IRRI Saturday Seminar
May 6, 1978
TESTING OF RAINFED LOWLAND RICE-BASED
CROPPING PATTERNS IN ILOILO
CROP YEAR 1977-78
R.D. Magbanua, M. Genesila, E.C. Price, H.G. Zandstra
and R.A. Morrisl/
Cropping systems research has been conducted in the munici-
palities of Oton and Tigbauan, Iloilo during the last three years
by the cooperative IRRI-BPI project. The descriptive phase started
in late 1974, while actual cropping pattern testing began in early
1975. The objectives and rationale-of the project, which are to
identify and develop intensified crop production technology for rainfed
and partially irrigated areas with conditions similar to Iloilo,have
been presented in previous seminars (Palada et al, 1976; Magbanua et
al, 1977). The Iloilo research area falls within a 2.3 climate pattern,
i.e., based on long term averages, there are 5-6 consecutive wet months
with more than 200 mm rain per month and 2-4 dry months with less than
100 mm per month (IRRI, 1974; IRRI, 1976). The site's climatic and
land systems characteristics are common and widespread in other rice
growing areas of Iloilo and nearby provinces. Although the Iloilo
municipalities of Tigbauan and Oton are primarily rainfed, an increas-
ing number of land units within the research area are receiving partial
irrigation of varying durations beyond the normal end of the wet season.
The main function of the cropping systems research site is to
test improved cropping patterns, including examination of the factors
determining pattern adaptation and the effectiveness of individual
management components within the pattern. Cropping patterns are tested
in farmer-cooperators' fields under the cooperator's management. The
project provides material inputs for each farmer-cooperator's field
(approximately 0.1 ha), but the cooperator is responsible for crop
establishment and other cultural operations.
Site Coordinator, Outreach Economist, Agricultural Economist,
Head, and Agronomist/Philippine Outreach Coordinator respectively;
Cropping Systems Program, International Rice Research Institute, Los
Experiments on various crop production components such as
pesticide, fertilizer and tillage experiments, were also conducted
in support of the cropping pattern testing.
The discussion in this paper will focus on the design, testing
and evaluation of agronomic and.economic performances and farmers'
reactions in the management of the test patterns during the 1977-78
1977-78 Cropping Pattern Design
On the basis of previous years' cropping pattern testing, the
1977-78 cropping patterns were reallocated to take into consideration
landscape and soil texture. During the 1977-78 crop year, there were
three general cropping pattern groups proposed for testing (Table 1).
The first group consisted of two rice crops followed by an upland crop.
In this pattern group, the first rice crop could be either dry-seeded
or wet-seeded, the second rice crop could be either wet-seeded or
transplanted, and the following upland crop could be either mungbeans
and cowpeas. The second group consisted of a single rice crop (WSR
or TPR) followed by upland crqps (sorghum, peanuts, mungbeans, cowpeas,
sweet potatoes and soybeans)_!. The third pattern group involved green
corn as the first crop, followed by either WSR or TPR, and then followed
by either mungbeans or cowpeas as a crop. These three pattern groups
were allocated to a total of 96 fields distributed to four classes of
landscapes. These landscape classes were plateau, plain, sideslope
and bottomland or waterways (Figure 1). Patterns which had been located
on slemits were discontinued in CY 77/78 because summits represent only
a minor area within the test site, no promising patterns had been found
for the summits, and only upland crops are grown in patterns designed
for summits. Also, in CY 77/78, a heavy emphasis was placed on pattern
testing under strictly rainfed conditions, and therefore only eight
irrigated fields were included among the tests. The assignment of
cropping patterns to fields was based on the description of landscape
and soil characteristics which had been made prior to the start of the
Description of the 1977-78 Rainfall Pattern
The 1977-78 rainfall distribution was exceptionally abnormal. It
was characterized by a delayed onset, an extremely high peak, and an
early and sharp end (Figure 2). There were only four months with more
than 200 mm of rain. April was extremely dry. The first rains occurred
-/Prior to planting upland crops, a decision was made to limit
upland crops to sorghum, peanuts, cowpeas or mungbeans, and dispense
with sweet potatoes and soybeans because of the unsuitability of the
latter two crops.
in May, and the total for May was 31 mm which is only 18% of the
long-term average. June, July, August and September were wetter
than the preceding crop year (1976-77) and the 25-year averages.
A series of typhoons in September produced a total of 834 mm for the
month, the highest recorded since 1950, the start of available records.
From October to March conditions were very dry (< 100 mm).
Based on a rainfall accumulation of 75 mm received (as an onset
criterion) and 100 mm accumulation remaining (as a termination
criterion), the length of the CY 77/78 growing season was only 151
days against 225 days in CY 76/77 and 212 days as the long term
average (Morris and Zandstra, 1978). The delayed onset and the
pattern of distribution of the early rains which occurred in June
affected greatly the establishment of the first crop as will be
AGRONOMIC PERFORMANCE OF CROPPING PATTERNS
AS INFLUENCED BY SITE VARIABLES
The pattern performance discussed in this section is the per-
formance of patterns grown by farmer-cooperators, following farmer-
cooperators alterations of proposed patterns in response to factors
such as unexpected weather or labor or power shortages. However,
the initial portion of the section discusses several major reasons
why farmer-cooperators altered the first crop in a proposed pattern
or planted later than planned.
Initial crops and planting dates as influenced by weather and
site variables. As indicated in the section describing the rainfall
pattern, the delay in early rains affected the establishment of the
first crop. Table 2 presents the actual crop components established
on different landscape positions. All proposed green corn plots were
shifted to WSR, eliminating the green corn-rice-upland crop pattern
group. The delay in the onset of the rain prompted the farmers to
plant rice because it was either (1) not possible to plant corn due
to ponding of the water in paddies or (2) a corn crop would delay
the planting of a following rice crop so that the latter would be
more vulnerable to drought stress. Of the nine fields proposed to
be dry-seeded, seven fields were actually planted and two were shifted
to WSR. Six of the seven dry-seeded rice plots were successfully
grown on plateau and sideslope positions where water takes longer to
accumulate. The rapidity of water accumulation on plains induced
farmers to puddle their fields and wet-seed or transplant instead of
The majority (72%) of the first crops were established as WSR.
The WSR establishment technique was used in all landscape positions.
