Group Title: TropSoils field research brief ;, 30
Title: Lime residual and maintenance; with special "blocking" design
CITATION THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00080606/00001
 Material Information
Title: Lime residual and maintenance; with special "blocking" design
Series Title: TropSoils field research brief ;, 30
Physical Description: 3, 2 leaves : ill. ; 28 cm.
Language: English
Creator: Wade, M. K. (Michael Karl)
Gedjer, I. P.
Wigena, Putu.
Lembaga Penelitian Tanah.
Soil Management Collaborative Research Support Program.
Publisher: Soil Management Collaborative Research Support Program, North Carolina State University,
Publication Date: 1986?
 Subjects
Subject: Liming of soils -- Indonesia.
Soil management -- Indonesia.
Spatial Coverage: Indonesia.
 Notes
General Note: Caption title.
General Note: At head of title: TropSoils-Indonesia, Centre for Soil Research.
 Record Information
Bibliographic ID: UF00080606
Volume ID: VID00001
Source Institution: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 156977663

Full Text


TROPSOILS-Indonesia
Centre for Soil Research
Field Research Brief No. 30


TITLE: Lime Residual and Maintenance; with special "blocking"
design

EXPERIMENT NO.: 1704

RESEARCHERS: Mike Wade, I P. Gedjer, Putu Wigena

SOIL: Typic Haplorthox/LCeak surface layer-sandy clay loam
subsoil-sandy clay


Introduction

The TROPSOILS team working in Sumatra has repeatedly
experienced extreme soi variability in research plots. Such
variability has also plagued other research projects, e.g. SARIF,
IFDC and IPB. The variability tends to not be directional or
following a gradient, rather it is random. Therefore blocking as
usually practiced is often ineffective in dealing with the soil
variability.
A research effort by CSR in Jambi (Kuamang Euning
transmigration area) has been initiated. Because of suspected high
soil variability at the Kuamang Kuning experimental location, it
was decided to conduct a uniformity trial at a pt-ential site
in order to quantify the variability by crop performance on the
unamended soil. Concurrently with this uniformi:y test crop, the
many-mini plots (FRB No. 29) were established to give a
better idea of the expected fertility responses a= this Kuamang
Kuning location.

Blocking

Methodology

Two blocks of 24 plots were located in the field. Plot size
was 4x6 m. The area had been cleared of forces= two years ago,
but left fallow since that time. The primary vegetation was
grass and shrubs. All vegetation was removed anc the area tilled
by hand-hoeing. Soybeans were planted after land preparation was
completed. No treatments were made, only crop protection from
weeds and pests were provided during the growing season.

Results

The plots and their "score" (average pod set/plant from 10
plants within each plot) are shown in Figure 1. As can be seen
there was rather extreme variability, and that variability tends
to be random about the area. There is no apparent fertility
gradient in any one direction, or in any regular pattern, in
either block.






Conclusions


Based on the results of another trial (FR3 !;o.29), a ime
experiment has bee designed with a rotation of soybean and maize.
The experimental design will be a randomized complete block,
except that plots to be included in a given replication will be
selected based on their soybean score rather than on their geo-
graphical location. The purpose of blocking is to divide the
experimental area, as much as possible, into blocks based on
known or suspected variability. The analysis of variance then
can take out the variance due to that known source of vari-
ability. Because the variability here has no pa:=ern or
gradient, it is impossible to group the known variability (soy-
bean scores) by geographical blocking. Therefore we will "block"
the experiment by individually selecting plots for a given repli-
cation based on the soybean scores. The 48 plots then are
divided into 4 replications, with 12 treatments each. The scores
were ranked from lowest to highest and divided into four groups,
each group becoming a replication regardless of the physical
location of the individual plots. The resulting plot allocation
is shown in Figure 2. This layout now has the lock of a complete-
ly randomized design, but it is not because the replications were
not randomly assigned, rather "blocked" according to their inher-
ent soil fertility. This methodology may seem unconventional,
but it is rational and it has accomplished in a r.ore precise
manner what is normally attempted to achieve when "blocking" a
field experiment by sight and without prior uniformity data.

Lime Exteriment

Treatments

The following treatments were assigned to the twelve plots
of each replication:
Lime Rates ( 'ha)
Trt. No. Year 1 Year 2 Year 3

1. 0 0 0
2. 0.5 0 0
3. 1.0 0 0
4. 1.5 0 0
5. 0 0.5 0
6. 0 1.0 0
7. 0 0 0.5
8. 0 0 1.0
9. 0.5 0.1 0.2
10. 0.5 0.2 0.2
11. 1.0 0.2 0.2
12. 1.0 0.4 0.4

The current (first) crop is soybean. Table 1 gives plant
height at 30 days by lime treatment and replication. Because
several treatments have the same lime rate the first year, the
data have been simplified by combining similar treatments i.e. 0
rate is the mean of treatment no. 1, 5, 6, 7 and 8; 0.5 rate is






the mean of no. 2, 9, and 10; 1.0 rate is the mean of 3, 11, and
12.

There is a consistent response to lime on all replications,
especially to the first rate. Both treatment and replication are
highly significant, and the CV is very low (by our standards).
Because of the manner of selecting plots to be included in a
given replication, the replications are progressively more fer-
tile. That is why there is a significant reokucatuib effect.
Because replications are a psuedo-fertility treatment, it might
be expected to have a treatment by replication interaction.
This, of course, cannot be tested statistically since that is the
error term itself. However a visual inspection of the data
indicate that there is a response in each replication, including
the most fertile (rep IV).

These initial results indicate that the blocking method used
has minimized variability and may be an improves method of laying
out plots when the soil exhibits extreme and random variability.


Table 1.


Effect of Lime Rates
replication).


on Soybean Heigh- at 30 Days (by


Lime Rate Replication
t/ha II III I X

0.0 29 31 38 39 34
0.5 33 39 42 44 39
1.0 33 40 41 47 40
1.5 44 40 44 51 45

X 35 38 41 45

CV=6.2%





Co m V 5,


FAgre. I. P(t l youi 0S Say.ear, Sco es


I 2 3 '-' 7 $

2. 3 S 8 '8 ii 2( z2z

9 8 1/ 6 9 8 3

1 7 S32. 0 (/ z I I t




o ? 5- z II

6 ( 6 ,9 3 ? A/ J











SfT T 3E iL jZ IZ
i Z l 3 -( I 9 q

q ? 3 11 8' ? 1)1

? 0o ( 10 I 9 6 ]o


_ 7- 2 3 7 -

2 37 : 2 IT r "r II

' o Z U I l) 3
I -~IIi


uitve 2.


euo ro7 -ssgvimeu:, I IA,.


" bc/aw a r fnid1 c0 I-rAma
*ew j-,i /ai k yr/[i'/Cd7s a--i /2 iPtS 4.




University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs