B-l I26
October 1972

 
  

 - Winter Wheat and Grain Sorghum

 g Systems
Northern High Plains of Texas

¥= Y!
A;

The Texas Agricultural Experiment Station
j. E. Miller, Director, College Station, Texas
A Texas A8cM University

Summary
Introduction 3 f
Purpose of Publication and v
Objectives of Study 3
Location of the Study 3 .4
Climate 3
Topography and Soils 5i
Description of the Study  5-ij
Results and Discussion t 7 i
Wheat 7 f
Grain Sorghum 11 a
Fallow l4;
Soil Properties 16‘
Organic Matter 165i
Bulk Density 17
Dry Aggregates _ 17'?
Wet Aggregates l8
Economics of the Different Cropping Systems .... ..l9
Acknowledgment 20 f‘
Literature Cited 20
Summary

A dryland winter wheat (Triticum aestivum L.)
and grain sorghum (Sorghum bicolor L.) cropping
system study was conducted at the USDA Southwestern
Great Plains Research Center, Bushland, from 1957 to
1970. Data for grain yields, soil water storage and use,
precipitation, fallow efficiency and soil properties are
presented.

Wheat grain yields were lowest for the continu-
ous wheat (CW), intermediate for wheat in the wheat-
sorghum-fallow (WSF) and highest for the wheat-
fallow (WF) cropping system on a harvested-area basis.
The yield differences seemingly resulted from differ-
ences in available soil water at seeding. Water-use
efficiency paralleled grain yields when soil water
changes and growing season precipitation were con-
sidered, but the trends were reversed when precipi-
tation during the fallow period was included. Includ-
ing fallow precipitation in total water-use efficiency
points out the low effectiveness of fallow for influ-
encing crop yields.

On a harvested-area basis, sorghum in the WSF
system significantly outyielded sorghum grown con-
tinuously (CS). The increased yields were related to
differences in available soil water at seeding. Trends
in water-use efficiency, whether based on . growing
season or fallow plus growing season precipitation,
were similar to the trends for wheat. However, grain

2

Contents

m ..\ .

   
  
  
   
   
 
  
  
  
   
   
   
  
   
   
  

sorghum used water more efficiently than did wheat
(based on pounds of grain per acre produced per acre-
inch of water used).

Storage efficiency decreased as the length of the
fallow period increased. Storage efficiences ranged
from 8.3 percent for the WF system to 20.1 percent for
the CS system during the 1959-70 period.

The soil organic matter content and the distri-
bution of dry and wet soil aggregates measured at
the end of the study were significantly affected by.’
the cropping systems. The organic matter of the
surface 6 inches of soil was highest (2.04 percent) for
a grass treatment, but significant differences also re-
sulted from the different wheat and sorghum cropping
systems, ranging from 1.64 percent for CS to 1.861
percent for CW. The CS system resulted in the high?
est percentage of fine (less than 0.84 millimeters), d =-
soil aggregates, indicating greater potential soil erodi-
bility by wind than for other systems. The distribu-
tion of wet soil aggregates was related to time since
harvest of the previous crop. Plant residues favored
the formation of larger aggregates, but the effective-A
ness of the residues decreased as the length of the
fallow period increased (time from harvest to sampling
for aggregate size distribution). Soil bulk density was
not affected by the cropping systems.

    
   
   
  
  
   
  
  
  
   
   
  
   
  
 
 
 
  
   
 
 
  

 PECT TO ACREAGE PLANTED, grain sorghum
 most important crop in Texas, and it is
iin order by cotton and wheat (Yearbook
 Committee, 1970). Of the state total,

i were planted on the Northern High Plains
" This area is bounded by the state line on
orth and east and by the southern boundary
l , Castro, Swisher, Briscoe, Donley and Col-
it- counties on the south. In this area, about
Q of the grain sorghum and 55 percent of
Q were grown on dryland in 1970 (New, 1970).

 ugh irrigated acreage has increased some-
f nt years, water tables and well yields are
f and projections are that much of the irri-
 e will revert to dryland crop production
erground water supply is eventually de-
1f hes and Harman, 1969). When this occurs,
 p production will increase in importance.
 bility of eventual water importation into
has not been overlooked. However, it is
' at water importation would not reduce the

op production acreage to below current

fi

of Publicatipn and Objectives of Study

a land winter." wheat and grain sorghum
 tem study conducted from 1957 to 1970
' bjectives:

_ determine the effects of selected cropping
tems on wheat and sorghum grain yields

 determine the effects of the cropping

, gacres of grain sorghum and 2,031,000 acres '

i; [and Winter Wheat and Grain Sorghum Cropping Systems - -
if Northern High Plains of Texas

Paul W. Unger*

systems on residue production for erosion
control

3. To determine the interrelationships of soil
water at seeding and growing season precipi-
tation in their effect on grain and residue
yields

4. To determine the efficiency of water storage
during the fallow period

5. To compare the effects of reseeded native
grass and the various cropping systems on
various physical and chemical soil properties.

Location of the Study

The study was conducted at the USDA South-
western Great Plains Research Center, Bushland. The
Center, located about l4 miles west of Amarillo, lies
in Potter and Randall counties. Its location is near
the center of the Northern High Plains of Texas. Soils
at the Center are representative of much of the land
used for wheat and grain sorghum production in the
area (Taylor et al., 1963).

Climate

Precipitation and temperature are the major
climatic factors influencing crop production in the
area. Precipitation averages from I6 inches at the
western edge to 24 inches at the eastern edge of the

‘Soil scientist, USDA Southwestern Great Plains Research Center,
Bushland.

Mention of a trademark or a proprietary product does not
constitute a guarantee or warranty of the product by the U.S.
Department of Agriculture or The Texas Agricultural Experi-
ment Station and does not imply its approval to the exclusion
of other products that may also be suitable. i

3

INCHES

YEARS

Figure l. Long-term precipitation at Amarillo, Texas, plotted as a l2-month moving total to show above- and below-average periods, i892-
1969. Long-term average is 20.92 inches. Points on the curve represent totals for the past l2 months (unpublished data from J. T. Musick).

area, but the precipitation at a given location is highly
variable (Bonnen, 1960). For example, yearly rainfall
at Bushland has ranged from 9.46 inches in 1970 to
32.87 inches in 1941 for the 1939 to 1970 period. The
precipitation variability from 1892 to 1969 at Amarillo
near the center of the Northern High Plains area is
shown in Figure 1.1

‘J. T. Musick, unpublished data.

IO '
I PRECIPITATION
U EVAPORATION
8 |
(D
Figure 2. Monthly rainfall u’ 6 '
and evaporation at the
USDA Southwestern Great I , T
P I a i n s Research Center, Q
Bushland, Texas. The pre-
cipitation shown is the av- Z 4 '

erage for a 32-year [1939-
70) period. Evaporation
shown for April through
September is a 31-year
(1940-70) average. For 2

the remaining months,
evaporation shown is an ,
IB-year (1951-68) CtV€r-
age.
0

  
   
 
 
 
 
 
 
 
  

I970

Precipitation at Bushland from 1939 to 1970 has,
averaged 18.26 inches per year. Precipitation has
averaged 8.66 and 9.94 inches during the grain sor-
ghum (June 16 to October l0) and wheat (October 11
to June l5) growing seasons, respectively. The aver-
age monthly distribution of precipitation and the
average pan evaporation [Young screen pan (Blood-
good, Patterson and Smith, 1954)] are shown in‘
Figure 2.

JAN. FEB. MAR. APR. MAY JUNE auur AUG. SEPT. ocr. nov. DEC.‘

 

MONTH

  

    
   
    
  
   
  
    
  
  
   
  
   
    
  
 
 
 
  
 
   
  
  
 

ure has its major effect on crop pro-
gh its influence on the length of the
_‘od, which averages from about 180 to
figthe northwest and southeast portions of
  tively (Bonnen, 1960). At Bushland,
  averages 190 days. Although not
 - rtance for winter wheat, low tempera-
influence the length of the growing
grain sorghum. The first killing frost
average, October 28, but occurred as
r 7 and as late as November 22 during
period. The last killing frost occurs, on
April l8,“ but has occurred as early as
nd as late as May l4. Average monthly
d minimum temperatures at Bushland
m Figure 3.

Topography and Soils

of the Northern High Plains of Texas is
ill feet above sea level and slopes toward
‘theast. ‘The land is nearly flat, but num-
9—n0rmally dry—dot the area. Much of
i drainage is into these lakes, but some
canyons that extend into the area and
gthe headwaters of the Brazos, Red and
ivers (Bonnen, 1960). The Canadian River
‘iv iated “breaks” of the river divide the
- major subareas. The soils are primarily
ay loams, but some sandy soils are in-
e principal soil at the Center is Pullman
aylor et al., 1963). The Pullman series
of the fine, mixed, thermic family of
leustalls (order Mollisols).

the Northern High Plains are subject to
ind, especially under dryland conditions.
ce the drouth of the 1930's, controlling
by maintaining crop residues on the soil

O
Q
O

00
O

 9e monthly
f; minimum
Kat thePUSDA
 Great ‘Plains
 er, Bushland,
‘libs 32 - year

TEMPERATURE
N b
O O

.)-

l

surface by stubble-mulch tillage has been studied.
Wheat yields with stubble-mulch tillage were equal
to or higher than yields with moldboard and one-way
tillage under dryland conditions in a previous study
at the Research Center (Johnson, 1950).

DESCRIPTION OF THE STUDY

The Pullman clay loam of the study area had
a slope of less than 1 percent. Plot size was 0.21 acre
(60 by 150 feet). The plots were bordered on three
sides, with natural runoff being permitted from the
fourth side. The following treatments, each replicated
three times, were randomly assigned to the treatment
blocks:

1. Continuous wheat (CW)——wheat seeded on the
same plots each year.

2. Wheat-fallow (WF)-wheat seeded on the
same plots in alternate years. This cropping
system provided for a fallow period of about
l5 months between harvest and seeding on a
particular plot. Two plots were used for this
treatment in each replication.

3. Wheat-sorghum-fallow (WSF)—altemate crops
of wheat and sorghum seeded on the plots,
resulting in two crops being grown during
the 3-year rotation. This system provided for
a fallow period of ll months between sor-
ghum harvest and wheat seeding or between
wheat harvest and sorghum seeding. Three
plots were used for this treatment in each
replication.

4. Wheat-sorghum-fallow with permanent ridges
and furrows on 40-inch spacings (WSF-RF)-
this cropping system was similar to the WSF
system above, except that the tillage methods
were different. This treatment was designed

 

n n A l l 1 n A

o n 4

 

JAN. FEB. MAR. APR. MAY JUNE JULY AUG. SEPT. OOT. NOV. DEG.