Due to lack of water for early seedling production, most farmers
chose to wet-seed so that only 20 fields were transplanted under the
farmers' WSR/TPR option.
The timing of the first crop establishment varied by landscape
position (Table 3). Fields located in the bottomlands were planted
as early as the second week of June (week 24) or about one week after
the occurrence of first rains. Most crops were established in this
landscape position in weeks 25 and 26. The fields in the plateau and
plain positions were planted between weeks 25 to 31, with a peak in
week 27. The latest plantings were observed in the sideslope position.
The differences in planting dates according to landscape position may
be explained by the differences in the earliness of water accumulation.
In general, water accumulation occurs in the order (1) bottomland,
(2) plain, (3) plateau and (4) sideslopes. It should be pointed out
however, that many of the patterns grown on the plateau and sideslopes
consisted of only one TPR rice crop followed by an upland crop. TPR
crops are normally established later because seedlings must be produced.
Given a method of establishment, the planting date of the first crop is
critical because it influences the planting dates of subsequent crops,
and therefore the rainfall of the subsequent crops are likely to receive
late in the wet season.
Crop yields as influenced by weather, establishment methods and
site variables. Yields obtained from rainfed first rice crops in the
different pattern groups varied by landscape position (Table 4). Fields
located on bottomlands produced the highest overall mean yield (6.1 t/ha).
Those grown on plateau and plains followed with average yields of 5.1 and
4,9 t/ha, respectively. Yields were lowest (4.6 t/ha) on sideslopes.
These differences in first rice crop yields, as found over the four
landscape positions, are believed to be associated with differences in
hydrologic regimes. During the growing season, flooding periodsvaried
in the different landscape position as shown in Table 5. Fields located
in rainfed bottomlands averaged 78 days of flooding, which was the longest
of-the four landscape positions. Fields on rainfed plain and plateau
positions followed with 76 days of flooding, in both cases. Rainfed
sideslopes fields received an average only of 61 days of flooding during
the growing period.
Across the different landscape positions, mean first rice crop
yield.and the period of flooding was longer in plots with partial
irrigation. In comparison to rainfed fields, average yields were
about 1.5 t/ha higher and the period of flooding was 13 days longer
under partial irrigation. Similar flooding behavior and yield per-
formances have been found in the past (Magbanua et al, 1977; Morris
et al, 1978).
Rice or upland crops were proposed as second crops, depending on
pattern groups. These second crops were also distributed to the four
landscape positions. Except for bottomlands, all landscape positions
were proposed to be planted to some patterns containing upland crops
following a single rice crop (Table 1). However, because of low October
rainfall, most of the rice crops in the plain, plateau and sideslope
positions were shifted to upland crops. Forty-one percent of the
proposed second rice crops across landscape position were shifted to
upland crops as shown in the distribution of crops in Table 6. Most
shifts to upland crops occurred in the plateau and sideslope positions
where water receded rapidly. On the other hand, a few fields in plain,
plateau and sideslope positions which were harvested early, but had
been proposed for upland crops, were shifted to rice. Because of lack
of soil moisture, fields that were harvested late because of late
plantings, and those that were planted to a late maturing variety (IR34)
during the first crop, were often not planted to any second crop,
although the fields had been prepared. A total of 83 or (86%) were
actually planted to either a second rice or upland crop, while 13 fields
were left unplanted after the first rice crop.
The method of establishment and the position of the field in the
landscape greatly affected the yields of the second crop. Under rainfed
conditions, transplanted second rice crops averaged higher yields than
wet-seeded second rice crops, viz., 2.2 vs. 1.5 t/ha (Table 7). As will
be mentioned in the discussion of turn-around, TPR crops were transplanted
only 3 days earlier than WSR crops. However, the main advantage of TPR
is that the crop avoids late drought by being harvested 2-4 weeks earlier,
depending on age of transplant.
Fields located on bottomlands yielded higher than fields elsewhere.
As shown in Table 8, time of drying varied by landscape position and
water source class within the landscape. Partially irrigated fields
dried up 2-4 weeks later than strictly rainfed fields. Earliest drying
occurred in the strictly rainfed sideslopes and plateaus, at about four
weeks after the second rice crop establishment. Fields on rainfed plains
dried up next, with average drying coming about one month later than in
fields located on plateaus and sideslopes. The irrigated plains dried up
4 weeks later than the rainfed plains. Fields on bottomlands dried up
last; in weeks 46 and 48, on rainfed and partially irrigated portions,
respectively. These differences in drying date explain much of the
difference in second rice crop yields found among fields on the four
Although four water source classes occur within the Iloilo test
site, test patterns were concentrated in rainfed fields in CY 77/78.
During the first crop, however, eigth partially irrigated test fields
were used and these fields had supplemental water for only one month
beyond the end of the dry season. To examine the impact of partial
irrigation on the yield of first and second rice crops, mean yields
of first and second rice crops are compared across landscape positions
in Table 9. First rice crop yields from these partially irrigated
fields were much higher than yields from strictly rainfed fields, as
noted earlier. Furthermore, compared to rainfed fields, a higher
percentage of second rice crops were attempted on partially irrigated
fields. However, about 30% of the second rice crops were failures
under both rainfed and partially irrigated condtiions. Nine of the
29 rainfed fields and two of the seven partially irrigated fields did
not produce any grain. The mean yield difference of 0.9 t/ha between
rainfed and partially irrigated fields was not statistically significant
at the 5% level.
Turn-around period. The interval between first crop harvest date
and second crop planting date, i.e., the turn-around period, varied
significantly among the different pattern groups. For rice-rice patterns.
the mean interval between crops was 17.7 days which was 4.3 days short.
than last year's mean interval. The turn-around period for this pattern
group was slightly shorter when the second rice crop was transplanted
compared to when it was wet-seeded; the intervals were 16.3 and 19.1 days,
respectively, for TPR and WSR plots (Table 10). A 35.0 day average delay
occurred between rice and upland crops in the rice-uplnad cropping pattern
group. For the rice-upland cropping pattern group, the tillage technique
used to establish upland crops affected the turn-around period. When
upland crops were planted with complete tillage, the average delay was
43.4 days, whereas an average delay of only 25.5 days occurred when
tillage was completely eliminated, i.e., when seeds were broadcast just
prior to the rice harvest or on rice stubble. The difference in the turn-
around period of the two tillage technique was highly significant using
the t-test. The length of the turn-around period has important implications
regarding the performance of second crops within the patterns. Generally
cropping patterns with short turn-around times have a greater chance of
yielding well because of the higher likelihood of precipitation critical
to good crop growth. In those patterns with longer delays, later crops
are placed in a less favorable environment, vulnerable to drought stress
and thus reduced yields. Other factors affecting the turn-around period
were discussed in a special study on Iloilo cropping patterns (Magbanua
et al, 1976).
