.

(2005), Pielke et al. (2002), Pitman and Zhao (2000), Pitman et al. (1999), Polcher and

Laval (1994a,b), Porporato et al. (2000), Ramankutty and Foley (1999), Ramankutty

et al. (2006), Ramírez and Senareth, (2000), Rodriguez-Iturbe et al. (1991a,b), Ruddiman

(2003), Snyder et al. (2004b), Sud et al. (1993, 1995, 1996), Texier et al. (2000), Turner

et al. (1990), van den Hurk et al. (2003),

Vitousek et al. (1997), Werth and Avissar

(2002, 2005), Williams (2003), Xue (1996, 1997), Xue and Fennessy (2002), Xue and

Shukla (1993, 1996), Xue et al. (1996), Zeng and Neelin (1999), Zeng et al. (1999, 2002),

Zhang et al. (1996a,b, 2001), Zhao and Pitman (2002a,b), Zhao et al. (2001a,b)

Information was provided by Dr. Rick Raddatz.

feedbacks in conjunction with an increased carbon dioxide radiative warming

produced an amplified regional and global surface temperature response. As dis-

cussed in Chapter 6, Eastman et al. (2001a,b) used a regional climate model

in a sensitivity study and suggested a cooler daytime and warmer nighttime in

the central Great Plains in response to greater plant growth in a doubled carbon

dioxide atmosphere.

The presence of drought and hydrological feedbacks associated with land-use

change locally or through teleconnections, therefore, has a direct impact on the

source–sink capabilities of the terrestrial ecosystem and therefore on climate.

An important conclusion from these studies is that land-use change directly

alters local and regional heat and moisture fluxes in two ways. Firstly, the local

and regional CAPE is changed since the Bowen ratio is changed as the surface

heat and moisture budgets are altered. Secondly, larger-scale heat and moisture

convergence, and associated large-scale wind circulations can be changed as a

232 Global effects of vegetation dynamics

result of changes in the large-scale atmospheric pressure field due to the landscape

change.

As a general conclusion, these regional and global model studies indicate that

the spatial patterning of land-use/land-cover change result in changes over time

in regional tropospheric diabatic heating patterns. These changes in the patterns

result in alterations in global circulation patterns. Indeed, it appears that such

heterogeneous diabatic heating changes on the regional scale have more of an

influence on the global climate scale than the more spatially homogeneous climate

forcing of the radiative effect of increasing CO

11.4 Quantifying land-use/land-cover forcing of climate

Carbon has become the currency used to assess the human intervention in the

Earth’s climate system. Impacts on climate are compared in terms of radiative

forcing, which can be considered as perturbations to the Earth’s radiation budget

prior to feedbacks from the rest of the climate system. The concept of global

warming potential (GWP) where

GWP =



xtdt



rtdt

 (11.1)

has been adopted to convert other atmospheric constituents into their equivalent

in terms of carbon dioxide atmospheric radiative forcing (Houghton et al., 2001).

Here, TH is the time period over which the calculation is considered, a

is the

radiative efficiency due to a unit change in atmospheric abundance of the substance

x (i.e., W m

−2

−1

), and xt is the time-dependent decay of the abundance from

an instantaneous release of the substance. The denominator is the same expression

but for the reference substance r, defined to be carbon dioxide.

The effects of land-surface albedo change can be quantified in terms of radiative

forcing (Hansen et al., 1997a; Betts, 2001), and this has been used in attempts

to compare the global significance of historical land-use change with that of

other drivers of climate change (Houghton et al., 2001). Betts (2000) suggested

that radiative forcing calculations could be used to translate albedo changes into

equivalent carbon emissions (Fig. 11.9); this could be useful for quantifying land-

use changes in regions where the main impact is on surface albedo, such as areas

subject to significant snow cover. However, in other regions, changes in other

land-surface properties may not exert a radiative forcing but still significantly

influence climate. For example, the partitioning of available energy into latent

and sensible heat fluxes exerts a direct impact on near-surface air temperature,

so a change in this partitioning should be considered a climate forcing. Radiative

forcing, and hence GWP, are therefore not able to represent the full impact

Quantifying land-use/land-cover forcing of climate 233

(a) Radiative forcing due to carbon sequestration (nW m

–2

–1

)