MONTH

for the ridges to function as increased runoff
zones and the furrows, because of deep chisel-
ing, to function as increased water intake
zones. Through increased water concentra-
tion and depth of water penetration in the
furrows, evaporation losses should be de-
creased and, thus, storage efficiency increased.
The system was first used for the 1966-67
crop. Plots for this cropping system were
those originally designated as wheat-optional
wheat, wheat-optional sorghum and sorghum-
optional wheat. The optional seedings were
based on 1.5 inches of water being available
for plant growth in the upper 2 feet of the
soil at seeding time. By 1966, the options
had not been exercised because the available
soil water at seeding time always exceeded
this amount. Data from the wheat-optional
wheat and wheat-optional sorghum plots were
combined with data from the continuous
wheat plots, and data from the sorghum-
optional wheat plots were combined with data
from the continuous sorghum plots for the
1957-66 period.

5. Continuous sorghum (CS)—sorghum seeded
on the same plots each year.

6. Grass-seeded to native grasses and used as
reference plots to determine the effects of
grass on physical and chemical properties of
soil.

All tillage before seeding wheat and sorghum,
except on the WSF-RF plots, was performed with
stubble-mulch equipment. This equipment had 30-
to 40-inch sweeps, and the tillage was limited to about
a 5-inch depth. On the WSF-RF plots, the furrows

 were chiseled after crop harvest, provided the soil

was dry, to a l-foot depth with a vibrating chisel.
Initial tillage when the soil was wet and subsequent
tillage after a chiseling operation were performed
with a sweep-row weeder with buffers attached to
the sweeps to maintain the ridges and furrows. The
24-inch sweeps tilled the furrows and lower sides of
the ridges, while the rod tilled the upper portion of
the ridges. In some years, a rolling cultivator was
used on the WSF-RF plots to control small weeds and
volunteer wheat.

Concho variety wheat was seeded in 1957 and
1958. Thereafter, Tascosa variety wheat was used.
Seeding rate was about 30 pounds per acre. Seeding
date depended on soil water availability for germi-
nation and ranged from September 10 in 1964 to
November 19 in 1957. On the CW, WF and WSF
plots, seeding was performed with a hoe-type (Noble)
drill which resulted in a 14-inch spacing between drill
rows. A single-disk, 10-inch grain drill was used to
seed wheat on the WSF-RF plots. Two rows were
seeded on the ridge and two in the furrows of the
40-inch spaced ridge-furrows.

6

   
 
    
 
   
 
 
 
 
 
 
 
 
 
 
 
  
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
  
 
 
 
   
  

Grain sorghum hybrids used were RS-610 from
1958 to 1964 and RS-626, a similar hybrid with head I
smut resistance, from 1965 to 1970. The seeding rate a
was about 2 pounds per acre. Seeding dates ranged
from June 4 in 1969 to July 9 in 1962. Lister planters ,
or unit planters mounted behind listers were used to
seed sorghum in furrows spaced 40 inches apart. ‘

A. mixture of buffalo (Buchloe dactyloides), blue
grama (Bouteloua gracilis) and sideoats grama (Boute- i
loua curtipendula) grasses was seeded on the grass.
plots April 3, 1958. Favorable soil water conditions *
at seeding resulted in a good stand of the grasses. p‘
Forage produced on the grass plots was not measured
or utilized. In some years, the plots were mowed or
sprayed with 2,4-D to control weeds.

In the interval between crops, weeds were con-
trolled with stubble-mulch or sweep-rod weeder tillage. 5
For weed control in sorghum, the initial cultivation
was with a knife sled on all plots. The second culti-l
vation was with a knife sled on the WSF-RF plots
and with a sweep cultivator on the other plots. Culti-
vations largely leveled ridges and furrows on the lister-
seeded plots. In some years, weeds were controlledl
by spraying with 2,4-D. Occasionally, hand hoeing
was used to control sparse weed populations. Grow-
ing-season weeds in wheat were controlled with 2,4-D
in years when heavy infestations occurred.

No fertilizers were used on either the wheat or
the grain sorghum during the study. Dryland wheat ;
has not responded to nitrogen or phosphorus ferti-
lization on Pullman clay loam (Eck and Fanning,
1962), and nutrient deficiencies have not been ob-i
served on dryland grain sorghum.

The soil water content at seeding and harvest.
was determined gravimetrically from cores obtainedi
by l-foot increments to a 6-foot depth at three 1oca- 5
tions per plot. Although soil water was measured to
a 6-foot depth, Musick and Sletten (1966) indicated -
that sorghum yields were more closely related to water
content and change to a 4-foot soil depth of Pullman
soil. Therefore, the soil water data presented, other‘
than in Figure 4, are based on the 4-foot depth.
Precipitation was measured at the plot area. i

In most years, grain yields were determined by I
combine harvesting a swath through the entire length
of each plot. In a few cases, yields were determined
by hand harvesting measured areas in the plots.

Residue yields were not measured in enough years '
to warrant inclusion and analysis of the data. How»,
ever, the residues produced, along with stubble-mulch‘
tillage, generally effectively controlled wind erosion
on the plot area. 1

In November 1970 soil cores were obtained at '
six locations in each plot to a 4-foot depth. The
cores were divided into 0- to 6-, 6- to 12-, 12- to 18-,
18- to 24-, 24- to 36- and 36- to 48-inch increments.‘

“composited by depth increments and
'ing soil bulk density and organic
_ g The organic matter content was
l" the Walkley-Black procedure (Piper,
nic, matter content was also deter-
w ce- samples taken at the initiation

   
  
   
   
 
  
  
  
   
  
  
   
   
  
  
  
   
   
   
    
   
  
    
  
   
   
  

p samples were obtained in November
and dry aggregate determinations. For
I f: size distribution determinations, a
ist soil was obtained from the 1- to
at three locations in each plot. The
-- through a 15-inch sieve while moist,
before the distribution of water-stable
a nest of sieves was determined accord-

 - ure outlined by Kemper and Chepil
I procedure was slightly modified by the
 and 4.00-millimeter (mm) sieves rather
- 4.76-mm sieves along with 1.00- and

ples for aggregate size distribution by
_‘ ere obtained from the surface 1 inch of
 locations in each plot. These samples
if- before the size distribution of the dry
 5 determined with a rotary sieve having
 The different sieves had 0.42-, 0.84-, 2.0-,
,1 square openings.

ULTS AND DISCUSSION

Wheat

water, precipitation, water storage and
a are given in Table 1 by individual
 wheat cropping systems. Average values
l‘, - s during which the cropping systems can
_ ‘  are given in Table 2. Winter wheat was

Q fall and harvested in late spring or early
fthe following year. A crop year is the
 the wheat was harvested. Although
VJ}- was harvested in 1958, pregrowing season
 data (precipitation from harvest of last
_' g of the current crop) were not available
U Data for all crops are presented, but
Tffiom the 1959-70 and the 1967-70 periods
, f ed. The wheat-sorghum-fallow rotation
 and furrows (WSF-RF) was used during
.; period only.

 grain yields were significantly affected
""_-- systems. Based on the area harvested,
 " acre was lowest for continuous wheat
 u diate for wheat-sorghum-fallow (WSF)
"J for wheat-fallow (WF) for the 1959-70
 averaged somewhat higher for the
 F systems during the 1967-70 period than
j 1959-70 period. For the 1967-70 period,
_ system increased the average yield by
 per acre over the WSF system, but the
 not statistically significant. Also, the
 ce (83 pounds per acre) between the

WSF-RF and WF systems was not significant. The
WSF-RF system was tested in only 4 crop years, a
period possibly too short to make a valid test of this
cropping system.

Available soil water content at seeding was lowest
for the CW system, averaging 2.29 inches to a 4-foot
depth for the 1959-70 period. The WSF and WF
systems resulted in 1.08 and 1.38 inches greater water
contents at seeding, respectively, than the CW system.
Soil water contents at seeding during the 1967-70
period averaged slightly lower than during the 1959-
70 period. The soil water content at seeding for the
WSF-RF system was similar to that of the WSF system.
A possible reason that the WSF-RF-system did not
increase water storage as anticipated was inadequate
weed control. Herbicides were not used during the
nongrowing period, and the sweep-rod weeder and
rolling cultivator were not effective tillage implements
for operation on the undisturbed permanent ridges.

Soil water changes between seeding and harvest
averaged 0.93, 1.94 and 2.11 inches for the CW, WSF
and WF systems, respectively, for the 1959-70 period.

The greater changes due to the WSF and WF systems 

were approximately equal to the additional water
stored at seeding due to these systems as compared
with the CW system. Changes during the 1967-70
period were somewhat greater than during the 1959-
70 period. Again, the greater changes due to the
WSF, WF and WSF-RF systems as compared with
the CW system were similar to the differences in water
content at seeding. Changes in soil water for the
WSF-RF system were similar to those of the WSF
system.

Assuming that the yield differences were associ-
ated only with differences in the soil water content
at seeding, each inch of additional soil water at seeding
resulted in 142 and 188 pounds more grain per acre
for the WSF and WF systems, respectively, during the
1959-70 period than for the CW system. During the
1967-70 period, each additional inch of stored water
resulted in 182, 270 and 345 pounds more grain per
acre for the WSF, WSF-RF and WF systems, respec-
tively. The increases during the 1959-70 period com-
pare favorably with those predicted by Johnson (1964).
The greater-than-predicted increases during the 1967-
70 period possibly were due to better distribution of
growing season precipitation. Another possibility may
have been better growing-season weed control. Weeds,
primarily Tansy mustard [Descurainia pinnata (Walt.)
Britt.], were controlled chemically in early spring
during the 1967-70 period.

Soil water contents at seeding and changes be-
tween seeding and harvest indicate that appreciable
amounts of “available” water were present in the soil
at harvest in some years (Figure 4). Much of the
available water present was due to late season rain-
fall, which benefited the crop only slightly. Rainfall
from June 1, June l1 and June 21 to harvest averaged

7

4.35, 2.66 and 1.04 inches, respectively, during the
study period. The average harvest date was June 28.