Dry-season tillage for the first crops. About one-third of the
total number of cropping pattern test fields were prepared during the
dry season following the last crop. Fields were plowed by some
farmers in anticipation of their first crop the following wet season,
especially those located on higher landscape positions, with lighter
soils and only a single rice crop or a rice-upland crop pattern the
previous year. Labor inputs, establishment weeks and weed weights
are compared in Table 11 for fields prepared with and without tillage
at the start of the dry season. Plowing before the rain increased
the total labor requirement and labor cost for land preparation, and
furthermore, 168 mh/ha of tillage were still required just prior to
planting. The dates of establishment of the first crop were not
influenced by dry season tillage. Whether the field had or had not
been prepared during the dry season, average establishment occurred
in week 27.3 and week 27.8, respectively. It should be pointed out
however, that this year's rainfall was greatly delayed but abrupt
when it did come, giving no opportunity to the farmers to work on
their fields during the early period as might be possible in a more
normal year. Weed weight was slightly higher in fields which had not
been prepared during the dry season. Mean yields were higher in fields
which had not been previously prepared. However, as noted above, fields
prepared during the dry season were located on higher landscape positions
and were often light-textured, and both these factors cause low yields
in comparison to heavy-textured lower-lying fields.
Performance of ratoon rice. Six farmer-cooperators grew ratoon
rice as an alternative to a full second rice crop or an upland crop.
The fields on which these ratoons were grown could not be planted to
uplnad crops due to the presence of standing water, but on the other
hand planting of a second rice crop would be risky because of the late
first crop harvest. Ratoon yield and maturity data is presented in
Table 12. Yields ranged from 639 to 1245 kg/ha within 35 to 49 days
from harvesting of the main crop. Except for one plot which was
fertilized with 20 kg N/ha for the ratoon, the rest of the plots
received no material inputs. The only labor expended was on harvesting.
Average production per day was only 19 kg/ha. Two farmer-cooperators
established upland crops following the ratooned rice.
Performance of upland crops. Two sets of plantings were proposed
for upland crops after rice. Mungbeans, cowpeas, sorghum and peanuts
following a single rice crop constituted an early planting while mung-
beans and cowpeas following two rice crops constituted a late planting.
However, because of the limited rainfall after the rice harvest, none
of the late plantings were established. Table 13 presents the yield
performance of the upland crops planted after a single rice crop, by
landscape position. Most upland crops-were successfully established
in the sideslopes and plateau positions because upland crops did not
compete with a second rice crop in these positions. For a given crop,
yields generally did not vary much over the different landscape positions.
Agronomic performance of pattern groups. A summary of rainfed
cropping pattern performance, as affected by different site variables,
is presented in Table 14. Because CY 77-78 contained only four months
with rainfall exceeding 200 am, no proposed rice-rice-upland crop
patterns were completed. Some farmer-cooperators grew only the rice-
rice portion and others planted a UC in place of the second rice crop.
However, there was sufficient rainfall to complete the proposed rice-
upland pattern group on most land units for which it has been designed.
With a few exceptions (rice-ratoon-UC and rice-fallow patterns;
bottomland position), yields of the first rice crop did not vary much
across the four landscape positions. However, yields of second rice
crops were substantially depressed especially in the higher landscape
positions. Moreover, many fewer farmer-cooperators attempted second
rice crops in the higher positions. The rice-upland crop pattern group
performed satisfactorily in the three landscape position in which it was
tested. Upland crops ranged from 31 to 49% in relative yields. Most
upland crop failures (45%) occurred on the plateau position, especially
when planting was as late as the first week of December (week 49) and
when crops were established with complete tillage. The rice-rice
pattern was most successful in partially irrigated fields (Table 15).
Only one rice-upland crop pattern was grown in a partially irrigated
field, indicating that partially irrigated fields were not only produced
higher yields, but were also better suited for two rice crops in the
opinions of farmer-cooperators.
ECONOMIC PERFORMANCE OF CROPPING PATTERNS
The economic performance of the different cropping pattern groups
were evaluated according to the method of crop establishment and land-
scape positions. To evaluate economic performance, net returns above
variable costs were determined and returns to scarce inputs, which include
cash and labor, were computed (Price, 1977).
To obtain the data necessary for economic evaluation, agronomy
barangay assistants kept daily log forms on which field operations and
input levels were recorded for each test pattern field. The material
and labor inputs were summarized weekly and used to compute costs
of production. From crop-cuts dried to standard moisture levles, yield
data were recorded at the end of each crop period. Costs and returns
were computed on a per hectare basis. Seeds, fertilizers and pesticide
costs were based on prevailing local market prices. Labor costs were
based on the local wage rates and crop prices were based on the monthly
prices obtained from local markets.
Economic evaluation of cropping pattern components. The costs and
returns of individual crops across patterns are discussed prior to a
general evaluation of patterns.
First rice crop. Farmer-cooperators used three techniques to establish
the first rice crop namely: dry-seeding (DSR), wet-seeding (WSR) and
transplanting (TPR). Table 16 summarizes the costs and returns of first
rice crop under these three establishment method. Among the three methods,
the WSR resulted in the highest net returns. The advantage of WSR was
mainly due to higher yield of the crop with this establishment method.
TPR fields obtained the lowest yield and highest labor cost and thus re-
sulted in lowest net return. The high labor cost of TPR was due to pre-
harvest labor operations. WSR and DSR rice were observed to require
higher material cost than transplanted rice due to herbicides. The yields
of DSR fields were intermediate between TPR and WSR fields.
The implication of the material and labor cost differences between
WSR and TPR, is that WSR would be advantageous when cash or credit is
available for materials (seeds and herbicides), there is a cost attached
to family as well as hired labor, and no severe hazards to WSR (such as
flooding or drought during grain filling) are present. The economic
performance of the first rice crop varied with the landscape positions.
