Analysis of the cassava yield variation at Cambodia- Thailand border

Abstract

Cassava (Manihot esculenta Crantz) is one of the most important upland crops in Cambodia. However, there is still a shortage of information and research about its yield variations and the causal factors, which is important to know if our goal is to increase the cassava production in Cambodia. The objectives of this study were to highlight the yield variation and causal factors in agro-practices of cassava production in Battambang and Pailin provinces. In these two provinces, 109 cassava farmers were randomly selected from a list of households provided by each village chief. The results showed that the cassava yield ranged from 10 to 34 tons per hectare with the mean yield ranging from 29.41% to 82.35% of the maximum yield, while the corresponding yield gap ranged from 6 to 24 tons per hectare. The variables of the agro-practices, such as the weed control and the usage of herbicides and liquid fertilizers are the significant factors of the cassava yield. They account for 83% of the total yield variability among the sample farmers. The main constraints to cassava production are drought, weed density, pests, and diseases. The farmers had, in general, a low education; 41% of them had only a primary school education diploma. So, it was difficult for them to understand information on new techniques, the weather factors, the pest mitigation, and appropriate agro-practices. The knowledge of all those factors could improve the yield and narrow the cassava yield gap. Therefore, it is undeniable that the yield gap varies substantially in Cambodia.

Full text

Asian Journal of Agricultural and Environmental Safety, 2020 (1): 17–27
ISSN: 2575423, https://www.ajaes.org
Asian Journal of Agricultural and Environmental Safety
Vol. 2020, No. 1
17
AJAES
Analysis of the cassava yield variation at Cambodia-
Thailand border
Vibol PEUO1*, Songsak MIMGRATOK1, Thaworn CHIMLIANG1, Yagura KENJIRO2, Sutisa CHAIKUL1,
Phirum PEUO3
1 Faculty of Agricultural Technology, Rambhai Barni Rajabhat University, Chanthaburi 22000, Thailand;
2 Faculty of Economics, Hannan University, Osaka 580-8502, Japan;
3 Department of Rural Development, Battambang 02200, Cambodia.
*Corresponding author: e-mail: pouevibol@yahoo.com
Abstract
Cassava (Manihot esculenta Crantz) is one of the most important upland crops in Cambodia. However, there is still a
shortage of information and research about its yield variations and the causal factors, which is important to know if
our goal is to increase the cassava production in Cambodia. The objectives of this study were to highlight the yield
variation and causal factors in agro-practices of cassava production in Battambang and Pailin provinces. In these two
provinces, 109 cassava farmers were randomly selected from a list of households provided by each village chief. The
results showed that the cassava yield ranged from 10 to 34 tons per hectare with the mean yield ranging from 29.41%
to 82.35% of the maximum yield, while the corresponding yield gap ranged from 6 to 24 tons per hectare. The variables
of the agro-practices, such as the weed control and the usage of herbicides and liquid fertilizers are the significant
factors of the cassava yield. They account for 83% of the total yield variability among the sample farmers. The main
constraints to cassava production are drought, weed density, pests, and diseases. The farmers had, in general, a low
education; 41% of them had only a primary school education diploma. So, it was difficult for them to understand
information on new techniques, the weather factors, the pest mitigation, and appropriate agro-practices. The
knowledge of all those factors could improve the yield and narrow the cassava yield gap. Therefore, it is undeniable
that the yield gap varies substantially in Cambodia.
Keywords: Agro-practices, analysis, Cambodia-Thailand, cassava, yield variation.
Introduction
Besides rice, which is the major agricultural product in
Cambodia, the farmers are also attracted to cassava
production. Cassava has now become an important cash
crop for resource-poor farmers in Cambodia (Sopheap et
al. 2008). For the last 5 years, the cassava production area
in Cambodia has expanded significantly from less than
515 thousand hectares in 2014 to more than 650 thousand
hectares in 2018 (MAFF 2017). Cassava is an essential
tropical crop and can be suitable to grow with an average
temperature of 25-29 0C (Onwueme and Sinha 1999) and
annual rainfall greater than 500 mm (MAFF 2015). In
2014, the average yield of cassava in Cambodia was 25
tons per hectare, which was the second-highest yield after
Laos in Southeast Asia (FAOSTAT 2017). The increase
in production has come from the expansion of the
planting areas, although the average yield had decreased
to 22.55 tons per hectare in 2017 (AFSIS 2017). In an
optimal growing environment, the yield of cassava could
reach 90 tons per hectare (El-Sharkawy 2004). However,
the cassava experiment under Cambodian conditions
obtained the highest yield at 36 tons per hectare (Sopheap
et al. 2008).
Improving crop management practices including
high-yielding varieties, good quality planting materials,
sufficient moisture, proper plant spacing, and pest and
disease management are needed to close the cassava yield
gap. However, farmers often based their practice more on
myths instead of facts, like thinking that cassava does not
need fertilizers or pesticides, so they grow cassava with
minimal or even no fertilizer application. Another study
conducted in Thailand showed that the yield decline is
due to continuous cassava production in the same
unfertilized plot over an 8-year period, which is similar
to the result of other experiments in Southeast Asia
(Howeler and Cadavid 1983). A study by Sopheap et al
(2012) in Cambodia showed a large yield variation
ranging from 12.7 to 37.2 tons per hectare, with the main
constraints in increasing the yield being the soil nutrient
deficit, short crop duration, and the weed competition.

