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

Подождите немного. Документ загружается.

Influence of irrigation 119

36 km

(a)

(b)

Figure 6.11 Composite of GOES-derived surface temperature at 1300 LST for

the period 1 August 1986 to 15 August 1986 for (a) northeast Colorado (FC,

Fort Collins; FM, Fort Morgan; GR, Greeley), and (b) the San Luis Valley in

Colorado (AL, Alamosa; AN, Antonito; DN, Del Norte; SA, Saguache). The

lower valley is outlined by a dark line separating it from significant elevated

terrain. Irrigated areas are shaded. Reproduced from Segal et al. (1988) with

permission from the American Meteorological Society.

120 Other land-use/land-cover changes

The first example (Fig. 6.11a) is for northeast Colorado which includes the agri-

cultural areas along the Front Range and the South Platte River. The topographical

variations in the area presented are in the range of 200 m.

The second example is for the San Luis Valley (Fig. 6.11b) where there is

intensive irrigated summer agricultural activity in the central domain. The valley is

nearly flat, and it is surrounded by steep mountains to the southwest and northeast

as indicated in the figure.

In both composites, there is a significant correspondence between the cooler areas

and vegetative cover. The highest temperature occur in the uncultivated areas. Maxi-

mum IR surface temperature gradients of 10



C and 12



C over distances of 10–20 km

were observed in northeast Colorado and the San Luis Valley, respectively. Since

the emittance of the Earth surface is less than 1, the actual surface temperatures

involved with these cases are somewhat higher than those presented. The emissivity

of the irrigated area, however, is likely to be somewhat higher than that of its dry-

land surroundings (e.g., Lee, 1978). Thus the actual surface temperature gradients

are suggested to be larger than those illustrated. Therefore noticeable circulations

should be expected to occur with such temperature gradients.

Figure 6.12 illustrates two soundings made over two locations in northeast

Colorado at 1213 LST on 28 July 1987 (Pielke and Zeng, 1989; Segal et al.,

250

300

400

500

700

850

–10 0 10 20

T (

P (mb)

3040

Figure 6.12 Radiosonde measurements of temperature (right side) and dewpoint

temperature (left side) for a dryland area (dashed line) and an irrigated area (solid

line) in northeast Colorado at 1213 LST on 28 July 1987. Reproduced from

Pielke and Zeng (1989) with permission from the National Weather Association.

Influence of irrigation 121

1989). The soundings were made prior to significant cloud development. The

radiosonde sounding over an irrigated location had a slightly cooler, but moister

lower troposphere than the sounding over the natural, shortgrass prairie location.

Aircraft flights at several levels between these two locations on 28 July 1987

(Figs. 6.13 and 6.14) demonstrate that the moistening and cooling occurred over

the entire region of irrigation. Using a convective index to 500 mb, the lifted

index, assuming surface parcel ascent, was −2 over the irrigated land but zero

over the shortgrass prairie (Pielke and Zeng, 1989). For this case, the moistening

of the lower atmosphere over the irrigated area was more important in increasing

CAPE than was the slight cooling in decreasing CAPE.

6.2.2 Nebraska

As discussed in Adegoke et al. (2003), over the last five decades, the total acreage

under irrigation in the US High Plains increased from less than 1.2 million

hectares to over 8 million hectares (Kuzelka, 1993). The rapid development of

320.0

316.0

317.0

θ (k)

320.0

317.0

01020304050

(d)

(c)

(b)

(a)

DISTANCE (km)

320.0

Figure 6.13 Measured potential temperature from Briggsdale to Windsor at the

altitude of (a) 140 m, (b) 240 m, (c) 345 m, and (d) 440 m above the ground.

The observed crop–dryland boundary is indicated by an arrow, with cropland to

its right. Adapted from Segal et al. (1989) with permission from the American

Meteorological Society.

122 Other land-use/land-cover changes

12.5

r (g kg

–1

)r (g kg

–1

)r (g kg

–1

)r (g kg

–1

)

13.0

8.0

12.0

7.0

13.0

8.0

0102030

DISTANCE (km)

(a)

(b)

(c)

(d)

7.5

Figure 6.14 Same as Fig. 6.13 except for moisture mixing ratio. Adapted from

Segal et al. (1989) with permission from the American Meteorological Society.

irrigation in this area, which stretches from Nebraska through western Kansas

to the Texas Panhandle, enabled the transformation of the area into one of the

major agricultural areas of the United States. In Nebraska, as in much of the

High Plains, corn is the dominant crop cultivated during the warm season months

(Williams and Murfield, 1977). Irrigated corn, which represented about 10% of

total corn-producing areas during the early 1950s, now composes nearly 60% of

the total corn-producing areas in Nebraska (Fig. 6.15). Figure 6.16 illustrates how

this irrigated landscape appears.

Data from the National Agricultural Statistics Services (NASS) of the United

States Department of Agriculture (USDA) for York County in east-central

Nebraska further underscore these changes. Between 1950 and 1998 the irrigated

corn area in York County increased from 3500 ha to 80 000 ha (a 2300% increase)

while the rain-fed corn area declined rapidly during the same period (National

Agricultural Statistics Service, 1998). This rapid land-use change was achieved

largely by converting rain-fed corn areas to irrigation. Land-use conversion of

this magnitude could affect energy and moisture exchanges between the land

surface and the lower atmosphere by altering transpiration and evaporation thus

generating complex changes in the surface energy budget.

Regional Atmospheric Modeling System (RAMS) (Pielke et al., 1992; Cotton

et al., 2003) simulations were performed which consisted of four land-use

Influence of irrigation 123

Figure 6.15 Area (%) of rainfed and irrigated corn farming in Nebraska (1950–

98). Reproduced from Adegoke et al. (2003) with permission from the American

Meteorological Society.

Figure 6.16 South of the South Platte River, south and west of North Platte,

Nebraska, looking north on 1 June 2004 at approximately 1300 LST. A number

of pivot irrigators are not watering. Photo courtesy of Kelly Redmond. See also

color plate.

scenarios covering the 15-day period from 1 to 15 July 1997. The first scenario

(control run) represented current farmland acreage under irrigation in Nebraska

as estimated from 1997 Landsat satellite and ancillary data (Fig. 6.17a). The

second and third scenarios (OGE wet and dry runs) represented the land-use con-

ditions from the Olson Global Ecosystem (OGE) vegetation dataset (Fig. 6.17b),

and the fourth scenario (natural vegetation run) represented the potential

124 Other land-use/land-cover changes

(c)

Shortgrass Tallgrass

Crop/mixed/farming Irrigated crop

Wooded

grassland

Mixed

woodland Scrubland

(a)

(b)

Figure 6.17 Land-cover datasets used for RAMS simulations for (a) 1997 Land-

sat and ancillary data irrigation, (b) OGE, and (c) Küchler potential vegetation.

From Adegoke et al. (2003) reproduced with permission from the American

Meteorological Society. See also color plate.

(i.e., pre-European settlement) land cover from the Küchler vegetation dataset

(Fig. 6.17c). In the control and OGE wet run simulations, the topsoil of the areas

under irrigation, up to a depth of 0.2 m, was saturated at 0000 UTC each day for

the duration of the experiment (1–15 July 1997). In both the OGE dry and natural

runs, the soil was allowed to dry out, except when replenished naturally by rainfall.

The “soil wetting” procedure for the control and OGE wet runs was constructed

to imitate the center-pivot irrigation scheduling under dry synoptic atmospheric

conditions such as observed in Nebraska during the first half of July 2000 (i.e.,

Influence of irrigation 125

0.9

0.8

(a)

(b)

(c)

0.7

0.6

0.5

0.4

0.3

0.2

0.1

–0.1

–0.2

–0.3

–0.4

–0.5

–5

–10

–15

–20

0.4

0.35

0.3

0.25

0.2

0.15

0.1

0.05

–0.05

–0.1

–0.15

–0.2

–0.25

–0.3

1JUL

1997

3JUL 5JUL 7JUL 9JUL 11JUL 13JUL 15JUL

1JUL

1997

3JUL 5JUL 7JUL 9JUL 11JUL 13JUL 15JUL

1JUL

1997

3JUL 5JUL 7JUL 9JUL 11JUL 13JUL 15JUL

Temperature (°C)

Latent heat (W m

–2

)

Moisture flux

–1

)

Figure 6.18 Control minus OGE wet run inner-domain area-averaged (a) 2 m

temperature, (b) surface latent heat, and (c) moisture flux at 500 m. Reproduced

from Adegoke et al. (2003) with permission from the American Meteorological

Society.

126 Other land-use/land-cover changes

Table 6.1 Inner-domain area-averaged model parameters for 7–15 July 1997

including scenario comparisons (% change)

Control OGE

(wet)

OGE

(dry)

Natural

vegetation

Temperature (



C) 241 24.9 (0.8%) 25.5 (1.4%) 27.6 (3.5%)

Surface sensible heat (W m

−2

)762 79.8 (4.7%) 86.9 (15%) 98.4 (29%)

Latent heat (W m

−2

) 1024 98.2 (4%) 74.5 (35%) 72.0 (42%)

Vapor flux at 500 m (g kg

−1

)111 10.4 (7%) 9.1 (22%) 8.2 (34%)

Source: From Adegoke et al. (2003).

little or no rainfall recorded throughout the state). The observed atmospheric con-

ditions from the National Centers for Environmental Prediction (NCEP) reanalysis

data (Kalnay et al., 1996) were used to create identical lateral boundary conditions

in the four cases. A two-grid nested model domain configuration was adopted with

a 10-km grid centered over Nebraska nested inside a larger 40-km grid centered

over Nebraska nested inside a larger 40-km grid, which extends over most of the

central United States.

Results for the inner-domain area-averaged model parameters between the control

run and OGE wet run (Fig. 6.18) showed very moderate differences (see Table 6.1

for a summary of scenario comparisons). This reflects the rather small change (less

than 10%) in the irrigated portion of the OGE vegetation data (Fig. 6.17b) compared

to the more recent Landsat satellite-based land-cover estimates (Fig. 6.17a). In

both simulations, the soil-wetting procedure was implemented for the irrigated

areas (i.e., land-use class 16 in LEAF-2 (Land-Ecosystem Atmospheric Feed-

back Model Version 2)). Larger changes were observed when the control run

was compared to the OGE dry run; midsummer 2 m temperature over Nebraska

might be cooler by as much as 34



C under current conditions (Fig. 6.19a).

The average difference between the control and OGE dry runs computed for the

6–15 July 2000 period was 12



C. The irrigation-induced surface cooling was

accompanied by a 36% increase in the surface in the surface latent heat flux

(Fig. 6.19b) and a significant increase (28%) in water vapor flux at 500 m above

the ground (Fig. 6.19c). A corresponding reduction in surface sensible heat (15%)

anda26



C elevation in dewpoint temperature were also observed (not shown).

The cooling effect and the surface energy budget differences identified above

intensified in magnitude when the control run results were compared to the poten-

tial natural vegetation scenario. For example, the near-ground average temperature

for 6–15 July 2000 was 33



C cooler, the surface latent heat flux was 42% higher,

and the water vapor flux (at 500 m) 38% greater in the control run compared

to the natural landscape run (Fig. 6.20). The first 5 days of the simulation were

Influence of irrigation 127

(a)

(b)

(c)

–1

–2

–3

–4

–50

–100

–150

100

150

200

3.5

2.5

1.5

0.5

–0.5

–1

–1.5

–2

–2.5

–3

1JUL

1997

3JUL 5JUL

7JUL 9JUL 11JUL 13JUL

15JUL

1JUL

1997

3JUL 5JUL 7JUL 9JUL 11JUL 13JUL 15JUL

1JUL

1997

3JUL 5JUL 7JUL 9JUL 11JUL 13JUL 15JUL

Temperature (°C)

Latent heat (W m

–2

)

Moisture flux (g

–1

)

Figure 6.19 Control minus OGE dry run inner-domain area-averaged (a) 2 m

temperature, (b) surface latent heat, and (c) moisture flux at 500 m. Reproduced

from Adegoke et al. (2003) with permission from the American Meteorological

Society.

128 Other land-use/land-cover changes

(a)

(b)

(c)

–2

–1

–3

–4

–6

–5

–7

Temperature (°C)

Latent heat (W

–2

)

Moisture flux (g

–1

)

–50

–100

100

150

200

250

–1

–2

–3

1JUL

1997

3JUL 5JUL

7JUL 9JUL 11JUL 13JUL

15JUL

1JUL

1997

3JUL 5JUL 7JUL 9JUL 11JUL 13JUL 15JUL

1JUL

1997

3JUL 5JUL 7JUL 9JUL 11JUL 13JUL 15JUL

Figure 6.20 Control minus natural vegetation run inner-domain area-averaged

(a) 2 m temperature, (b) surface latent heat, and (c) moisture flux at 500 m.

Reproduced from Adegoke et al. (2003) with permission from the American

Meteorological Society.