Because costs were about equal for all landscape positions, net returns
from combined WSR and TPR first crops strongly reflected the yield diffe-
rences across landscape positions (Table 17). Based on net returns, WSR
was superior to TPR in sideslope, plateau and plain positions. No farmer-
cooperators used TPR in the bottomland position, where net returns for
WSR were highest. Among the three landscape positions, net returns for
TPR were highest in the sideslope. Labor costs for TPR were highest in
the sideslope, which was mainly due to high pre-harvest labor require-
ments. DSR is thought to be suitable for light-textured soils in side-
slope or plateau positions, where dry tillage is easier,and standing water
takes longer to accumulate. Cost-wise, DSR appears competitive or superior
to WSR or TPR under these conditions.
Water source classes did not significantly alter the input compo-
nents of the first crop; total variable costs were almost similar across
fields in rainfed and partially irrigated areas. There was a slight net
return advantage for partially irrigated first rice crops due to higher
Second rice crop. The performance of second rice crops, by landscape
position, is presented in Table 18. The ratoon crops grown in plateau
and sideslope positions, produced the highest return over variable cost.
These ratooned crops also had the lowest material and labor costs compared
with WSR or TPR second crops in any landscape position. Full second crops
in sideslope and plateau positions were obviously not competitive with a
ratoon rice crop in CY 77/78. Based on the limited data, it appears that
a ratoon crop would have even been competitive with a full second crop in
the lower lying positions in CY 77/78.
Water source class influenced the performance of second rice crop.
Partially irrigated fields outyielded rainfed fields and because total
variable costs were not different between the classes, higher net returns
were obtained under partial irrigation.
Upland crops. The economic performance of upland crops planted after one
rice crop was affected by landscape position but relationships were in-
consistent (Table 19). Mungbeansreceived the overall highest net returns,
followed distantly by sorghum. The average yields of both mungbean and
sorghum were only moderate. Mungbeanscommand a relatively higher price
in the local market. On the other hand, seed costs for mungbeansare
high compared to sorghum seed costs. Peanut yields were only low to
moderate, but material costs (mainly seeds) were the highest of the four
upland crops tested. Fields suitable for peanut appear limited in the
Economic performance of pattern groups. Four cropping pattern groups
were evaluated: rice-fallow, rice-rice, rice-upland crops and rice-ratoon-
upland crops. These were the resulting actual patternsgrown following
farmer-cooperator's pattern alterations in response to prevailing condi-
tions. The labor and cash components for the different pattern groups
are summarized in Table 20. Labor requirements per hectare were observed
to be the highest in rice-rice pattern (1,222 man-hrs), followed by the
other groups in the order of (1) rice-ratoon-upland crops, (2) rice-upland
crop, and (3) rice-fallow. Cash requirement was highest in the rice-upland
crop pattern, primarily because of the high cost of upland crop seeds and
insecticides. However, among the rice-rice, rice-ratoon-upland crop and
rice-upland crop patterns, combined material costs and labor inputs dif-
fered only slightly. By comparison, the rice-fallow pattern had much lower
material costs and labor inputs, as would be expected.
Within those rice-rice patterns starting with WSR, the WSR-TPR
sequence achieved the highest net return per hectare (Table 21).
Total cash requirement was lowest in WSR-ratoon, causing the return
per unit cash to be highest.
Within the rice-upland pattern group, the rice-mungbean patterns
produced the highest net return, while the lowest net return was obtained
from rice-peanut pattern group (Table 22). The rice-peanut pattern re-
quired the highest labor and material inputs due to land preparation
and expensive peanut seeds. When coupled with low rice and peanut yields.
the rice-peanut pattern ranked lowest in terms of returns above variable
costs, returns to cash and returns to labor. The low rice yields re-
sulted because fields which are good for peanut (light-textured) are poor
for rice. Net returns for the rice-mungbean and rice-sorghum patterns
differed by only 105 in favor of the former. The rice-mungbean pattern
group had a slight advantage over the rice-sorghum pattern for returns-to-
labor but the reverse was true for returns-to-cash, reflecting sorghum's
lower material costs.
Costs and returns of cropping pattern groups are summarized in
Table 23 according to landscape position. Several points are obvious
from the table. Based on net returns, the rice-fallow pattern was
inferior to the other across the sideslope, plateau and plain positions.
The net returns of rice-upland crop patterns varied only slightly across
the three landscape positions and net returns of rice-rice patterns did
not differe by more than 17% from the rice-upland crop pattern. The
rice-ratoon-upland crop patterns produced the highest net returns but
only twoobservations were made. The lowest landscape position (bottom-
land) produced the highest net returns for the rice-rice pattern group.
Finally it should be noted that the yields, and to some extent the inputs
going into the pattern groups in the various landscape positions reflect
the farmer-cooperator's judgement as to what the management strategy for
the second crop should be, if a decision is even made to plant a second
crop given conditions prevailing at the time the decision must be made
and the farmer-cooperator's prognosis for the future. In the discussion
on pattern alterations made by farmer-cooperators, the point was made
that greatest difficulty in growing two rice crops was encountered in
the higher landscapes. Thus the net returns presented in Table 23
for the high landscape units (sideslopes and plateaus) must be considered
as representing the more favorable extremes within those two landscape
classes. This also applies to a lesser degree for the lower landscape
Losses from second crops which fail. Due to an unusually dry November-
through-January period, some second crops which had been planted produced
no grain, i.e., they were complete failures. To examine losses farmers
would likely suffer in the event of a failure, cash and pre-harvest labor
inputs were abstracted from the data to determine how much equity a farmer
would have in a second crop. The upper part of Table 24 shows the average
levels of cash and labor input losses suffered when crops completely failed.
The lower part of Table 24 shows the average levels of cash and labor
inputs when some grain was harvested. As indicated by the consistently
lower material input levels on crops that failed, it is clear that ma-
terial inputs were cut back when crop failure was imminent. The cut-backs
were primarily in insecticide applications. It is also clear that there
was much higher equity in the peanut crops in comparison to other crops, and
peanut crops were therefore more vulnerable to greater cash losses in
the event of failure.
Rice ratoons had by far the lowest equity. Of the full second
crops, sorghum and cowpea had the lowest cash inputs, followed by WSR
and TPR. Species-for-species,upland crop failures all had higher pre-
harvest labor costs than the upland crop non-failures, which reflects
a problem of confounding. Upland crops which failed were often planted
following labor-consuming full tillage operations. The pre-harvest labor
input in TPR crops appeared to be reduced only slightly in failing crops.