Growing season precipitation was identical for
all systems, and total growing season water use by
the crops was considered equal to growing season

precipitation plus soil water changes. Water use
values include some water lost by surface runoff in
some seasons. Using this total, grain production
averaged 54, 60 and 69 pounds per acre-inch of water
used during the 1959-70 period for the CW, WSF

TABLE 1. WHEAT YIELD, AVAILABLE SOIL WATER,‘ PRECIPITATION, WATER STORAGE AND WATER USE DATA BYTdNDlVlDUAL YEARS FOR
CONTINUOUS WHEAT, WHEAT-SORGHUM-FALLOW AND WHEAT-FALLOW CROPPING SYSTEMS ON DRYLAND

Water-use
Total water used efficiency
Pre-GS Pre-GS
es and es es and es
Avcniubie so“ precip- precip- precip- precip-
wme, itation itation itation itation
_ Precipitation Water storage *1‘ $011 + soil + soil + soil
cfOppiflg Grain At water water water water
system Year yield seedIng Change’ GS“ Pre-GS‘ Amount Erficiency” change change change change
Lb./acre
Lb. /acre Inches Inches Inches Percent Inches Inch
CW 1958 885 3.93 6——5.1 3 8.17 13.30 67
1959 862 2.82 -1.98 9.14 9.88 4.02 40.7 11.12 21.00 78 41
1960 1440 1.05 + 2.01 17.03 5.44 0.21 3.9 15.02 20.46 96 70
1961 663 5.20 —3.44 11.49 15.12 2.14 14.2 14.93 30.05 44 22
T962 98 1.36 +1.56 13.65 5.64 -0.40 — 7.1 12.09 17.73 8 6
1963 481 2.82 —-0.15 10.10 8.29 —0.1O — 1.2 10.25 18.54 47 26
1964 676 2.89 —-2.30 6.99 7.68 0.22 2.9 9.29 16.97 73 40
1965 214 0.70 +1.43 19.77 2.76 0.11 4.0 18.34 21.10 12 10
1966 412 2.95 '——3.48 5.53 4.75 0.82 17.3 9.01 13.76 46 30
1967 478 1.07 --0.72 12.38 6.43 1.60 24.9 13.10 19.53 36 24
1968 802 0.52 '-—0.60 8.79 2.83 0.17 6.0 9.39 12.22 85 66
1969 486 1.43 +0.22 11.84 7.13 1.51 21.0 11.62 18.75 42 26
1970 672 4.70 —3.80 4.38 14.13 3.05 21.6 8.18 22.31 82 30
WSF 1958 793 3.87 '—5.41 8.17 13.58 58
1959 1108 3.16 '—3.96 9.14 19.49 13.10 32.59 85 34
1960 1326 2.34 0.00 17.03 13.98 2.69 19.2 17.03 31.01 78 43
1961 676 5.38 —3.4O 11.49 31.03 4.75 15.3 14.89 45.92 45 15
1962 504 4.08 —0.68 13.65 10.88 1.39 12.8 14.33 25.21 35 20
1963 458 3.55 —0.69 10.10 20.81 2.90 13.9 10.79 31.60 42 14
1964 705 2.94 -—2.39 6.99 17.08 2.78 16.3 9.38 26.46 _75 27
1965 261 2.58 +3.15 19.77 8.67 —0.67 — 7.7 16.62 25.29 16 10
1966 668 3.47 '—4.06 5.53 20.48 1.53 7.5 9.59 30.07 70 22
1967 590 1.98 —1.82 12.38 10.03 0.29 2.9 14.20 24.23 42 24
1968 , 756 2.06 -1.90 8.79 15.05 2.67 17.7 10.69 25.74 71 29
1969 1508 3.98 —-3.06 11.84 16.22 4.12 25.4 14.90 31.12 101 48
1970 540 4.96 ——4.53 4.38 24.75 3.91 15.8 8.91 33.66 61 16
WSF-RF 1967 623 2.84 ——2.76 12.38 10.03 0.29 2.9 15.14 25.17 41 25
1968 1244 2.41 —2.33 8.79 15.05 2.80 18.6 11.12 26.17 112 48
1969 1234 3.11 —2.11 11.84 16.22 3.49 21.5 13.95 30.17 88 41
1970 690 4.37 —3.94 4.38 24.75 3.16 12.8 8.32 33.07 83 21
WF 1958 859 3.28 '—4.47 8.17 12.64 68
1959 1116 3.80 --3.07 9.14 26.77 12.21 38.98 91 29
1960 1589 3.5T —0.70 17.03 24.46 4.70 19.2 17.73 42.19 90 38
1961 715 4.78 —3.09 11.49 37.59 4.05 10.8 14.58 52.17 49 14
1962 409 4.86 —2.07 13.65 32.24 2.05 6.4 15.72 47.96 26 9
1963 573 3.45 ——1.15 10.10 27.58 1.76 6.4 11.25 38.83 51 15
1964 873 4.10 ——2.96 6.99 26.07 1.31 5 O 9.95 36.02 88 24
1965 260 2.87 +2.19 19.77 17.43 0.57 3.3 17.58 35.01 15 7
1966 744 4.15 '——4.88 5.53 27.28 3.01 11.0 10.4.1 37.69 71 20
1967 1029 1.31 +0.25 12.38 16.71 —3.75 —22.4 12.13 28.84 85 36
1968 956 1.89 ——1.70 8.79 24.89 2.62 10.5 10.49 35.38 91 27
1969 1466 4.00 -—-2.81 11.84 18.74 2.44 13.0 14.65 33.39 100 44
1970 672 5.41 —5.36 4.38 33.08 5.22 15.8 9.74 42.82 69 16

‘Determined to a 4-foot depth and based on wilting point values determined by the sunflower technique. Unavailable water to 4 feet totals,
9.97 inches. 1
‘Based on soil water changes between crop seeding and harvest.

‘Growing season.

‘Pregrowing season (from harvest of previous crop to seeding of current crop).

‘Based on soil water changes and precipitation occurring during the fallow period that preceded seeding of the indicated crop.

‘Changes in available soil water content exceeding the available soil water content at seeding apparently resulted from the inadequacy of the
sunflower technique for establishing a precise wilting point for wheat and from soil drying to below the wilting point due to evaporation.

   
 
   
   
   
     
    
    
  
   
 
  
  
  
   

iWHEAT YIELD, AVAILABLE SOIL WATER,‘ PRECIPITATION, WATER STORAGE AND WATER USE DATA BY PERIODS FOR
T, WHEAT-SORGHUM-FALLOW AND WHEAT-FALLOW CROPPING SYSTEMS ON DRYLAND

Water-use
Total water used efficiency
Pre-GS Pre-GS
GS and GS GS and GS
Available soil 9'99")‘ PWCIP‘ PTQCIP‘ PTWIP‘
water itation itation itation itation
Precipitation Water storage ‘l’ $9" + soil + soil + soil
Grain At water water water water
yield seeding Change’ GS” Pre-GS‘ Amount Efficiency“ change change change change
Lb./ acre Inches Inches Inches Percent Inches Lb. I acre - inch
'62s? 2.41 -1.2e 10.71 11.97 s5
75a" 3.41 -2.21 10.71 12.92 so
866' 3.64 —2.29 10.71 ’ 13.00 69
'607' 2.29 -0.93 10.92 7.50 1.03 14.8 11.85 19.35 54 33
761' 3.37 —1.94 10.92 17.38 2.40 13.9 12.86 30.24 60 25
867' 3.67 —-2.11 10.92 26.07 2.18 8.3 13.03 39.10 69 22
'610' 1.93 —-1.33 9.34 7.63 1.58 20.7 10.67 18.30 62 37
849' 3.24 —2.82 9.34 16.52 2.75 16.6 12.16 28.68 69 29
F 948'" 3.18 -—2.78 9.34 16.52 2.44 14.7 12.12 28.64 81 34
1031' 3.15 —2.40 9.34 23.36 1.63 7.0 11.74 35.10 86 31

4-foot depth and based on wilting point values determined by the sunflower technique. Unavailable water to 4 feet totals
 changes between crop seeding and harvest.
 (from harvest of previous crop to seeding of current crop). "

stored divided by total precipitation received during all fallow periods.
In a group followed by the same letter or letters are not significantly different (Duncan's Multiple Range Test--5-percent

SOIL WATER VCONTENT-“INOHES/FOOT
2.o 2.5 3.0 5.5 _ 2.0 2.5 5.0 5.5

I I I ‘ T l *'I7

I ‘U

2~° 2-5 3-0 3-5 WATER content T0 s Fr.

(AGTUAL- wm
' svsram HARV. seen.
. wsF-wn. 2.1a 4.65
cw 1.14 5.22
' 1:41- 255 5.51
wsr- _ cs wsF-soa. 2.21 4.51
$°R6HUM 0s 2.2a see

 soil water contents at the beginning (harvest) and end (seeding) of the fallow periods for the dryland wheat and grain sor-
V ng systems; also wilting point values, based on the sunflower technique and field observations. The actual minus wilting point
31in estimate of the plant available water in the soil at harvest of the previous crop and seeding of the indicated crop.

m 140'
l.l.l
'2
I20 '
a SORGHUM-‘GS
E
t_ too-
< i
\.
{I ao -
6 I
....l
| so - WHEAT-GS
3 ' SORGHUM-TOTAL
g 4o -
Z v  AT—TOTAL
< 2o - i‘
O!
w INCREASING FALLOW LENGTH
o I I l
0W WSF WF
0S
OROPPING SYSTEM
Figure 5. Water efficiency for grain production as influenced by

wheat and grain sorghum in the different dryland cropping systems
(lengths of fallow periods). The values are based on growing sea-
son soil water change and precipitation (GS) or growing season soil
water change and total precipitation between crop harvests (total).

and WF systems, respectively (Figure 5). Water-use
efficiencies were 62, 69, 81 and 86 pounds of grain
per acre-inch of water during the 1967-70 period for
the CW, WSF, WSF-RF and WF systems, respectively.
The greater water-use efficiency during the 1967-70
period again possibly was due to better growing season
precipitation distribution and weed control during
this period than during the 1959-70 period. Reason
for the greater efficiency for the WSF-RF system as

  
   
 
 
   
 
  
 
 
 
 
 
 
 
 
 
 
   
 
  
 
 
 
 
 
 
 
 
 
 
 
 
 

compared with the WSF system and the greater in- _.
crease in efficiency for the WF system as compared i
with the other systems for the 1967-70 period over d
the 1959-70 period is not readily apparent. Values I
for the 1959-70 period should be more reliable be-
cause the 1967-70 period may have been too short to >
make valid tests of the systems, especially under the
highly variable climatic conditions prevalent in the
region. 

