Centre for Soil Research; JI Ir. H. Juanda 98; Boqor: Indonesia (0251) 23012 Contact: TROPSOILS; Box 02; Sitiung 1A; Sumatera Barat DATE: SEPTEMBER 1986 NUMBER: 33
TITLE: Alley Cropping Experiment 1985/86 Growing Season
EXPERIMENT NO.: 3502
RESEARCHERS: Carl Ev*nsen and Russell Yost
1. To determine Al tolerance of 3 legume tree species under
2. To determine their nitrogen, leaf, and wood production.
3. To measure the effects of green leaf manure on rice and
cowpea yields when intercropped with the trees.
4. To measure effects of these organic matter additions on
soil chemical properties (Al saturation, ECEC, organic C,
etc.) and their interactions with crop growth.
5. To select appropriate legume tree species and liming
levels for farmer testing.
SOIL: Typic Paleudult; clayey, kaolinitic, isohyperthermic.
FCC: Ceak. Cleared by manual tree felling and then bulldozed during 82/83 wet season but never cropped
previously; heavily eroded in parts.
The experiment'is laid out in a split plot design with four
replications. The treatments are as follows:
Main Plots Tree Species
1) Albizia falcataria (grown from seed)
2) Calliandra calothyrsus (grown from seed)
3) Gliricidia sepium (grown from hardwood cuttings)
4) No trees (control)
Subplot Liming Levels
1) No lime
2) tow lime (Ca supplied as a nutrient at 375 kg lime/ha/
year to provide 0.5 cmol Ca / kg soil to 15 cm depth)
3) High liming rate to reduce Al saturation to 25 %
(2 T CaCO3/ha applied first year; about 1/2 T lime/ha
applied second year)
This report covers the crops grown during the 1985/86 rainy season, including upland rice, cowpeas, and the growth of the tree hedges themselves. For a description of the establishment of the tree hedges, see FRB no. 20.
The trees were first pruned on September 17, 1985, nine months after planting. They were sampled to determine leaf fraction, wood fraction, and dry matter content, and the prunings spread in the alleys to dry. After four days, leaves were shaken off branches and wood removed from the plots. Lime was reapplied on the high lime plots to bring acid saturation to 25 % (about 1/2 T lime/ha this season) and at 375 kg lime/ha on the low lime plots (to make a total of 750 kg lime/ha applied to these plots). Also, TSP and KCL were broadcast, each at 50 kg/ha (i.e. 10 kg P and 25 kg K per ha). All these fertilizers and amendments were incorporated with hoes to a depth of about 15cm.
A local variety of upland rice (disease tolerant and
preferred by local farmers) was planted at a spacing of 40 x 15 cm on September 25, 1985. No rows of rice were skipped because of tree hedges (i.e. rows of rice were planted 20 cm to either side of the tree hedges). Seed was planted in dibble holes along with 3 % carbofuran granules (1.3 kg a.i./ha). Germination was rapid and uniform. All tree species had resprouted well by two weeks after pruning. The trees were pruned again on November 26, 1985 to reduce shading of the rice; the prunings were used as mulch between the rows of rice. Diazinon insecticide was sprayed on October 12, December 6, and on January 2, 12, and 17.
Blast and Helminthosporium Brown Spot caused some leaf
damage which looked serious in November, but the rice recovered well and produced a good crop of grain. Heavy rains in January caused-some serious lodging, but stems to be included in the harvested areas were carefully sorted out. Mature panicles were harvested on-February. 3, 1986. The straw and the remainder of the later maturing panicles were harvested on February 13. Harvest areas were 12 m which consisted of 2 m to either side of the center tree hedge. The straw was returned to the plots and buried.
The trees were pruned again on February 18, the prunings
placed in the alleys to dry and drop the leaves, and after four days the wood was removed. TSP was broadcast at 50 kg/ha (10 kg P/ha) and leaves and fertilizer incorporated by hoe. A local variety of semi-determinate cowpea was planted at 40 x 20 cm spacing on March 6, 1986. Germination and growth was very good. On April 12, the tree hedges were pruned and the prunings applied as mulch between the rows of cowpeas. Mature pods were harvested from the 12 m harvest areas on May 5 and vines and the remaining pods harvested on May 15, 1986. The trees were allowed to grow uncut into the dry season.