On the other hand, pre-harvest labor in WSR crops was reduced by 31%
(perhaps because of the higher weeding and spraying costs in the non-failing
crop). Except for peanut, the average pre-harvest labor put into upland
crops was much less than that going into rice crops.
In the period between late October and mid-January, rainfall is
erratic but the success of rainfed late crops, either rice or upland
crops, is very dependent on the amount and distribution of rain over
this period. Even partially irrigated fields are strongly dependent on
rainfall during this period (1) to supplement often low irrigation water
flows to the fields and (2) to maintain stream-flow from the catchment
area feeding the river from which water is diverted. To gauge the riski-
ness of low rainfall over the second crop period, a wet and dry pentade-1
diagram was constructed following the procedure of Griffiths (1959), but
using a slightly higher set of criteria to distinguish between wet and
dry pentades. A wet pentade is the middle pentade of a group of three
consecutive pentades, in which the total rainfall for the three pentades
either exceeds 53 mm or any two of the three pentades have individual
rainfall totals exceeding 9 mm each. Figure 3 shows that year-to-year
conditions over the 23 years of observations did indeed differ over the
November-January period. Only 3 of the 21 years for which observations
were available was drier during the second crop period than CY 77/78. Only
7 of the 21 years had 7 or more wet pentades out of the 15-pentade
period starting with November 3. Twelve of the 21 had no wet pentades
out of the 6-pentade period starting with December 18. Based on this data
and other analyses (Morris and Zandstra, 1978), it appears that rice as a
second crop will be risky in rainfed Iloilo areas, except in lower positions
which receive delayed through-flow and base-flow from higher elevations.
Even grain legumes appear vulnerable in many years unless planted early.
Given the erratic nature of rainfall during the second crop period, farmers
should be interested in technology which (1) reduces the costs associated
with cultural operations in the early phases of crop production (tillage,
basal fertilization, planting, early pest management), and (2) moves the
harvest period ahead (early varieties, minimum tillage and reduced turn-
1/A pentade is a five-day period.
-A pentade is a five-day period.
Based on the preceding agronomic and economic evaluations of
introduced cropping patterns tested in 1977-78 crop year, relevant
conclusions are as follows:
Landscape position, water source, and soil texture are important
factors that influence both the agronomic and economic performance of
In rainfed areas, two rice crops can be successfully grown
on lower landscape position such as plains and waterways, especially
on heavy-textured soils, while rice-upland crop patterns fit best in
upper landscape position (sideslope and plateau).
Where short duration partial irrigation exists, farmer-cooperators
are able to produce additional rice crops with yields sufficient to make
rice-rice and rice-rice-UC patterns more promising and economically
acceptable than rice-UC patterns, even in years with growing seasons
which are extremely short as was the case in the 1977-78 crop year.
Early establishment technique for the first crop such as dry
seeding (DSR) and turn-around time reductions between crops are
important factors in obtaining successful rice-rice and rice-rice-UC
patterns. Early plantings and short TAT result in higher yields of
the later crops due to reduction of drought stress. Future research
on second crops to be grown under the risky environment should con-
centrate on technology which reduces the costs associated with cultural
operations in the early phases of crop production and moves the harvest
Flexibility of the design of the cropping patterns to allow the
farmer-cooperators to react to unpredictable rainfall is important in
cropping pattern introduction. The choice of the method of establishment
for rice and adaptability of green corn as first crop depends on behavior
of rainfall distribution at the start of the growing season. Upland
crops may do poorly after the rice crop due to early cessation of rains
at the end of the growing season. This causes the previously puddled
soil to dry to a very unfavorable physical environment for the root
system of upland crops.
To take advantage of the domestic market potential, more studies
should be conducted to find methods to grow soybeans successfully in a
post-rice environment, using farmers' resources. Marketing outlets
should be established for sweet potatoes and sorghum. If these limita-
tions are overcome, soybeans, sorghum and sweet potatoes could constitute
major upland crops following lowland rice.
PATTERN RECOMMENDATIONS FOR ILOILO APPLIED RESEARCH TRIALS
As a result cropping pattern tests and trials with associated
component technology conducted between 1975 and the present, recommen-
dations are outlined below based on landscape position, soil texture,
and water source class, after consideration of the long-term average
rainfall. The upland crop to be planted after rice should be selected
on the basis of farmers' present resources, local market limitations and
A. Bottomlands and lower plains.
Cropping pattern: Rice-Rice-Upland crops
Specifications :For medium to light soil of the plain, the
first crop should be dry-seeded. Wet-seeding
should be employed for heavy soils and for
bottomlands. Dry seeding should start from
the third week of April to second week of
May; while wet-seeding should proceed from
the third week of May to third week of June,
depending on rainfall. Under rainfed
conditions, the second rice crop should be
transplanted from a relayed seedbed, while
partially irrigated fields can be either wet-
seeded or transplanted. Cowpea and mungbean
should be established at zero tillage (i.e.,
broadcast relayed to rice) while the soil is
B. Middle and lower sideslopes, plateaus and upper plains.
1. Soil texture:
Heavy (and partially irrigated upper sideslopes,
regardless of texture)
Cropping pattern: Rice-Ratoon-Upland Crops
2. Soil texture:
: Under strictly rainfed conditions, the
first crop should be wet-seeded while
transplanting should be used on partially
irrigated fields. Mungbeans and cowpeas
should be established at zero tillage
(i.e, broadcast-relayed into the ratoon).
Medium to light (and rainfed upper sideslopes,
regardless of texture)
Cropping pattern: Rice-Upland crops
: The first crop should be dry-seeded in a
strictly rainfed environment, while wet-
seeding may be appropriate for partially
irrigated conditions. Mungbeans, cowpeas
and peanuts should be established after
rice with complete tillage levels.
Although upland crop recommendations are limited to mungbeans,
cowpeas and peanuts, other crops as melons, bush sitao, and some fruits
and leaf vegetables may be alternatively used depending on local market
demand and crop adaptation.
These recommendations are summarized in Table 25. Crop manage-
ment practices for individual crops are presented in Tables 26, 27 and 28.