When fallow precipitation, that received from
harvest of the previous crop to seeding of the current
crop, was included with water used by the current.
crop, total water-use values averaged 19.35, 30.24 and -
39.10 inches per crop during the 1959-70 period for g
the CW, WSF and WF systems, respectively. Corre- ,
sponding total water-use efficiency values were 33, 25 I
and 22 pounds per acre-inch, respectively (Figure 5). l
The lower total water-use efficiency values obtained .
by including fallow precipitation indicate the low
effectiveness of fallow precipitation for increasing crop .
yields. Trends during the 1967-70 period generally
were similar to trends during the 1959-70 period, but y
the values were slightly higher. The values were 37,
29, 34 and 31 pounds per acre-inch for the CW, WSF, g
WSF-RF and WF systems, respectively. '

Multiple linear regression analysis (Ezekiel and
Fox, 1959) was used to establish relationships between A
available soil water at seeding, growing season precipi- a
tation and wheat grain yields. For this analysis, thef
growing season was divided into six growth periods. _
The periods were germination, seedling establishment i
and fall growth—seeding to December 31; winter main- ‘I
tenance and early spring growth-January 1 to April v
15; rapid spring growth to boot-April 16 to April 30; -'
late boot and flowering-May 1 to May 20; grain-
filling-May 21 to June l0; and grain hardening to f
harvest-June 11 to harvest. The net regression co-
efficients, along with their standard errors, and the .
coefficients of multiple determination (R2 values) for
the CW, WSF and WF systems are given in Table 3. f

 

TABLE 3. SUMMARY OF MULTIPLE LINEAR REGRESSION ANALYSIS ASSOCIATING WHEAT GRAIN YIELDS IN THE C\N, WSF AND WF SYSTEMS
ON DRYLAND WITH AVAILABLE SOIL WATER AT SEEDING‘ AND PRECIPITATION DURING DIFFERENT PERIODS OF THE GROWING SEASON:
Coefficient
Net regression coefficients’ 0f mulllple
Cropping determination '
system bu‘ b1 b: I): b4 b6 be b1 (R2) I
CW 161 12 51 98 184 52 76 -— 76 0.797‘
(Standard error) i117 i 92 i111 i 271 i106 i108 i 69
WSF 1092 - 199 ‘14<2* 202 —1112** 161* 25 -122* o.a42*
(Standard error) i110 i 68 i 82 i 180 i 71 i126 i 48
F 681 —— 85 81 173 —— 533 116 67 — 98 0.745
(Standard error) i 147 i 104 i 109 i 276 i 103 i 130 i 71

‘Determined to a 4-foot depth.

to harvest (b1).

‘The b coefficients indicate pounds of groin per acre per acre-inch of water.

'Y--inte rcept.

‘Asterisks denote statistical significance (one—-5~percent level; two--1-percent level).

l0

Unavailable water to 4 feet totals 9.97 inches.
‘The growing season periods and associated regression coefficients were germination, seedling establishment and fall growth—seeding to De?
cember 31 (b2); winter maintenance and early spring growth—-Januory 1 to April 15 lbs); rapid spring growth to boot—-April 16 to April j
30 (b4); late boot and flowering-—May 1 to May 20 (b5); grain filling—-May 21 to June 10 (bit); and grain hardening to harvest—June 11 a


 

The b1 regression coefficient is associated with this factor.

  
   
  
   
 
   
  
   
  
   
  
  
 
 
 
   
  
 
 
   
   
  
  
   
 
    
   
 
  
 
 
 
  
  
   
   
    
  
  

.. net regression coefficients were ob-
‘the WSF system. The coefficients were
" seeding to December 31, January 1
May l to May 15 periods and negative
‘to April 30 and the June ll to harvest
n for the highly significant negative
ipitation on grain yields during the
'1 30 period (rapid spring growth to
pparent but may be associated with
'ty later in the growing season. For
.- uate precipitation occurs during the
pril 30 period to permit the develop-
plants, “normal” or “below normal”

l?- later in the growing season may be
 mature the grain properly. The nega-
 of precipitation during the grain hard-
 est period was associated with plant
H; possibly, hail damage as the crop ap-
 ' i urity. Although not significant, coeffi-
-?= te season precipitation indicated yield
the CW and WF cropping systems also.

-- ted finding from the analyses was the
‘cant positive influences of available soil
If“ on wheat grain yields. Actually,
it not significant, coefficients were ob-
 WSF and WF systems. According to
2, available soil water at seeding and
est for the CW, intennediate for the
"Jy est for the WF system, but the influence
 at seeding on yields was not reflected in
 ion coefficients. Random variation
v L» a factor, but other workers (Eck and
‘  ) also have experienced difficulty in re-
 yields to soil water and precipitation

 although the data in Tables 1 and 2

“'1 the greater yields for the WSF and WF
'  related to greater available soil water at
 these systems than for the CW system,
i’ during the growing season possibly ob-
g uences of available soil water at seeding.

l‘ icients of multiple determination (R2
ff significant for the CW and WSF systems,
ghat significant amounts of the variations
if accounted for by available soil water at
,- by growing season precipitation. The
 the WF system (0.745) approached the
 for statistical significance (0.754).

Grain Sorghum
il water, precipitation, water storage and
ta for the grain sorghum cropping systems
, Table 4 for individual years. Average
igivenlfin Table 5 for periods in which the
pping systems can be compared. Although
[*1 systems were not fully in sequence in
~ for that year are included. Pregrowing
fpitation (that received between harvest
‘ ifus crop and seeding of the current crop)
{p crop was determined by establishing

“normal” harvest dates for the previous crops (July 1
for wheat and November 1 for grain sorghum).

For the 1958-70 period, grain sorghum yields for
the VJSF treatment exceeded yields for the CS treat-
ment by 4-20 pounds per acre, and the difference was
statistically significant. Based on the area harvested,
the yields averaged 1,137 and 1,557 pounds per acre
for the CS and WSF systems, respectively. The differ-
ences are attributed to greater soil water contents at
seeding and changes during the growing season for the
WSF system as compared with the CS system since
precipitation for both systems was identical. The
greater water change for the WVSF system was approx-
imately equal to the greater water content for this
system at seeding.

Average grain yields for the 1967-70 period when
the WSF-RF system was in effect were 970, 1,135 and
965 pounds per acre for the CS, WSF and WSF-RF
systems, respectively. The lower yield for the WSF-
RF system as compared with the WSF system possibly
resulted from the lower soil water content and change
for the WSF-RF system. The smaller water change
resulted from a lower water content at seeding with
the WSF-RF system, but reasons for the lower water
content are not clear. Inadequate weed control on
the permanent ridge-furrows as mentioned in the
wheat section, may have been a factor.

For comparable systems (CS and WSF), yields
were lower during the 1967-70 period than. during
the 1958-70 period (Table 5). Water content at seed-
ing and growing season change were lower during the
1967-70 period as was growing-season precipitation.
Consequently, yields also were lower. However, data
for the 1958-70 period should be more reliable when
comparing cropping systems because of the greater
number of years involved.

Based on the differences in available soil water at
seeding between the CS and WSF systems, each inch
of additional stored water resulted in a 609-pound-
per-acre increase in grain yields for the WSF system
during the 1958-70 period and a 358-pound-per-acre
increase during the 1967-70 period. Water contents
at seeding for the CS and WSF-RF systems were sim-
ilar during the 1967-70 period, and yields also were
similar for the two systems.

The yield increases due to additional stored water
at seeding were in the range reported by Bond, Army
and Lehman (1964). It is doubtful, however, that the
yield increases were due to the amount of additional
stored water per se. Of possibly greater importance
was the distribution of water in the profile and the
depth of wetting of the profile (Figure 4). For cotton,
Fisher and Burnett (1953) reported marked increases
in lint yield with increases in water in the second and
third foot of soil. Similar results can be expected for
grain sorghum because deep profile water would be
available to the plants in latter growth stages when
water is important during heading and grain filling.

ll

 
 

TABLE 4. GRAIN SORGHUM YIELD, AVAILABLE SOIL WATER,‘ PRECIPITATlON, WATER STORAGE AND WATER USE DATA BY INDIVIDUAL
YEARS FOR CONTINUOUS SORGHUM AND WHEAT-SORGHUM-FALLOW CROPPING SYSTEMS ON DRYLAND

er-‘Based on soil water changes between crop seeding and harvest.

 
 
 
 
  
  
  
 
 
  
 
  
 
  
 
 
 
 
  
     
 
 
  

Water-use
Total water used efficiency
Pre-GS Pre-GS p
GS and GS GS and GS
Available soil presip- presip- Presip- Prede-
wafer itation itqtion itation itation‘
. . . -|— soil .—l-', soil + soil + soi
Cropping Grain At Precmmcfmn Water slorqqe water wlater water water
system Year yield seeding Change’ GS’ Pre-GS‘ Amount Efficiency“ change change change change
Lb. lacre Inches Inches Inches Percent Inches Lb./acre - inch
CS 1958 1948 4.38 °——4.76 10.80 8.93 15.56 24.49 125 80w

1959 674 1.72 -1.46 8.60 6.50 2.18 33.5 10.06 16.56 67 412'

1960 1675 3.37 +0.10 19.81 15.57 3.11 20.0 19.71 35.28 85 47

1961 2060 4.56 —3.51 8.81 5.10 1.09 21.4 12.32 17.42 167 118

1962 715 2.49 —2.46 9.37 12.65 1.44 11.4 11.83 24.48 60 29

1963 1599 3.66 ——0.86 8.80 8.85 3.23 36.5 9.66 18.51 166 86

1964 500 2.47 -—-0.85 8.70 4.01 -—0.33 —- 8.2 9.55 13.56 52 37

1965 1376 3.70 -2.29 6.76 15.65 2.08 13.3 9.05 24.70 152 56

1966 360 1.98 '——2.56 6.72 3.47 0.57 16.4 9.28 12.75 A 39 28

1967 384 1.52 °—-1.67 7.19 7.56 2.10 27.8 8.86 16.42 43 23

1968 1350 2.48 —0.93 8.72 8.72 2.63 30.2 9.65 18.37 140 73

1969 959 1.95 +0.57 13.76 7.77 0.40 5.1 13.19 20.96 73 46

1970 1187 4.86 —4.1 6 3.21 7.60 2.34 30.8 7.37 14.97 161 79

WSF 1958 2026 4.07 “—4.42 10.80 17.04 15.22 32.26 133 63

1959 1964 3.13 —2.5O 8.60 16.67 4.67 28.0 11.10 27.77 177 71

1960 1990 2.77 —0.08 19.81 22.61 3.57 15.8 19.89 42.50 100 47

1961 2540 5.06 —4.41 8.81 24.57 2.72 11.1 13.22 37.79 192 67

1962 1325 3.49 —3.33 9.37 19.12 1.51 7.9 12.70 31.82 104 42

1963 1726 4.63 —-1.38 8.80 18.35 1.23 6.7 10.18 28.53 170 60

1964 1050 3.39 —1.45 8.70 12.77 0.53 4.2 10.15 22.92 103 46

1965 1545 5.05 —-3.36 6.76 22.45 4.50 20.0 10.12 32.57 153 47

1966 1542 3.96 °—4.57 6.72 10.15 —-1.77 -—17.4 11.29 21.44 137 72

1967 449 1.67 '—1.81 7.19 14.15 2.26 16.0 9.00 23.15 50 19

1968 1682 2.81 —1.76 8.72 11.25 2.65 23.6 10.48 21.73 160 77

1969 785 3.23 —0.99 13.76 16.22) 3.07 18.9 14.75 30.97 53 25

1970 1622 4.93 —4.61 3.21 18.51 4.01 21.7 7.82 26.33 207 62

WSF-RF 1967 235 1.02 -—1.40 7.19 14.15 1.57 11.1 8.59 22.74 27 10
1968 1586 2.03 —O.82 8.72 11.25 1.95 17 3 9.54 20.79 166 76

1969 802 2.30 —0.07 13.76 16.22 2.22 13.7 13.83 30.05 58 27

1970 1238 4.93 —4.65 3.21 18.51 3.93 21.2 7.86 26.37 158 47

‘Determined to a 4-foot depth and based on wilting point values determined by the sunflower technique. Unavailable water to 4 feet:
totals 9.97 inches.

‘Growing season.
‘Pregrowing season (from harvest of previous crop to seeding of current crop).
‘Based on soil water changes and precipitation occurring during the fallow period that preceded seeding of the indicated crop.

‘Changes in available soil water content exceeding the available soil water content at seeding apparently resulted from the inadequacy of the-
sunflower technique for establishing a precise wilting point for grain sorghum and from soil drying to below the wilting point due to.
evaporation.

The influence of depth of moist soil on yields has or factors for influencing yields. One of these could
been illustrated by Fisher and Burnett (1953) for be greater water intake during the growing season;
cotton and by Brown and Shrader (1959) for grain (See section on Soil Properties.) The coefficients for
sorghum. ' the regression equations and the R2 values were no V)
greatly different when the available soil water at seed-
ing was considered to a 6~foot depth, indicating a
relatively minor influence of water in soil at a 5- o .
6-foot depth on sorghum yields. i

Another possible factor involved in the differences
in yields between the CS and WSF systems was water
intake during the growing season. Regression equa-
tions based on available soil water at seeding (X) and a
yields (Y) were Y = — 104.4 + 412.5X (R2 = 0.646) for Total growing-season water use by the crops w _
continuous sorghum and Y = 449.9 + 298.7X (R2 considered equal to growing-season precipitation pl ~

 
 

= 0.305) for sorghum in the WSF system. Yields for soil water change between seeding and harvest. Usin
the CS system were more dependent on stored soil this total, grain production averaged 102 and l3
water at seeding than for the WSF system. The lesser pounds per acre-inch of water used during the 1958-7
dependency on stored soil water for the WSF system period for the CS and WSF systems, respective] y.

points toward a greater influence of some other factor (Figure 5). During the 1967-70 period, the efficien ~

12

  
   
  
  
  
  

~ IN SORGHUM YIELD, AVAILABLE sort WATER,‘ PRECIPITATION, WATER STORAGE AND WATER USE DATA BY PE-
sORGI-IUM AND WHEAT-SORGHUM-FALLOW CROPPWG sYsTEMs ON DRYLAND

Water-use
Total water used efficiency
Pre-GS Pre-GS
GS and GS GS and GS
Available soil Pmflp‘ Pmilp‘ Preclp‘ Praflp‘
water nation itation itation itation
_ Precipitation Water storaae + 5°51 "l- SOll + soil + soil
Grain A: water water water water
iyleld seeding Change’ GS“ Pre~GS‘ Amount Effidency‘ change change change change

Lb./acre - inch
.1137‘ 3.01 -—1.91 9.32 8.65 1.73 20.1 11.23 19.88 102 57

Lb. I acre Inches lnches Inches Percent Inches

 
 

1557" 3.70 —2.66 9.32 17.23 2.41 14.0 11.98 29.21 134 54
' 970‘ 2.70 -—-l.54 8.22 7.91 1.86 23.6 9.76 17.67 104 56
1135" 3.16 —2.29 8.22 15.03 2.99 19.9 10.51 25.54 118 46

965' 2.57 —- 1 .18 8.22 15.03 2.41 16.1 9.40 24.43 102 40

 
  
  
  
   
   
    
   
  
  
 
 
  
  
   
  
 
 
   
  
   
  
   
   
  
    
  

changes between crop seeding and harvest.

118 and 102 for the CS, WSF and
A respectively. The greater efficiency
SF system than for the CS system
soil water content at seeding and
eased yields for the WSF system.
» e greater water content was deeper
itgure 4). Brown and Shrader (1959)
-- water-use efficiencies with greater
. Possibly also involved with the
.. for the WSF system may have been
-season water intake as mentioned

efficiency based on pounds of grain
acre per acre-inch of water used was
um than for wheat. Also, the increase

sorghum in the WSF system over
um was greater than the increase in
heat in the WSF system over continu-
. - 5). This difference suggests that
more responsive to fallow than wheat
jgrain production. Similar conclusions
Luebs (1962). Timeliness of precipi-
pect to the growing season may have
_ ble for grain sorghum than for wheat.
 the water stored in soil at winter wheat
Hm used or evaporated during the long
t period, and spring vegetative growth
 pendent upon precipitation. For grain
“- staged water at seeding along with pre-
more readily available for vegetative
'n production.

ow precipitation was included with
the current crop, total water-use values
'gher (19.88 vs. 11.23 inches and 29.21
for the CS and WSF systems, respec-

depth and based on wilting point values determined by the sunflower technique. Unavailable water to 4 feet

mm harvest of previous crop to seeding of current crop).
~ - divided by total precipitation received during all fallow periods.
(b group followed by the same letter are not significantly different (Duncan's Multiple Range Test——5-percent level).

tively) and water-use efficiency values were much
lower (57 vs. 102 pounds per acre-inch and 54 vs. 134
pounds per acre-inch for the CS and WSF systems,
respectively) than where growing season precipitation
alone was used. The low efficiency values for grain
sorghum production when including fallow precipi-
tat-ion further substantiate the low effectiveness of
fallow for increasing crop production. However,
efficiency values were higher for grain sorghum than
for wheat when fallow precipitation was included,
again suggesting that grain sorghum was more re-
sponsive than wheat to fallow precipitation with
respect to grain production (Figure 5).

Multiple linear regression analysis was used to
establish relationships between available soil water at
seeding, precipitation during the growing season and
grain yields for the sorghum. The growing season
was divided into five periods which corresponded to
major plant growth stages. These were germination
and seedling establishment—seeding to 20 days after
seeding; rapid vegetative growth-21 to 50 days after
seeding; late boot and flowering—51 to 65 days after
seeding; grain filling—66 to 80 days after seeding; and
grain hardening to harvest—81 days after seeding to
harvest. The net regression coefficients and coeffi-
cients of multiple determination (R2 values) for
sorghum in the CS and WSF systems are given in
Table 6. Standard errors were calculated for the
individual net regression coefficients.

The coefficients (Table 6) indicate that grain
yields for the CS and WSF systems were influenced
most by the available soil water content at seeding.
Although the coefficient (b1) for available water at
seeding was higher for the WSF system than for the
CS system, the standard error associated with this

13

coefficient also was higher for the WSF system. Based
on the coefficient and the standard error, an inch of
available water at seeding would cause a range in
grain yields from 383 to 431 pounds per acre for the
CS system and a range from 368 to 530 pounds per
acre for the WSF system in about two of three cases.

For the CS system, yields were influenced most
by precipitation during the period of rapid vegetative
growth, as indicated by the regression coefficient being
greater than the standard error for the coefficient.
Undoubtedly, yields were also influenced by precipi-
tation during other periods. However, the high
variability of precipitation during the study period,
which is typical for the study area, resulted in the
high standard errors. The regression coefficients
exceeded the standard errors for the germination and
seedling establishment (b2) and grain filling (b5) pre-
cipitation periods for the WSF system, indicating a
positive influence of precipitation on yields during
these periods.

The coefficients of multiple determination (R2
values) were 0.893 and 0.656 for the CS and WSF
systems, respectively. The R2 values suggest a higher
correlation between yields, soil water and precipitation
for the CS system than for the WSF system and a
greater influence of some other factors on yields for
the WSF system than for the CS system. Soil fertility
may have been a factor. Although dryland grain
sorghum on Pullman clay loam has shown no nutrient
deficiencies, it is possible that grain sorghum would
respond to fertilizer in years of above-average precip-
itation.

Fallow

Fallowing (the practice of allowing land to remain
idle and weed-free for a growing season) has been
widely used to increase yield levels. As an illustration,
it was arbitrarily assumed that a crop producing less
than 600 pounds of grain per acre harvested would
not be profitable. Data for this 13-year study show
that wheat in the CW, WSF and WF systems produced
less than 600 pounds of grain per acre in 6, 5 and 3
years, respectively. For grain sorghum, yields were

TABLE 6. SUMMARY OF MULTIPLE LINEAR REGRESSION ANALYSIS ASSOCIATING GRAIN YIELDS OF CONTINUOUS SORGHUM AND SORGHUMl
IN A WSF SYSTEM ON DRYLAND WITH AVAILABLE SOIL WATER AT SEEDING‘ AND PRECIPITATION DURING DIFFERENT PERIODS OF THE GROW-i

less than 600 pounds per acre in 3 years for continu- l"
ous sorghum and in l year for sorghum in the WSF -
system. Thus, fallow did increase the reliability of ,
grain production by wheat and sorghum during the

study period.

Fallowing increased the yields of wheat and grain ,
sorghum but decreased the efficiency of total water‘
use for grain production. This inefficiency of fallow I
is widely recognized. For the study area, precipitation 
storage has generally been around l5 percent of the
precipitation received during the fallow period for i
Precipitation, water,‘
storage and water storage efficiency values during the
fallow period preceding wheat are included in Tables
l and 2 and preceding grain sorghum in Tables 4.
and 5. The distributions of water at the beginning;
(harvest of previous crop) and end (seeding of crop) j

winter wheat (Johnson, 1966).

of the fallow periods are shown in Figure 4.

For wheat, the fallow periods (interval between
crops for CW) were about 3, ll and 15 months for
the CW, WSF and WF systems, respectively. For
grain sorghum, the fallow periods (interval between
crops for CS) were about 8 and ll months for the

CS and WSF systems, respectively.

Precipitation amounts were directly related to}
length of the fallow period, and, in general, water
storage increased as length of the fallow period in-»_
creased. A marked exception was evident for the fr
WF cropping system. Although the fallow period
for WF was 4 months longer than for WSF and pre-%
cipitation averaged about 9 inches more for WF than
for WSF, the WF system resulted in slightly less water"
storage and considerably lower storage efficiency than‘
the WSF system. Possible reasons may have been the
soil water content at harvest and distribution of pre?
cipitation during the fallow period. High precipita-i
tion in late May and June (Figure 2) when wheat
approaches maturity sometimes results in relatively?
high soil water contents at wheat harvest, thus re-"f
ducing the potential for water storage during the
subsequent fallow period. On the other hand, they
prevalence of lower precipitation as grain sorghum.
approaches maturity results in low soil water contents;

ING SEASON:
Coefficient
Cropping Net regression coefficients‘ °f mulilPle 1
determination ,
system bu‘ b1 b: b: b4 bs be i
CS -—847 5407*‘: — IO I9) 226 4O 68 0.893**
(Standard error) i 24 i I 2) i I 04 i 564 i 275 i 276
WSF -s64 500*" 17o 132 -s0s 283 - 39 0856*
(Standard error) i I 32 i I Q4 i I 37 i 524 i 243 i 266

‘Determined to a 4-foot depth. Unavailable water to 4 feet totals 9.97 inches. The b1 regression coefficient is associated with this factor.