(b) Radiative forcing due to albedo change (nW m

–2

–1

)

–1

)

(d) Net radiative forcing by afforestation (nW m

–2

–1

)

–0.8

–0.6 –0.4 –0.2 –0.1 –0.05 0.05 0.1 0.15 0.2

–0.6 –0.5 –0.4 –0.3 –0.2

0.05

30 40 50 60 70 80 90 100 120

0.1 0.15 0.2 0.25 0.3 0.35 0.4

Figure 11.9 Radiative forcing of climate by afforestation, considering illus-

trative 1-ha plantations in the temperate and boreal forest zones. Calculations

apply to the time at the end of one forestry rotation period, relative to the

start of the rotation period with plantation areas unforested. (a) Global mean

longwave radiative forcing due to carbon dioxide removal through sequestra-

tion nWm

−2

−1

. (b) Global mean shortwave radiative forcing due to albedo

reduction nWm

−2

−1

. (c) Carbon emissions that would give the same mag-

nitude of radiative forcing as the albedo reduction tC ha

−1

. (d) Net radiative

forcing due to afforestation, found by summing (a) and (b) nWm

−2

−1

.

Positive forcing implies a warming influence; where (d) shows positive values,

afforestation would warm climate rather than cooling it as would be expected

by considering carbon sequestration alone. After Betts (2000); from Pielke et al.

(2002). See also color plate.

234 Global effects of vegetation dynamics

of land-cover change in all regions. Some new means of quantifying land-use

forcing is therefore required. Separation of the components of the surface energy

budget could provide a possible starting point, with the surface heat energy being

separated into

+ Q

= 0 (11.2)

where

= Q

1 − A + Q

↓

↑

 (11.3)

Here, Q

is the net radiative flux, Q

is the turbulent sensible heat flux, Q

the turbulent latent heat flux (evaporation/transpiration), Q

is the heat flux into

the Earth’s surface, Q

is the solar irradiance, A is the surface albedo, Q

↓

is the

downward atmospheric irradiance, and Q

↑

is the upward surface irradiance. The

magnitude of these fluxes can be expressed in units of watts per meter squared

or joules per unit of time (for example, a globally averaged value of 1 W m

−2

is equal to 161 10

J per decade; 1 W = 1J s

−1

). One measure of land-cover

change forcing of climate could be the perturbation to one of the components of

the surface energy balance equation (Eq. 11.2) prior to feedbacks from the rest of

the climate system.

In the past, climate change metrics have been concerned with globally aver-

aged responses as exemplified in the GWP. However, as discussed previously in

this chapter, global-scale climate changes can also occur due to regional land-use

changes but where global averaged values are unchanged. This occurred in the

situation described by Chase et al. (1996, 2000) where there was no significant

global averaged change in these quantities. Figure 11.10 illustrates the anthro-

pogenically caused change in LAI (see Nemani et al., 1996), which were used

in the Chase et al. studies. It was the spatial redistribution of the land surface

sensible and latent heat flux pattern that resulted in the global climate change.

Globally averaged climate change may therefore bear no well-defined relation

to the real changes experienced in any region and these regional changes, which

can be of any sign, are what impact people and will stimulate mitigation strategies

to be applied (Matsui and Pielke, 2006). Therefore, we identify a surface “regional

climate change potential” (RCCP) that addresses this deficiency. The RCCP would

be defined to quantify where a direct human-caused change (either positive or

negative) alters any of the individual terms in Eq. (11.2). These changes could be

scaled by the surface area of the Earth to place them in a global context.

To provide a land-use forcing term free of feedbacks, as in the case of radiative

forcing, a land-surface scheme could be used to calculate perturbations to the sur-

face energy budget excluding feedbacks from the atmosphere. However, assessing

the true global significance of these feedbacks is not trivial, since atmospheric

circulation changes could give rise to remote climate changes that do not relate

Quantifying land-use/land-cover forcing of climate 235

scale: 0.5 × 0.5 /lat/lon

–2

LAI

+2 –2 –40

Figure 11.10 Effect of land-use changes on plant canopy density (potential LAI

− actual LAI). Scale 0.5 latitude × 0.5 longitude. From Pielke et al. (2002),

based on Nemani et al. (1996). See also color plate.

linearly to the perturbation in the region of land-use change (e.g., Gedney and

Valdes, 2000). To illustrate this point, the model results from Chase et al. (2000)

are used to show the 10-year average absolute-value differences in the surface

energy flux for January and July (i.e., Q

and Q

from Eq. 11.2) between the

climate with the current landscape and the climate with potential landscape that

should exist under current atmospheric conditions without human intervention.

Figure 11.11 presents the 10-year average value difference in the surface turbu-

lent heat fluxes between the two experiments for just the locations where the

human-caused landscape change was prescribed in the model, while Fig. 11.12

presents the 10-year average difference values in the surface turbulent heat fluxes

worldwide.

The 10-year average change for the locations where only the land cover was

altered were 53Wm

−2

and 47Wm

−2

in January, and 88Wm

−2

and 67Wm

−2

in July for surface sensible heat flux and latent heat flux, respectively. The

globally weighted changes in the sum of the absolute values of the surface

sensible and latent heat flux changes for Fig. 11.11 were 07Wm

−2

in January

and 108 W m

−2

in July. With teleconnections changes also included, however,

the globally averaged changes in the surface fluxes were significantly larger

with values of 947 W m

−2

in January and 890 W m

−2

in July (Fig. 11.12).

These results clearly demonstrate that regional landscape change can result in an

amplified perturbation to surface fluxes elsewhere in the world through non-linear

feedbacks within the atmosphere’s global circulation. If the other terms in the

236 Global effects of vegetation dynamics

Figure 11.11 The 10-year average absolute value change in (a) January surface

latent turbulent heat flux, (b) July surface latent turbulent heat flux, (c) January

surface sensible heat flux, and (d) July surface sensible heat flux in W m

−2

the locations where land-use change occurred. Based on Chase et al. (2000);

from Pielke et al. (2002). See also color plate.

surface heat budget (Eq. 11.2) were included, the magnitude of the differences

(i.e., the RCCP) would presumably be even larger.

A potential based purely on surface flux perturbations would not be analo-

gous to GWP, because the latter is an index requiring an expression in terms

of the abundance of atmospheric constituents. Direct comparison of land-cover

change effects with greenhouse gas emissions therefore remains a challenge. One

possibility might be to dispense with quantification in terms of forcings, and com-

pare anthropogenic influences by modeling whole-system responses to individual

causes of climate change.

For example, Claussen et al. (2001) compared biogeophysical and biogeochem-

ical effects of land-cover change in terms of the overall climate change (as opposed

to the forcing). A number of simulations were performed with an intermediate

complexity Earth system model, in each run altering a small part of the land

surface and simulating its effects on global mean temperature (through changes in

60° N

(a)

60° E 120° E

4 8 12 16 20 24

180° 120° W60° W

30°

60° S

0°

0° 0°

60° N

(b)

60° E 120° E

4 8 12 16 20 24

180° 120° W60° W

30°

60°

0°

0° 0°

60° N

(c)

60° E 120° E

4 8 12 16 20 24

180° 120° W60° W

30°

30° N

60°

0°

0° 0°

60° N

(d)

60° E 120° E

4 8 12 16 20 24

180° 120° W60° W

30° S

30°

60°

0°

0° 0°

Atmosphere–vegetation interactions 237

60° N

(a)

60° E 120° E

4 8 12 16 20 24

180° 120° W60° W

30°

60°

0°

0° 0°

60° N

(b)

60° E 120° E

4 8 12 16 20 24

180° 120° W60° W

30°

60°

0°

0° 0°

60° N

(c)