The authors wish to thank the valuable help and contribution
of Drs. V.R. Carangal, J.A. Litsinger, Keith Moody, in the design and
testing phases of cropping patterns and associated component technology.
The cooperation and efforts of the Iloilo research staff, office and
field personnel are gratefully acknowledged.
Griffiths, J.F. 1959. Bioclimatology and the meteorological services. In
Proceedings of the Symposium on Tropical Meteorology in Africa.
Muntilap Foundation and WMO. Nairobi, Kenya. pp 282-300.
International Rice Research Institute. 1974. An agto-climatic classifi-
cation for evaluating cropping systems potentials in Southeast Asian
rice growing regions, IRRI, Los Basos, Laguna, Philippines.
International Rice Research Institute. 1976. Annual Report for 1975.
IRRI, Los Baiios, Laguna, Philippines.
Magbanua, R.D., N.M. Roxas and H.G. Zandstra. 1976. Comparison of the
turn-around period of the different groups of patterns in Iloilo.
A paper presented at the Eight Annual Scientific Meeting of CSSP,
MSAC, La Trinidad, Benguet, May 5-7, 1977.
Magbanua, R.D., N.M. Roxas, M.E. Raymund. and H.G. Zandstra. 1977. Testing
of rainfed lowland rice cropping patterns in Iloilo, crop year 1976-77.
IRRI Saturday Seminar, June 18, 1977.
Magbanua, R.D., R.A. Morris and H.G. Zandstra. 1978. Influence of site
variables on performance of cropping pattern groups in Iloilo. Paper
presented to the 9th Annual Scientific Meeting of the CSSP, Iloilo
City, May 11-13, 1978.
Morris, R.A. and H.G. Zandstra. 1978. Soil and climatic determinants in
relations to cropping patterns. Paper presented at the International
Rice Research Conference. IRRI, Los Baios, Laguna, Philippines,
Morris, R.A., R.D. Magbanua, H.C. Gines and R. L. Tinsley. 1978. The
dynamics of water regimes in Iloilo and Pangasinan land systems.
IRRI Saturday Seminar, March 18, 1978.
Palada, M.C., R. L. Tinsley and R.R. Harwood. 1976. Cropping systems
agronomy program for rainfed lowland rice areas in Iloilo. IRRI
Saturday Seminar, April 24, 1976.
Price, E.C., 1977. Research on the economic of cropping systems at IRRI.
IRRI Annual Review, January 28, 1977.
Table 1. Distribution of the proposed cropping patterns, by landscapes,
for the 1977-78 crop year.
Cropping Number of Fields per Landscape Position
patterns Plain Plateau Sideslope Bottomland Total
DSR-WSR-(M, CP) 4 2 2 8
WSR-TPR-(M, CP) 6 2 2 6 16
WSR-WSR-(M, CP) 4 4
(WSR,TPR)-S 2 3 3 8
(WSR,TPR)-P 2 3 3 8
(WSR,TPR)-M 2 3 3 8
(WSR,TPR)-CP 2 3 3 8
(WSR,TPR)-SP 2 3 3 8
(WSR,TPR)-SB 2 3 3 8
G. Corn-Rice-Upland Crops
GC-(WSR, TPR)-M 2 5 3 10
GC-(WSR, TPR)-CP 2 5 3 10
Total Number of fields 96
Table 2. Summary of the initial crops actually grown. Iloilo 1977-78
Pattern Number of Fields per Landscape Position
component- Bottomland Plain Plateau Sideslope Total
DSR 1 4 2 7
WSR 10 25 22 12 69
TPR 2 4 14 20
TOTALS 10 28 30 28
GRAND TOTAL 96
- wet-seeded rice;
GC= Green corn; DSR = dry-seeded rice; WSR
TPR = transplanted rice.
Table 3. Timing of first crop establishment by landscape
positions. Iloilo 1977-78 crop year.
Landscape Time range Average time
position Week No. Week No.
Bottomland 24 28 25 26
Plain 25 31 27
Plateau 25 31 27
Sideslope 25 32 28 29
Table 4. Yield performance Ct/ha) of rainfed first rice crops (R1), by
landscape positions, Iloilo 1977-78 crop year.
Cropping Landscape position
pattern Bottomland Plain Plateau Sideslope
Ricel-Rice2 6.1(8) 5.5(15) 6.2( 5) 5.2(1)
Ricei-UC 4.6( 9) 4.9(25) 4.1(6)
Overall means 6.1 4.9 5.1 4.6
Overall means 6.1 4.9 5.1 4.6
Table 5. Average flooded days of rainfed first rice crops by landscape
position. Iloilo 1977-78 crop year.
Landscape position Average flooded days-/
Bottomland 78 ( 8)b
Plain 76 (24)
Plateau 76 (30)
Sideslope 61 (26)
A field is considered flooded when there is standing water on
field at least 1 mm.
Number of observations
Table 6. Breakdown of proposed, shifted, unplanted, and actual crop
component of rainfed cropping patterns during the second
crop, by landscape position. Iloilo 1977-78 crop year.
Landscape Number of fields
position Proposed Shifted Unplanted Actual
Rice UO- Rice UC Rice UC Rice UC
Bottomland 8 8 -
Plain 16 9 2UC 4Rice 2 2 16 5
Plateau 15 14 11UC 2Rice 2 3 4 20
Sideslope 9 17 9UC lRice 4 1 21
Total 48 40 22 7 4 9 29 46
a/ Upland crops
Table 7. Mean yields (t/ha) of rainfed second rice crop by establish-
ment method and landscape position. Iloilo 1977-78 crop
Method of Landscape position
establishment Bottomland Plain Mean
No. of failures
No. of failures
-- - - m m lm mm m m -- -- -- -- m m -------- m - -- - -
Mean 2.4 1.4
aNumbers in the parentheses represent the number of observa-
tions in the means including failures. Failures were included as
zero yield in the computation of the means.
Table 8. Average time of dryingA/ of the fields after the last rains
by landscape position and water source class. Iloilo
1977-78 crop year.
Landscape Water Source Class
position Strictly Rainfed Partially irrigated
Bottomland 46 (8) 48 (2)
Plain 45 (15) 49 (3)
Plateau 41 (5)
Sideslope 41 (1) 46 (2)
!/The time of drying was reckoned three days after the fields
lost standing water after the last rains.