‘The growing season periods and associated regression coefficients were germination and seedling establishment—seeding to 20 days after.
seeding (b2); rapid vegetative growth-—2l to 50 days after seeding lbs); late boot to flowering—-5I to 65 days after seeding (b4); grain ,

filIing--66 to 80 days after seeding lbs); and grain hardening to harvest—8l days after seeding to harvest Ibo).
‘The b coefficients indicate pounds of grain per acre per acre-inch of water.

‘Y-—intercept.

‘Asterisks denote statistical sign-ificance (one-S-percent level; two-—l-percent level).

14

  
   
  
  
   
   
  
 

_ a greater potential for water storage
_ ow period. Thus, although storage

~  was less for the WF than for
-- actual water contents at seeding were
- WF than for the WSF system (Table 2).

precipitation amounts during the fallow
CS and CW systems were similar (7.50
), water storage was about 68 percent
_ than for CW. The resultant storage
a 14.8 and 20.1 percent for the CW
’_ respectively. The soil water content
the two crops, as discussed in the pre-
 ph, was a major factor influencing
31in the interval between crops. Another
- ce undoubtedly was the distribution
4‘- Lower precipitation prevailed as
approached (August and September)
i- sorghum seeding (May and June).

Except for the WF system, water storage effi-
ciencies were somewhat higher during the 1967-70
period than during the 1959-70 period. However,
data from the longer period should give a better
indication of the treatment effects on water storage
than from the 1967-70 period. Water storage effi-
ciency for the WSF-RF system was slightly less than
for the WSF system. The permanent ridge-furrow
system did not enhance water storage as anticipated,
but the period during which this system was included
possibly was too short to validly test the system.

Multiple linear regression analysis was used to
establish relationships between available water re-
maining in the soil at harvest of, the previous crop,
precipitation during the fallow period and available
soil water at seeding. A summary of the results for
the wheat and grain sorghum cropping systems is given
in Table 7.

g -' OF MULTIPLE LINEAR REGRESSION ANALYSIS ASSOCIATING AVAILABLE SOIL WATER AT SEEDING OF WHEAT AND
r ON DRYLAND WITH AVAILABLE SOIL WATER AT HARVEST‘ AND PRECIPITATION DURING DIFFERENT PORTIONS OF THE

Factor or
precipitation period

Y-intercepi”
and net
regression
coefficients‘

Symbol Value

Coefficient
of multiple
determination

WSF

Available soil water at harvest
Precipitation—-harvest to July 3I
Precipitatiom-August
Precipitatiom-Sept. I to seeding

Available soil water at sorghum
Precipitatiom-sorghum harvest to

Precipitatiom-Dec. I to April 3O

Precipitatiom-May I to June 30
Precipitation-July I to Aug. 3I
Precipitation-Sept. I to seeding

Available soil water at harvest
Precipitation—-harvest to July 3I
Precipitation-Aug. I to Nov. 3O
Precipitatiom-Dec. I to April 30
Precipitation-—May I to June 3O
Precipitation-July I to Aug. 3I
Precipitation—Sept. I to seeding

Available soil wafer at harvest
Precipitation-harvest to Nov. 3O
Precipitatiom-Dec. I to Feb. 28

Precipitation—-Mar. I to April 30

Precipitation—-May I to seeding

Available soil water at wheat harvest

Precipitation-harvest to July 3I

Precipitatiom-Aug. I to Sept. 30
Precipitation-—Oct. I to Nov. 30

Precipitation—-Dec. I to Feb. 28
Precipitation-Mar. I to April 30
Precipitation-May I to seeding

—O.8I2 ‘0.873**
.316
.337
.366

.357

0.945 0.8I I**
.297
.489
.095
.OI 2
.225
.I 64

harvest
Nov. 3O

0.303 0732*
.096
.126
. I OI
— .203
.l I3
.357
.I 38

I .260 O.552N.S.
0.545
0.345
0. I O4
0.072
0.093

2.329 0.672*
--0.039

0.009

0.386

0.226
—0.l 80
—0.6I7
—0.074

???????? 9'91’??? 9'??????? 9'?????9' 9'9‘???

  
    

-foot depth. Unavailable water to 4 feet totals 9.97 inches.

Indicate Inches of water storage per inch of soil water at harvest or inch of precipitation received.
statistical significance Ione-S-percent level; two-—I-percent level).

15

The net regression coefficients suggest relatively
high water storage from precipitation occurring soon
after sorghum harvest and relatively low storage after
wheat harvest except for the continuous wheat system.
The coefficients also suggest rather high storage of
precipitation as wheat seeding is approached and
rather low storage as sorghum seeding is approached.
This latter suggestion is contrary to expectations as
discussed earlier. A possible explanation would be
the effects of temperature and other climatic condi-
tions on evaporation. Potential evaporation is higher
before sorghum seeding than before wheat seeding.
Consequently, water storage as a portion of precipi-
tation received may indeed be greater before wheat
seeding than before sorghum seeding, but total water
stored may still be greater before sorghum seeding
than before wheat seeding due to the greater amounts
of precipitation occurring before sorghum seeding.
Of course, the limitations of multiple regression anal-
ysis are realized, and the suggested trends may be
coincidental rather than real.

For continuous wheat, precipitation between crops
had a relatively constant effect on available soil water
at seeding as indicated by the net regression coeffi-
cients. Also, the high coefficient of multiple determi-
nation (R2 = 0.873) suggests a minimum true correla-
tion between the dependent and independent variables
of about 0.65 (P = 0.95) as determined from graphs
and discussion presented by Ezekiel and Fox (1959,
pp. 295-298). The high coefficient of multiple de-
termination for the CW system further suggests that
available soil water at seeding can be estimated with
a fair degree of accuracy from a knowledge of avail-
able water remaining in the soil at harvest and the
precipitation during the nongrowing season. By hav-
ing a fairly reliable estimate of available soil water at
seeding time, the producer could decide whether to

" seed wheat (continuous wheat) or whether to fallow

the land with hopes of better returns from a sorghum
crop the following year or a wheat crop a year later.

The low efficiency of fallow for storing precipi-
tation as soil water and increasing crop production
points to a need for flexible crop management and
cropping systems. When the soil water content is
high at harvest, water storage during the fallow period
is low or water may actually be lost from the soil. In
such cases, the water in the soil and subsequent pre-
cipitation may be more efficiently used if another
crop is seeded immediately after harvest. After wheat,
grain sorghum or a forage crop for livestock could be
used. The increasing cattle industry in the area
presents attractive possibilities for forage production.
Wheat for grain or grazing could be seeded after grain
sorghum when soil water conditions are favorable.
Other possibilities would be to seed wheat or sorghum
continuously if soil water conditions are favorable
rather than seed the alternate crop later in the fallow
period.

Without drastic alterations of the soil surface
(microwatersheds, waterproofing, continuous mulches,

16

and so forth) and possibly profile modification, it is
doubtful that the efficiency of fallow with respect to
water storage and crop production can be markedly

increased by current cropping practices and systems. .

Flexible management and cropping systems will have

to be used to make the most efficient use of all avail- .

able water supplies.

Soil Properties
Organic Matter

The organic matter content in the surface 6 inches A
of soil was significantly higher in 1970 for the grass s

treatment than for any of the other treatments (Talble

8). The difference in organic matter content due to 1
the CW and CS treatments was significant also. _

Although not necessarily significant, all samples for

the 0- to 6-inch depth from treatments with only wheat’
in the cropping system had higher organic matter 1
contents than those from systems with wheat and sor- a
ghum or sorghum alone. Apparently, wheat is more.
conducive to maintaining the organic matter level .
of a soil than grain sorghum. Similar conclusions

were reached by Hobbs and Thompson (1971).

In comparison with the organic matter content.
of samples obtained from the surface 6 inches of soil
when the study was started in 1957, only the CS treat-A‘
ment resulted in an organic matter content decreasef
The increase in organic matter contents as a result-
of the grass and CW treatments were significant. The‘
increases due to the WF, WSF and WSF-RF treat-
ments were not significant according to the unpaired 3

"t" tCSL

The organic matter contents of samples from
other depths were not significantly different as a result

of the treatments except for the 24- to 36-inch depth‘
of the WF treatment over the CW treatment. This
difference evidently was due to random variation
because there was no logical reason for that difference

TABLE 8. SOIL ORGANIC MATTER IN 1970 AS INFLUENCED B V
WHEAT AND GRAIN SORGHUM CROPPlNG SYSTEMS AND BY GRASS

ON DRYLAND; MEAN VALUES FOR SAMPLES COLLECTED AT lNlTlA-
TlON OF THE STUDY IN 1957 1

Soil depth (inches)

Treatment 0-6 6-12 12-18 18-24 24-86 36-4

Percent

cw 11.86"‘ 1.40" 1.22"‘ 0.86"" 0.53" 0.44‘
w1= 1.77"‘ 1.84" 1.08"‘ .89“ .80“ .47‘ 1
ws1= 1.70"" 1.40" 1.09" 85°“ .68“ .45‘ ».
WSF-RF 1.70"‘ 1.30"" 1.18" 92"" .66“ .46‘
cs 1.64" 1.80"" 1.20"‘ .79“ .68“ .47‘ .
Grass 2.04‘ 1.45"‘ 1.14" .84“ .70“ .52‘
Mean (all .

treatments) 11.79‘ 1.87‘ 1.14‘ .85‘ .68‘ .47‘ A
Mecm (all

samples——

19571 1.66 1.22 1.08 0.86 0.64

‘Column or row values or mean values followed by the same lett
or letters are not significantly different at the S-percent lev
(Duncan's Multiple Range Test). ;

  
    
 
  
   
  
   
   
   
  
   
  
   
    
   
  
   
   
  
   
  
  
   
 
   
   
  
   
 
  
   
 
  
  

organic matter contents at other depths for
f‘ ents were similar

0, the organic matter content of samples
A 6- to 12-inch depth of the grass plots was
‘t greater than that of the initial soil
(m this depth, and this increase was sig-
or the other treatments and depths, the
, tter contents were similar to those of the
_ samples. The increase in soil organic
'1 ent for the grass plots apparently resulted
turn to the soil of all forage produced.
(was not grazed or removed as hay. The
 to the CW treatment and the tendency
 ses due to the WF and WSF treatments
cted. Generally, it is considered diffi-
tain soil organic matter contents under
nditions and even more difficult to in-
 Hobbs and Thompson (1971), however,
 reversal of the downward trend in soil
ter when a change was made from con-
hum or sorghum-fallow cropping systems
-wheat-sorghum cropping system. They
the increase to different equilibrium or-
' levels for the different systems. Namely,
E’ H systems had lower equilibrium levels
llow-wheat sorghum system-hence, the
en the latter system was introduced.

 ild area used for the study of this report
pped to wheat continuously or alternately
,_ since it was broken from sod in 1927.
 initially broken from sod in 1919 but
"1 grass after 2 years of sorghum.) Since the
n cropped to wheat previously, the reasons
ses in organic matter for the 1970 sam-
f: 1957 samples are not readily apparent.
Vle reason may be that the organic matter
‘ 'unduly low in 1957. A major drouth
preceding 7 years resulted in either crop
low residue production. Microbial activity
gluring the drouth, and the low residue
resulted in a net reduction of soil organic
 er the drouth, higher residue production
soil organic matter level. Data reported
1968) for an adjacent dryland wheat tillage
q g practices study (CW and WF) were simi-
,ing the reliability of the organic matter
ble 8.

h not statistically significant except for
lues for depth, several trends in soil bulk
.ble 9) are apparent. At all except the 0-
d the 36- and 48-inch depths, the bulk
ighest  the CS treatment. The higher
 for this treatment possibly was related
 organic matter content.