The yields of leaf, leaf nitrogen, and wood are shown in Tables la, lb, and Ic, respectively. The tree yields were calculated on the basis of total intercropped land area, not just yield per hedgerow area. Clearly Albizia was the most productive tree species in the first several prunings, but by February, Calliandra had become the most vigorous and productive species. Because of the huge amount of leaf and wood produced by the Albizia during early growth, this species produced highest yields during the first year of pruning (from planting in December 1984 to the last pruning of the 1985/86 season in April 1986). Albizia wood yields were especially higher than the other species, however, the wood is soft and (according to local farmers) is not as desirable a fuelwood as is Calliandra. The Gliricidia was far less vigorous and productive. It is probably not as well adapted to the infertile soils in Sitiung, although it may be at a disadvantage in this experiment since it was started from cuttings. The Albizia and Calliandra were grown from seed and therefore have the benefit of strong taproots. Analysis of variance of leaf, leaf N, and wood yields revealed that neither lime nor lime x species interactions caused significant differences in yields. There was however a tendency for the Gliricidia yields to increase with increasing liming levels. Heights of the trees at pruning times are shown in Table 2.
The yields of grain and stover for both rice and dowpea
crops increased significantly with increasing lime rates (Table 3). Rice grain yields increased with both increments of lime, but were not significantly different in response to application of green leaf manures from the different tree species (Table 4). However, there was a tendency for rice yield to decrease in conjunction with more vigorous growth of the tree hedges. The tree hedges did shade the closest rows of-rice, although this did not cause a significant rice yield decrease. Timely pruning of the tree hedges to minimize this shading would seem to be very important. Table 5 shows that cowpea grain yield, unlike rice, increased significantly with only the first increment of lime. Also, the application of Albizia green leaf manure caused significantly greater cowpea grain yields than the other green leaf manure.species or the control (Table 5).
The interaction of green leaf manure species x lime caused significant differences only of rice grain yields (Table 3). This indicates that only in the case of rice grain, the crop response to green leaf manure application differed at the different lime rates. It is especially interesting to note (Table 4) that for rice grain at the zero rate of liming, there is a strong tendency (although not significant) for grain yield to increase with increasing amounts of green leaf manure application. This indicates the possibility of reducing liming requirements by green leaf manure application. Cowpea grain yields (Table 5) were higher in response to Albizia application at all liming levels (although not significantly so at the zero lime level). The possibility of reducing lime requirements by
application of green leaf manure is merely suggested by this data and confirmation will have to be sought in the next season's results, along with information on other possible effects of green leaf manures.
It is also important to note the rather high coefficients of variation of experimental error as shown in Table 3. The soil micro-variability on this site is high, which makes any tests of significance imprecise. This is a constant problem on these newly cleared forest sites and in a low input trial such as this, only very strong treatment effects will be evident.
In Table 6, soil analyses are shown for composite samples from all plots, taken before the start of and during the experiment. There was little difference in exchangeable bases or in acid saturation between different soil depths before the start of the experiment. However, extractable acidity as well as ECEC seemed to increase with depth, possibly due to greater organic matter content in the surface layers. Extractable P was extremely low. Analysis of samples taken on September 10, 1985 (before the second lime application) and on April 12, 1986 (after the second lime application) show that the lime treatments had produced distinct acid saturations of approximately the desired levels. At a later date, the samples will be analysed for organic C.
An important analysis which can't be done on site in Sitiung, is the determination of the breakdown products of decomposition of the green manures and crop residues. The labile organic acids and polyphenolic compounds may be especially important in completing with aluminium, thereby reducing liming requirements. An assessment of these decomposition products will possibly be carried out at the University of Hawaii.
The leguminous tree species, Albizia falcataria-and
Calliandra calothyrsus, show potential for use in alley cropping under the soil and climatic conditions in Sitiung (i.e. acidic soils, low in bases and a warm, -humid climate). The species, Gliricidia sepium, does not grow as vigorously under these conditions. During the 1985/86 growing season, the upland rice crop did not respond significantly to these green leaf manure additions, but cowpea crop yields were increased by addition of Albizia prunings. These results indicate that alley cropping provides only a marginal benefit to farmers during the first year of cropping. However, as the tree pruning and cropping sequence continues on this site, it is quite likely that the frequent additions of green leaf manure will cause increased soil fertility and crop yields. Also, the hedges of trees can provide other benefits, including an animal forage source, fuelwood, and erosion control when planted along contours. These considerations indicate that further study of alley cropping in Sitiung is warranted.