60° E 120° E

4 8 12 16 20 24

180° 120° W60° W

30°

60°

0°

0° 0°

60° N

(d)

60° E 120° E

4 8 12 16 20 24

180° 120° W60° W

30°

60°

0°

0° 0°

Figure 11.12 The 10-year average absolute value change in surface sensible

and latent turbulent heat flux in W m

−2

worldwide as a result of the land-use

changes. (a) January surface latent turbulent heat flux, (b) July surface latent

turbulent heat flux, (c) January sensible turbulent heat flux, and (d) July sensible

turbulent heat flux. Based on Chase et al. (2000); from Pielke et al. (2002). See

also color plate.

surface properties and carbon exchange). However, as stated above, global mean

temperature changes conceal important local and regional changes, so applications

of this type of modeling approach again require a global quantification of regional

changes.

11.5 Atmosphere–vegetation interactions

The atmosphere and biosphere are coupled, dynamic systems. On the long time-

scales, the major vegetation biomes of the Earth propagate in response to climate

change. During the Pleistocene, for example, spruce and fir forests covered what is

now northern Florida, as a vast continental ice sheet covered much of northeastern

North America. In the eastern plains of Colorado, large sand dunes developed

about 3000 to 1500 years before the present (BP) at a time of greater aridity

than exists today (Muhs, 1985). Human agricultural systems are also affected.

238 Global effects of vegetation dynamics

Neumann and Sigrist (1978) conclude that in Babylonia about 3800 to 3650 years

BP the barley harvest was 10–20 days earlier than it is now due to a warmer

climate, while cooler weather from the years 2600 to 2400 years BP resulted in

a harvest 10–20 days later than under the current climate in this region.

On shorter timescales, Schwartz and Karl (1990) report that the seasonal upward

progression of temperature in the eastern United States is temporarily halted

as vegetation greens in the spring. At least a 35



C reduction in surface daily

maximum temperature occurred over the subsequent 2-week period after the

beginning of leafing in agricultural inland areas of the central and eastern United

States, as contrasted with the 2 weeks prior to first leaf. This occurred even when

the overlying atmosphere had the same lower tropospheric average temperature.

On a hemispheric scale, seasonal pulsations of carbon dioxide concentrations

occur as vegetation extracts carbon dioxide in spring and summer in the vast land

areas of the Northern Hemisphere.

On daily timescales, vegetation interacts with the atmosphere through its direct

influence on the partitioning between latent and sensible heat fluxes, as discussed

in Chapter 6. During the day, over regions of transpiring vegetation, huge fluxes of

moisture to the atmosphere and a significant reduced surface heating effect occurs.

At night the stomata on the leaves close and respiration of oxygen dominates over

photosynthesis and the uptake of carbon dioxide by plants.

The importance of biological interactions with the atmosphere and as a compo-

nent of the climate system is only now becoming recognized. The ability of veg-

etation to effectively utilize increased carbon dioxide to increase global biomass

may result in reduced carbon dioxide warming from what otherwise would occur.

We need to determine, however, the impact of deforestation and conversion of

vegetated land to asphalt, roads, industrial centers, etc. on this potential storage

mechanism for carbon. Also, how are methane and nitrogen oxide inputs to the

atmosphere from industrial and agricultural practices influenced by the presence

of plants? Does enhanced nitrogen, for example, permit enriched fertilization for

plant growth?

Several non-radiative forcings involve the biological components of the climate

system. They can be categorized into three types.

(1) Biophysical forcing involves changes in the fluxes of trace gases and heat between

vegetation, soils, and the atmosphere. For example, a GCM experiment by Sellers

et al. (1997) showed that, in the presence of increased carbon dioxide, plants open

their stomata less and are therefore more water efficient. Thus, increased carbon

dioxide impacts the hydrological cycle, in addition to its well-known direct radiative

impacts.

(2) Biogeochemical forcing involves changes in vegetation biomass and soils. For exam-

ple, increased nitrogen deposition caused by greater anthropogenic emissions of