 0- to 6-inch depth, the bulk density was

the grass treatment. This depth was the
or all except the grass treatment, which

TABLE 9. SOIL 8uu< DENSITY IN 1-1 readily moves into soil

AND GRAIN SORGHUM CROPPlNG fol-es and’ in many Cases,

DRYLAND
further water entry.
5°" detnces in the distribu-
Treatment 0-6 6-12 12-18 ‘ound in this study,
in water intake for
G/Cm ystems could be
cw ‘0.95 1.50 1.56 1.5 the differenees
w1= 1.09 1.47 1.51 . 6
WSF 1.04 1.52 1.56 1.59 ‘it’ ‘(ere Ida‘
WSF-RF 1.07 1.57 1.55 1.64 11g influence
cs 1.13 1.65 1.65 1.69 ‘q d -
Grass 1.51 1.57 1.46 1.59 $6.0m ’ 5°11
Mean (all Hons was
treatments) '1.13' 1.55” 1.55“ 1.61‘ 1.5: reduces
a. soil is

‘Column and row and 'mean values are not significantly a.
aMean values followed by the same letter are not significaiqccurs‘
ferent at the 5-percent level (Duncan's Multiple Range Test) loam

been

explains the lower bulk densities for the tillage p?“
The grass plots had not been tilled since the grills
was established in April 1958. The mean bulk densitlfi
for the 0- to 6-inch depth was significantly lower than
for the other depths. Again, these differences resulted
from tilling the surface layer of all plots except those
1n grass.

Dry Aggregates

The percentages of aggregates in the different size
ranges were significantly affected by the different treat-
ments imposed during the study period (Table l0).
According to Woodruff and Siddoway (1965), about
75 percent of the aggregates (clods) on large, bare,
smooth, unprotected fields should be greater than (>)
0.84 mm in order to hold average annual soil losses
by wind erosion to less than the tolerable level of
5 tons per acre. All cropping systems resulted in less
than the required amount of large aggregates to con-
trol wind erosion effectively (Table 10). For the less
than (<) 0.84-mm fraction, the amounts ranged from
35.8 percent for the T/VF system to 46.0 percent for
the CS system, and the differences were significant as
indicated in Table l0. Soil of the CS plots would
be more erodible than that of other plots, while the

TABLE 10. DRY AGGREGATE SlZE DISTRIBUTION IN 1970 AS INFLU-
ENCED BY WHEAT AND GRAlN SORGHUM CROPPING SYSTEMS AND
BY GRASS ON DRYLAND

Soil fraction size (mm)

Treatment <0.84 0.84-2.0 2.0-6.4 64-183 >18.3 MWD’

i Percent mm
cw ‘39.4* 13.1"‘ 17.9" 20.8"“ 8.8‘ 7.87
WF 35.8‘ 12.7"‘ 17.9" 22.5‘ 11.3" 9.26
ws1= 39.6‘ 13.1"‘ 17.6" 21.6"‘ 8.1” 7.63
WSF-RF 41.5‘ 15.0"’ 17.7“ 17.8“ 8.1‘ 7.20
cs 46.0‘ 14.1"‘ 183°" 18.7" 2.8‘ 4.81
Grass 13.2"‘ 13.1"‘ 22.2“ 28.2‘ 23.3‘ 15.78
Mean (all

treatments) 235.9‘ 13.51’ 18.6” 21.6“ 10.4‘

‘Mean weight diameter.

zColumn or row values or mean values followed by the same letter
or letters are not significantly different at the 5-percent level (Dun-
can's Multiple Range Test).

17

The net regression .1'Odibl8. CIOP residues on [116
high water stgrage frQmVid€d gOOd PYOICCtiOII during
after sorghum harvgst 1T1)’ spring erosion period. The
wheat; harvest excgpt were not tilled until April, and
The Cggfficientg alat that time maintained most of
precipitation as W sllffaCe-

rather 10W 5mm‘ the grass treatment, there were no
This latter S119

, yces in the percentages of aggregates in
‘t1QS°“5fSf°dteaf‘2.0-, 2.0- to 0.4- or 6.4- to l8.3-mm size
t. e e 6c S fto treatments. For the >l8.3-mm range,
1on5 on eyeatment resulted in a significantly lower
before so‘

Conseqw‘ percentage than the other treatments.

tation though the data are not shown, separate statis-
seedininalyses were made for the dry aggregate distri-
storedn data from the WF, WSF and WSF-RF cropping
than-ms. For the WF system, the wheat plots (plots
of m which wheat was harvested in 1970) had signifi-
Ofltly less fine aggregates (<0.84 mm) than fallow
yrlots (plots that were fallowed during the 1969-70

 rwheat season). Also, the wheat plots had significantly

more aggregates in the >l8.3-mm range than the
fallow plots. For the WSF-RF system, significant
differences were found between the percentages of
aggregates in the <0.84-mm range for the wheat,
fallow and sorghum plots, with wheat having the
lowest and sorghum having the highest percentage of
aggregates in this size range. The wheat plots had
significantly more aggregates in the >l8.3-mm range
than either the fallow or sorghum plots. Although
not significant, the trends for aggregates from the
WSF system were similar to those of the WSF-RF
system. These data suggest that wheat was more
conducive to stabilizing dry soil aggregates than sor-
ghum but that the effects of wheat on the stability
of dry aggregates were relatively short lived.

The grass treatment resulted in significantly fewer
aggregates in the <0.84-mm size range than any of the
other treatments and significantly more aggregates in
the 6.4- to 18.?» and the >l8.3-mm size ranges than
the other treatments.

Another method of indicating the differences in
dry soil aggregation between treatments is through the
calculation of a mean weight diameter (MWD) for
each treatment (Kempe and Chepil, 1965). The MWD
is equal to the sum of the products of the mean
diameter (Y1) in millimeters of each size fraction and
the proportion of the total weight (wi) occurring in
the corresponding fraction. The amount passing
through the finest sieve is included. A maximum
diameter of 76.2 mm was assumed for the largest frac-
tion. The equation used was

n
 '= 2 Kiwi
i=1

Size distribution data (Table l0) for the different
treatments were used to calculate the MWD values.
Data for the CW treatment were used in the follow-
ing example of the calculations:

l8

MWD = [(0.42 mm X 0.394) + (1.42 mm
>< 0.131) + (4.20 mm x 0.179)
+ (12.35 mm X 0.208) + (47.75
mm >< 0.088)]
= 7.87 mm.

The calculated MWD’s are included in Table l0. By
the nature of the calculations, the percentages of A1

coarse aggregates have a greater; influence on the

MWD than those of the fine aggregates, and high a
MWD values reflect greater amounts of coarse aggre-
gates for a treatment than low MWD values. The?
MWD was higher for the grass treatment than for’
the tillage treatments. For the different croppiqg“
treatments, the MWD was highest for WF and lowest.
for CS, again indicating greater erosion susceptibility
for the CS treatment than for the other treatments.

The dry aggregate size distribution data indicate
that all cropping systems on dryland could lead to

serious wind erosion problems. However, through use.

of stubble-mulch tillage which maintained most of
the crop residues on the surface, wind erosion gen-
erally was controlled adequately under the prevailing

conditions. However, in some years, residue produc-
tion is low, and erosion by wind may be severetj

Under such conditions, tillage which increases thef
cloddiness and roughness of the soil surface may be
necessary to reduce wind erosion to tolerable levels.‘

Samples for the dry aggregates size distribution‘

determinations were collected in November, and it:
is recognized that different size distributions may have:
been found had the samples been collected at some:

other time. For example, freezing and thawing a w"
known to pulverize the soil surface. However, the.
data presented are satisfactory for determining th;
relative influence of the cropping systems on soi?
erosion susceptibility but should not be used to indi.
cate the absolute erosion potential during the criti ,_
late winter-early spring erosion period. g

Wet Aggregates g

The size distribution of soil aggregates wetti
under vacuum was determined by wet sieving the soil
under water. The percentages of aggregates in th
different size ranges as determined by the amoun
retained on the different sieves and the amount pa w
ing through the finest sieve are given in Table 1l_
Also given are mean weight diameters calculat
according to the procedure illustrated in the sectio 
pertaining to dry aggregates.

The different wheat and sorghum cropping sy
tems significantly affected the percentages of aggr,
gates in the <0.25- and the 4.0- to 12.7-mm size rang_
but not in the intermediate ranges. The grass trea
ment resulted in significant differences as compar
with other treatments in the <0.25-, 2.0- to 4.0- an
4.0- to 12.7-mm size ranges. ,

The grass treatment resulted in the lowest vif
the WF and WSF treatments resulted in the highesj?‘
percentages of aggregates in the <0.25-mm size ran_
The low percentage of fine aggregates (<0.25 f

   
 
  
   
 
 
  
  
  
   
 
   
    
  
 
 
  
   
   
  
  
   
   
   
  
  
   
  
   
  
  

AGGREGATE SIZE DISTRIBUTION IN I970 AS IN-
‘ T AND GRAIN SORGHUM CROPPING SYSTEMS
DRYLAND

SoiI fraction size (mm)

’ 0.25-I.O I.0-2.0 2.0-4.0 4.0-I2.7

MWD‘

Percent mm

27.2‘ 7.6‘ 8.9‘ 19.9"‘ 2.26
27.7‘ 7.2‘ 8.0‘ 15.1"‘ 1.4a
25.0" 7.4‘ 8.3‘ 17.2"‘ 1.60
29.3‘ 7.3‘ 7.6‘ 16.8"“ 1.58
27.9‘ 7.7‘ 8.9‘ 19.9“ 2.27
27.1‘ 12.2"’ 12.9"‘ 22.4“ 2.64

27.4‘ .8.2‘ 9.1‘ 18.6“

I Ives or mean values followed by the same letter
Ignlficantly different at the 5-percent level (Dun-
- Test).

wtment was attributed to the organic
) accumulation (Table 8) under the
__--r was beneficial for soil aggregation.
 of grass on soil aggregation was also
'5'“ higher percentage of aggregates for
ent in the 2.0- to 4.0- and 4.0- to
ges. The organic matter under grass
together in larger aggregates. This
of sod on soil aggregation is widely

were higher percentages of fine aggre-
, WSF and WSF-RF treatments than
~ CS treatments apparently was related
 Data in Table ll for the WF,
_ -RF treatments are averages for the
"it in the crop rotations. Although
' ' 'vidual sequences are not shown, the
ne aggregates was greater for plots
WSF-RF systems that were {allowed
g -pping season than for plots that were
heat or grain sorghum. Apparently,

ad a beneficial effect on soil aggre-
{irelatively short period. The fallow
p. sufficiently long so that most of the
“ athered by the time of sampling and,
dues no longer benefited soil aggre-
ahe CW and CS systems, the intervals
“gwere relatively short, and greater aggre-
i‘ 1 - nt.

s in the percentages of aggregates for
atments for the 4.0- to 12.7-mm range
_ *the trends for the <0.25-mm range.
V" ect the ability of residues from recent
 ~~ to promote the formation of large
y This ability was also reflected in the
I for  different treatments.

h higher percentages of water-stable
fills-the larger size ranges should permit
 intake than soils having higher
i'_ fine aggregates. The fine aggregates
 ult from dispersion of the soil upon

wetting. This fine material readily moves into soil
pores with water, clogs the pores and, in many cases,
virtually seals the soil against further water entry.