Table 9. Productivity of the first and second rice crops across
pattern groups by water source class. Iloilo 1977-78
Water Source Yield (t/ha)
Class Crop 1 Crop 2
Strictly rainfed (1)
No. of failures
Partially irrigated (II)
No. of failures
** Not significant at 5% level by "t" test.
( ) Number of observation.
Table 10. Average turn-around periods in the rice-rice and rice-
upland crop patterns. Iloilo 1977-78 crop year.
Cropping No. of Average turn-
pattern fields around time Difference
WSR 15 19.1
TPR 15 16.3 2.8ns
with tillage 22 43.4
without tillage 25 25.5 17.9**
** Highly significant difference using "t" test.
Comparison between fields with and without dry season
land preparation. Iloilo 1977-78 crop year.
With Dry Season Without Dry Season
Component Land Preparation Land Preparation
Number of fields 28 68
Total tillage labor
requirement (mh/ha) 270 202
(120 + 168)a
Total tillage labor
cost (1/ha) 454 339
Average time of rice
establishment (week no.) 27.3 27.8
Weed weight ,b first rice
crop (gm/m)- 19 21
Yield (t/ha) 4.9 5.2
a/The first figure
season tillage, the second
represents labor requirement for dry
figure for tillage immediately before
Sampled 40-45 DAE/DAT/DAS
Table 12. Performance of rice ratoon. Iloilo 1977-78 crop year.
Farmer Landscape Yield Days to
Cooperator position (kg/ha) Maturity
a/All of the rice ratoon was done in plots under rainfed condition
b/Received 20 kg N/ha as topdressing on ratoon.
Table 13. Performance of upland crops planted after one rice crop by
landscape position. Iloilo 1977-78 crop year.
Landscape No. of Field
Crop position fields (t/ha)
Mungbeans Sideslope 7 0.7 (2)
Plateau 7 0.6 (2)
Plain 2 0.9 (0)
Cowpeas Sideslope 7 1.1 (2)
Plateau 7 0.2 (4)
Plain 3 0.6 (1)
Sorghum Sideslope 5 2.7 (0)
Plateau 4 2.2 (1)
Plain 1 2.8 (0)
Peanuts Sideslope 2 0.5 (1)
Plateau 2 0.9 (1)
( ) Number of failures.
in the computation of means.
The failures were included as zero yield
Summary of performance of different rainfed pattern groups
by landscape position. Iloilo 1977-78 crop year.
Landscape Pattern No. of Crop 1 Crop 2 Crop 3
position group fields (t/ha) (t/ha/%) (%)
Sideslope Rice-Rice 1 5.2 0.8
Rice-UC 21 5.0 48 (4)b/
Rice 4 3.6 -
Plateau Rice-Ratoon-UC 2 7.1 0.9 37 (1)
Rice-Rice 3 5.7 0.5
Rice-UC 20 5.0 31 (9)
Rice 5 4.3 -
Plain Rice-Rice 15 5.5 1.8 (7)
Rice-UC 5 4.6 52 (1)
Rice 4 4.9 -
Bottomland Rice-Rice 8 6.1 2.1 (2)
a/Yields of upland crops are on a relative yield basis (Mungbean -
t/ha = 100%; Cowpea 1.7 t/ha = 100%; Sorghum 8.3 t/ha = 100%
peanut 2.8 t/ha = 100%. These were the highest yield obtained in
of the previous cropping pattern trials and are believed to reflect
maximum performance obtainable by farmer-cooperators.
b/Integers in the parentheses represent the numbers of field that
failed. The failures were included as zero yield in the computation of
:Abl 15. Yield summary of the pattern groups in two water source
classes. Iioilo 1977-78 crop year.
Pattern Water Source No. of Crop Yields (kg/ha)
group Classes fields Crop 1 Crop 2
Rice-Rice I 29 5781 1489 (9)/
II 7 7140 2403 (2)
Rice-UC I 33 4950 60 (39)b/
II 1 2676 31 (0)
Rice-Fallow I 13 4297
A/Number in the parenthesis
b/Relative yield basis (%).
1200 kg/ha = 100%; Sorghum 8326
represent the percentage failures.
Cowpea 1700 = 100%; Mungbean -
kg/ha = 100%; Peanut 2779 kg/ha =
Table 16. Cost and return summary for first crop in cropping pattern
trials, Iloilo, 1977-78 crop year.
Components Method of establishment
DSR WSR TPR
No. of plots 7 (0) 69 (7) 20(1)
Yield (t/ha) 4.7 5.5 4.0
Material cost 591 691 442
Labor cost Ci/ha)
Pre-harvest 333 451 82b
Harvest/thresh 708 818 659
Total variable cost (?/ha) 1632 1962 1927
Total return (V/ha) 4246 5022 4117
Net return over variable cost (W/ha) 2614 3060 2190
( ) No. of plots with partial irrigation
Table 17. Cost and returns
(f/ha) summary of TPR and WSR first rice crop by landscape position. Iloilo 1977-78
Landscape No, of Yield Material Labor cost Total variable Total Return over
position plots (t/ha) cost pre-harvest Harvest/ coat return variable cost
WSR 12 5.2 684 591 790 2065 4740 2675
TPR 14 4.9 439 909 743 2091 4459 2368
Mean 5.0 2510
WSR 22 5.4 712 445 817 1974 4906 2932
TPR 4 3.9 479 703 597 1779 3578 1779
Mean 5.2 2755
WSR 25 5.4 712 434 832 1978 4992 3014
TPR 2 3.1 484 490 467 1441 2798 1357
Mean 5.2 2891
WSR 10 6.2 616 341 947 1904 5683 3779
(g/ha) of second rice crop by landscape position. Iloilo 1977-78 crop year.
Material Pre- Harvest/
cost harvest thresh
( ) no. of plots with partial irrigation
Table 18. Cost and returns
Table 19. Cost and returns (/ha) of upland crops planted after one rice crop by landscape position. Iloilo 1977-78
Landscape No.of Yield Matenial Labor cost Total Total Net
Species position plots (tlha) cost Pre-harvest Harvest variable cost return return
Table 20. Summary of labor and cash components of cropping pattern groups.
Iloilo, 1977-78 crop year.