Yields of tree hedgerows during the 1985/86 season.
a. Leaf Yield
--------- Tree Pruning Date ---------- Total
Tree Species Sept.'85 Nov.'85 Feb.'86 Apr.'86 Annual
------------------- kg / ha ------------------Albizia 1489 538 264 220 2511
Calliandra 397 531 590 332 1850
Gliricidia 172 196 90 68 526
LSD (0.05) 335 154 141 87 600
b. Nitrogen Yield in Leaf
--------- Tree Pruning Date ---------- Total
Tree Species Sept.'85 Nov.'85 Feb.'86 Apr.'86 Annual
------------------- kg / ha------------------Albizia 48.5 17.6 8.6 7.2 82.1
Calliandra 11.2 15.0 16.5 9.4 52.3
Gliricidia 6.8 7.7 3.5 2.7 20.7
LSD (0.05) 10.7 5.0 4.4 2.6 18.6
c. Wood Yield
------- Tree Pruning Date --------- Total Fuel
Tree-Species Sept.'85 Nov.'85 Feb.'86 Apr.'86 Annual Wood
Albizia 2685 416 408 101 3610 3093
Calliandra 496 391 460 150 1497 956
Gliricidia 180 111 77 23 391 257
LSD (0.05) 969 120 139 30 1092 1069
Yield in kg/ha is calculated in the basis of total intercropped
land area, not just yield per hedgerow. (To calculate kg
yields per meter of hedgerow, multiply yields/ha x 0.0004).
Fuel wood is the sum of only the Sept.'85 and Feb.'86 harvests,
since the wood was taken off the plots only at these times.
Tree heights at pruning (in cm above ground level).
---------- Tree Pruning Dates ----------Tree Species Sept.'85 Nov.'85 Feb.'86 Apr.'86
------------------ cm------------------Albizia 253 161 173 92
Calliandra 131 143 174 94
Gliricidia 109 94 98 57
LSD (0.05) 71 20 23 9
Table 3. Analysis of
variance for grain and and cowpeas during the
stover yields of 1985/86 season.
---- Upland Rice ---- ------ Cowpea -----Source df Grain MS Stover MS Grain MS Stover MS
Replication Tree Species Error A Lime Rate Lime x Tree Spp. Error B
58478 ns 217301 ns 445646 7394238 *** 443317 155329
558156 ns 274648 ns 474338 9958957 *** 759350 ns 737151
53946 ns 333560 50970 868025 ***
3776 ns 38597
118189 ns 250841 58810 1395855 ** 27049 ns 28799
CV (%) Main-plot 45.3
CV (%) Subplot 26.7
* Significant at 0.05 level
*** Significant at 0.001 level
ns = not significant
Table 4. Effects of lime rates and green leaf manure species on
grain yields of upland rice during the 1985/86 season. Tree Species High lime Low lime Zero lime Species Means
------------------- kg / ha ---------------------Albizia 1804 1219 939 1321
Calliandra 1901 1511 883 1432
Gliricidia 2013 1938 562 1504
No Tree 2349 2113 464 1642
Lime Rate Means 2017 1695 712
LSD Between species means = ns
(0.05) Between lime rates means = 287
Between lime rate means for the same species = 576
Between species for the same or different lime rates = 774
Table 5. Effects of lime rates and green leaf manure species on
grain yields of cowpeas during the 1985/86 season.
Tree Species High lime Low lime Zero lime Species Means
------------------- kg / ha---------------------Albizia 921 849 427 732
Calliandra 652 510 213 458
Gliricidia 556 459 138 384
No Tree 536 478 121 209
Lime Rate Means 666 574 225
LSD Between species means = 209
(0.05) Between lime rates means = 143
Between lime rate means for the same species = 287
Between species for the same or different lime rates = 313
Table 6. Soil analyses at the start of and during the experiment. Sampling Sample AI+Hd Cad Mgd Ke ECEC pe % Acid
Time Type ---------- cmol / kg ----------- ppm Sat.
09/1/84a 0-15cm 2.21 0.76 0.28 0.07 3.32 0.9 66
15-30cm 1.93 0.77 0.25 0.05 3.01 1.6 63
30-60cm 1.53 0.85 0.25 0.04 2.67 0 57
09/10/85b High lime 1.12 2.19 0.07 0.09 3.47 N/A 33
Low lime 1.89 0.75 0.11 0.08 2.83 N/A 68
Zero lime 2.21 0.38 0.06 0.09 2.73 N/A 81
04/12/86c High lime 0.74 1.66 0.12 0.21 2.74 5.8 27
Low lime 1.33 0.83 0.16 0.20 2.52 5.8 54
Zero lime 1.71 0.27 0.10 0.21 2.29 6.3 75
a Before start of experiment
Before second lime application (Sept. 23, 1986) and planting. All samples 0-15 cm. C After second lime application. All samples 0-15 cm.
Extracted with IN KC1.
e Extracted with Mehlich I (double acid) extractant.