Although significant differences in the distribu-
tion of the wet aggregates were found in this study,
it is doubtful whether differences in water intake for
the wheat and sorghum cropping systems could be
detected under field conditions. First, the differences
between treatments, although significant, were rela-
tively small, and presumably the resulting influence
on water intake would be small also. Second, soil
for the aggregate distribution determinations was
wetted under vacuum. This type of wetting reduces
soil dispersion. ,Under field conditions, the soil is
rapidly wetted by precipitation, and dispersion occurs.
Differences in aggregate stability of Pullman clay loam
as a result of rapid and vacuum wetting have been
shown by Unger (1969). Also, raindrops striking bare
soil enhance dispersion, whereas even small amounts
of surface residues intercept raindrops and reduce soil
dispersion. Weathered residues serve this purpose
but evidently had little influence on the size distribu-
tion of wet aggregates as discussed previously. Conse-
quently, water intake on the rotation plots (WF, WSF

and WSF-RF) under field conditions may be greater”

than on the continuous cropping plots (CW and CS)
because the rotation plots, even at the end of the
fallow period, have weathered residues on the surface
which intercept the raindrops. This possible differ-
ence in water intake during the growing season by
soil of the rotation and continuous cropping plots
may have been a major factor in greater grain pro-
duction on the rotation plots as compared with the
continuous cropping plots. It is doubtful that the
small differences in available soil water at seeding
alone were responsible for the differences in yield.

Economics of the Different Cropping Systems
In Tables 2 and 5, average grain yields for the
different cropping systems are presented. The aver-
age yields, however, were based on the area harvested

TABLE I2. MEAN ANNUAL YIELDS BASED ON THE AREA HARVESTED
AND AREA IN THE DIFFERENT CROPPING SYSTEMS FOR WHEAT AND
GRAIN SORGHUM ON DRYLAND (I958 THROUGH I970)

Mean yield Mean yield
(area harvested) (area in cropping system)
Cropping Total
system Wheat Sorghum Wheat Sorghum grain‘
Lb./acre
Continuous
wheat 628 628 628
Wheat-
sorghum-
fallow’ 758 I557 253 5I9 772
Wheat-
fallow“ 866 433 433
Continuous
sorghum I I37 I I37 I I37

‘Accounts for differences in area harvested each year and area in
the cropping systems.

23-year system.

'2-year system.

I9

TABLE 13. SUMMARY OF FIELD OPERATIONS‘ PERFORMED FOR THE
WHEAT AND GRAIN SORGHUM CROPPING SYSTEMS ON DRYLAND.
THE VALUES SHOWN ARE THE AVERAGE NUMBERS PERFORMED FOR
THE PRODUCTION OF A CROP

 

Wheat Sorghum
Operation CW WF WSF CS WSF
No. / year
Seeding 1.0 1.0 1.0 21.3 ‘L3
Sweep tillage 2.8 7.2 4.6 1.9 4.6
Cultivations 1.0 1.0

‘Occasional operations such as spraying for weeds and insects and
use of the rotary hoe to aid emergence of grain sorghum are not
included because they were not performed a sufficient number of
times to establish trends.

‘Sorghum was reseeded in about 1 year in 3.

each year and did not take into account the differ-
ences in acreages devoted to the different cropping
systems. For an economic analysis of the cropping
systems, differences in acreages along with production
requirements and retums for the different systems
must be considered. In Table 12, average yields are
presented on the basis of the harvested area and the
area in each cropping system. Table 13 summarizes
the field operations performed for each cropping
system during the study period.

Data in Tables 12 and 13 along with prevailing
costs and grain prices can be used to obtain estimated
incomes and expenses per harvested acre for the differ-
ent cropping systems. A complete economic analysis
of the cropping systems, however, is beyond the scope
of this paper. For those interested, the guidelines
published by Grubb, Moore and I..acewel1 (1967) and
by Osborn, Moore and Ethridge (1969) for the pro-
duction of dryland wheat and grain sorghum would
be useful for determining the system most suitable
for a particular production enterprise.

ACKNOWLEDGMENT

This study was a cooperative effort between the
Agricultural Research Service, U.S. Department of
Agriculture, and The Texas Agricultural Experiment
Station, Texas A8cM University.

Numerous persons have been involved in this
study since it was initiated in 1957. The author is
indebted to all of them and especially to T. J. Army
and J. J. Bond, who planned and initiated the study
and managed it the first 4 years, and to W. C. Johnson
and C. E. Van Doren, who managed it in subsequent
years; also to F. O. Wood, who performed most of
the field operations in the latter years of the study,
and to J. J. Parker, who assisted in performing the
soil physical analysis at the conclusion of the field
study.

LITERATURE CITED

l. Bloodgood, D. W., R._E. Patterson and R. L. Smith. 1954.
Water evaporation studies in Texas. Texas Agr. Expt. Sta.
Bull. 787.

2. Bond, J. J., T. J. Army and 0. R. Lehman. 1964. Row
spacing, plant populations and moisture supply as factors
in dryland grain sorghum production. Agron. Jour. 56:3—6.

20

l0.

12.

l3.

14.

15.

16.

17.

18.

19.

21.

23.

24.

Bonnen, C. A. 1960. Types of farming in Texas. Texas J

Agr. Expt. Sta. Bull. 964.

Brown, Paul L., and W. D. Shrader.
evapotranspiration, and water use efficiency of grain sor-
ghum under different cultural practices. Agron. Jour.
51:339-343.

Eck, Harold V., and Carl Fanning.

1959. Grain yields, ‘

1962. Fertilization of .

dryland winter wheat under stubble-mulch and one-way 

tillage on Pullman silty clay loam. _ Texas Agr. Expt. Sta. ii

Misc. Pub. 584.

Eck, Harold v., and Billy B. Tucker‘: 196s. Winter wheat

yields and response to nitrogen as affected by soil and
climatic factors. Agron. Jour. 60:663-666.

Ezekiel, Mordecai, and Karl A. Fox. 1959. Methods of
correlation and regression analysis, 3rd edition. John Wiley
and Sons, Inc., New York. l‘
Fisher, C. E., and Earl Burnett. 1953. Conservation and;
utilization of soil moisture. Texas Agr. Expt. Sta. Bull. 767."
Grubb, Herbert W., D. S. Moore and R. D. Lacewell. 1967.»
Expected production requirements, costs and returns for.

major agricultural crops; fine-textured soils—Texas High‘.

Plains. Texas Agr. Expt. Sta. Misc. Pub. 848.
Hobbs, J. A., and C. A. Thompson.

Kansas Argiustoll (Chernozem). Agron. Jour. 63:66-68. .
Hughes, William F., and Wyatte L. Harman. 1969. Pro-i
jected economic life of water resources, subdivision number
1, High Plains Underground Water Reservoir. Texas Agrl
Expt. Sta. Tech. Monograph 6. _
Johnson, Wendell C. 1950. Stubble-much farming on:
wheatlands of the Southern High Plains. United States,
Dept. Agr. Circ. No. 860. 3

Johnson, Wendell C., 1964. Some observations on the con j A
tribution of an inch of seeding time soil moisture to wheat _

yield in the Great Plains. Agron. Jour. 56:29-35.

Johnson, Wendell C. 1966. Soil moisture conservati H:
during fallow periods on wheatlands in the Southern Grea
Plains. Proc. Fourth West Texas Water Conference. Te t 
Tech University, Lubbock, Texas, pp. 37-43. "-
Kemper, W. D., and W. S. Chepil. 1965. Size distributi -;
of aggregates. In Methods of Soil Analysis, Part I. C. ' i
Black, Editordn-Chief. Agron. 99199-510.
Luebs, R. E. 1962. Investigations of cropping system
tillage methods, and cultural practices for dryland farmi 
at the Fort Hays (Kansas) Branch Experiment Statio '
Kansas Agr. Expt. Sta. Bull. 449. j
Musick, J. T., and W. H. Sletten. 1966. Grain sorgh w
irrigation-water management on Richfield and Pullm 
soils. Trans. ASAE 9:369-371, 373. _
New, Leon. 1970. High Plains irrigation survey. T t
Ext. Service (mimeo report).

Osborn, James E., Donald S. Moore and Don E. Ethridge
1969. Expected production requirements, costs and retu ~-
for major crops; medium-textured soils-Texas High P1'-_
Texas Agr. Expt. Sta. Misc. Pub. 922.

Piper, C. S. 1944. Soil and plant analysis. Interscien
Pub., Inc., New York, pp. 223-227. .-
Taylor, Howard M., C. E. Van Doren, Curtis L. Godfr
and James R. Coover. 1963. Soils of the Southweste -~,
Great Plains Field Station. Texas Agr. Expt. Sta. M’
Pub. 669.

Unger, Paul W. 1968. Soil organic matter and v;-
changes during 24 years of dryland wheat tillage an
cropping practices. Soil Sci. Soc. Amer. Proc. 32z427-4 '1
Unger, Paul W. 1969. Physical properties of Pullman sil l
clay loam as affected by dryland wheat managament pra
tices. Texas Agr. Expt. Sta. Misc. Pub. 933. y.
Woodruff, N. P., and F. H. Siddoway. 1965. A win
erosion equation. Soil Sci. Soc. Amer. Proc. 29:602-608.
Yearbook Statistical Committee, James R. Kendall, Chair-l
man. 1970. Agricultural Statistics. United States Govt.‘
Printing Office, Washington. a

1971. Effect of culti-i
vation on the nitrogen and organic carbon contents of £7‘