Variable Rice-fallow Rice-rice Rice-upland upland
cost (13) (34) (47) (2)
Land preparation 255 377 390 362
nursery 211 261 256 322
Fertilizing 11 21 37 21
Pest application 13 22 36 20
Handweeding 28 152 84 157
Others 7 1 7 -
Harvest/thresh 202 388 403 597
Total 727 1222 1212 1479
Seeds 184 371 548 434
Fertilizer 286 563 612 667
Insecticide 53 205 93 11
Herbicide 88 153 131 100
Total 611 1292 1384 1212
SIncludes 4 ratooned rice crops
( ) no. of observations
Table 21. Cost and return summary of WSR-rice patterns by method of
second crop establishment.
Method of second crop establishment
WSR TPR Ratoon
No. of plots
Material cost (Y/ha)
Labor cost (W/ha)
Total variable cost (f/ha)
Total returns (W/ha)
Net returns (W/ha)
Return/unit cash/ (/fl)
/ Return per unit labor = total returns = material cost
Return per unit cash = total returns divided by total
( ) no. of plots with partial irrigation.
Table 22. Cost and returns (W/ha) summary for rice-upland crop pattern groups.
Iloilo 1977-78 crop year.
Components Rice-mungbean Rice-cowpea Rice-sorghum Rice-peanut
No. of plots 16 17 10 4
First crop 4.93 4.94 4.75 2.92
Second crop 0.86 1.07 2.76 0.68
Material cost (W/ha) 1417 1219 1162 2172
Labor cost (1/ha)
Pre-harvest 808 691 676 1742
Harvest/thresh 1091 901 919 690
Total variable cost (V/ha) 3316 2811 2757 4604
Total return (W/ha) 6546 5392 5882 5593
Net return over variable
cost (W/ha) 3230 2581 3125 989
Return per unit labor (V/ha)9.46 9.02 8.62 5.18
Return per unit cash (Y/ha) 4.62 4.42 5.02 2.44
Costs and returns (Y/ha) summary of cropping pattern groups by
lloilo 1977-78 crop year.
Labor cost Total
Landscape Pattern No. of Yield (t/ha) Material Pre- Harvest/ Variable Total Net
position group fields 1st 2nd 3rd cost harvest Thresh cost Returns Returns
10 6.2 2.1 1306 849
1295 3450 7768 4318
Table 24. Average material and pre-harvest labor costs of failed and
non-failed second crops. Iloilo 1977-78 crop year.
No. of Material Pre-harvest Total Variable
Crop fields Cost Labor Cost Cost
A/Percent of average input applied to a non-failed crop.
Table 26. Recommended agronomic practices for cropping patterns. Iloilo 1978-79 crop year.
St Seeding rates/spacing Fertilization
plant population Basal Top dressing
Rice DSR IR36 120 kg/ha (30cm row-spacing 0-30-0/40-0-0-0-/ 40 N-
WSR IR36 120 kg/ha 30-30-0 30 N/30 N
TPR IR36/IR42a 20 x 20 cm (1st crop)
15 x 15 cm (2nd crop) 30-30-0 30 N/30 N
25-30-0 30 N/25 N/
High tillage CES ID-21 50 kg/50 x 10 (2)- 20-40-40-/
Zero tillage CES 55
High tillage EG#2 50 kg/50 x 10 (2)- 20-40-40-
Peanuts CES 101 80 kg/60 x 20 (2)-c/ 45-45-45
-a/IR36 for two rice crop patterns; IR42 for one rice crop patterns and waterways.
Should be employed for one rice crop patterns only.
-/Number in the parentheses represent the number of plants per hill
d/Application of 40 kg N/ha will be delayed 15-20 days after rice emergence.
e/For zero tillage application of basal fertilizer should be done one week before rice harvest.
For high tillage fertilizer will be drilled on furrows prior to seeding.
Top dressing of 40 kg N/ha should be done at panicle initiation only.
-gNitrogen recommendation for IR42 is reduced at 80 kg/ha applied at the following; 25 basal,
30 first top dressing and 25 second top dressing.
Table 25. Summary of pattern recommendations. Iloilo 1978-79 crop year.
Landscape Soils a Cropping
position (15-30 cm) pattern
Bottomland, lower plains
Middle and lower sideslopes
plateaus and upper plains
- Heavy soils include sandy clay, silty clays and clays
Medium soils include sandy loam, silty clay loam, clay loam
Light soils include sandy clay loam, loam, loamy sand, and sandy loam
Establishment technique for rice option: at the left of the slash (/) for
strictly rainfed condition; at the right for partially irrigated fields.
Table 27. Insect pest and disease control recommendations for cropping pattern.
Iloilo 1978-79 crop year.
Crop Recommendations Target pest
Rice first crop (DSR)
First crop (WSR,TPR)
Carbofuran (0.5 kg at
Carbofuran (1.0 kg soil
Maneb seed treatment
Flea, beetle, leafhopper
Table 28. Recommended weed control practices for cropping patterns, Iloilo 1978-79 crop year.
C r o p Weed Control Method Time of Application
Dry-seeded Butachlor fb one handweeding Apply herbicide immediately if
soil is moist. If soil is dry
after germinating rain fb as
Wet-seeded Butachlor2 if standing water 8-10 DAE (or as needed)
at suggested application time;
otherwise, spot weed
Transplanted Spot weeding 21-28 DAT4
Field not plowed Paraquat 0.75 kg a.i./ha Immediately before seeding
Field plowed Interrow cultivation (hilling- 14 DAE for mungbeans
up) fb spot weeding 14-21 DAE for cowpeas
21-28 DAE for peanuts
-/DAE = days after emergence
/1.5 kg a.i./ha for heavy soils; 1.0 kg a.i./ha for light soils
/-fb = followed by
- DAT = days after transplanting
Landscape position Summit Side- Plateau
*,-;..;c.- Loamy texture or
.. non-expanding clay
( I ) Pluvic, (2) Fluxic,
(3) Fluxi-cumulic, (4) Cumulic, ( 5) Cumulo- delugic
Schematic presentation of geomorphic and pedologic conditions
in Iloilo outreach site. (Source: M.E. Raymundo, 1976)
M t 1976-77
Comparison of 1977-78 monthly rainfall with the
1976-77 and the 25-year average.
Aug Sept Oct Nov
31 25 20 14
Figure 3. Wet and dry pentades for crop years 1950-51 to 1969-70 and 1975-76 to
1977-78, Tigbauan, Iloilo. (See text for definition of wet and dry