2020 Peuo et al.
Asian Journal of Agricultural and Environmental Safety
Vol. 2020, No. 1
18
Surveys of farming practices, supplemented by
measurements of soil properties and crop performance,
have the potential to provide a valuable means of
assessing yield constraints in farmers’ fields (Calvino
and Sandras 2002; Inthavong et al. 2011). Neumann et al
(2010) pointed out that minimizing the yield gap requires
an understanding of nature and the constraints of the
regions. Therefore, separated assessment is needed to
increase the yield in a specific region. At the same time,
information on yield variations, coefficients, and causal
factors are indispensable for increasing the cassava
production in Cambodia.
The objective of the study is to highlight the yield
variation and causal factors in relation to the aspects of
agro-practices of the cassava production: varieties,
fertilizers, herbicides, pesticides, weed control, and
knowledge (Education and farmer experience). The study
is also an importance tool that highlights the prevalence
of a yield gap; that information is not readily available in
many regions of Cambodia, due to a lack of data in the
agro-practices of cassava production.
Methods
Study area
This study was conducted in two provinces of Cambodia,
Battambang and Pailin which are located near the
Cambodia-Thailand border. They have more than 50,000
hectares of cassava fields per province and are among the
top 10 provinces that planted the most cassava in
Cambodia, according to the Ministry of Forestry and
Fisheries, Cambodia (MAFF 2017). The study area is
situated in one particular agro-ecological zone which has
many enterprises for the storage, processing, and export
of cassava, mainly to Thailand. The major source of
household income in the study area is from cassava,
maize, mungbean, soybean, and fruit trees. These
provinces both have a long history of cassava production
and are currently the largest cassava growing areas in
Cambodia.
Sampling method
A multi-stage sampling method developed by Yamane
formula (created in 1967 and 1973) was used to select
109 cassava farmer samples with a sampling fraction of
at least 10% of the population of cassava farming
(standard error of 10%). These respondents were chosen
randomly from a list collected from the village chiefs
across 3 districts of those 2 provinces as shown in Table
1 and Figure 1. The cassava farmers who collaborated
with this study had to indicate a willingness to be
interviewed and allow a study of their crop. The first
survey was conducted between February and March 2019
in Battambang province and the second one was from
June to July 2019 in Pailin province.
Fig. 1: Study area in Battambang and Pailin provinces, Cambodia.

2020 Peuo et al.
Asian Journal of Agricultural and Environmental Safety
Vol. 2020, No. 1
19
Table 1. Number of household samples in each target village.
Provinces
Districts
Communes
Villages
Households
Households
Cultivated
Selection
Battambang
Sompovlun
Serei Mean Chey
OuKandal
455
308
28
PhnomPhrek
PhnomPhrek
PhnomPhrek
901
350
32
Pailin
Salakrav
Salakrav
Phnomkuy
386
295
27
OuAndoung
Ouchetbram
290
241
22
Total
1,637
1,194
109
Data collection
Both qualitative and quantitative methods of data
analysis were applied in this study to understand the
actions of cassava farming. Primary data was collected to
get information about cassava production, the agro-
practices, the varieties, pests and diseases, and the history
of cassava production. This was achieved by using both
semi-structured and guided questionnaire interviews with
the members of the households who were part of sample.
Secondary data gathered information about temperature,
humidity level, rainfall, and any other data relevant to the
study.
Data analysis
The data collected from the questionnaire surveys were
coded and analyzed using SPSS Version 20 statistical
program to generate cross-tabulation of variables and
calculate descriptive statistics. The correlation and
regression analysis were also conducted to examine how
agro-practices significantly affect cassava yield and
whether the farmer’s knowledge level, represented by
education level and farm experience, has an influence on
the cassava yield. Microsoft Excel was used for
frequencies, charts, and tables to show visible findings of
agro-practices and causal factors.
Results
Education of head households
The level of education is one of the factors that affect
agricultural productivity (Asadullah and Rahman 2005).
Likewise, the major finding in a previous study showed
that as the level of education increases, the productivity
increases as well. The farmers with a secondary school
education got the highest returns on agricultural
productivity (Oduro et al. 2014). The results of our
findings, as in Figure 2, showed that a large number of
cassava farmers in our samples have a low level of
education: 41.29% of them have a primary school
education and 16.5% are illiterate. Only 3% of them
completed college education.
Fig. 2: Education level of head of households’ cassava farmers in both
provinces.
Characteristics of the Agro-practices
Land use
The sample farmers have an average farming area of 7.63
hectares, which is divided into 58.03% of cassava
cultivation, 30% of maize, 6.11% of rice, and 5.86% of
fruit trees such as mango, longan, and cashew nuts (See
the detail in Fig. 3). This indicates that cassava is the most
important crop. It provides a major income to farmers in
this study area.
Of the land planted with cassava, 43.66% was
previously used to plant maize. Most farmers grew
cassava as a mono-crop. Only a few farmers were
growing cassava as a mixed crop. Our findings showed
that only 28.77% of cassava is processed locally and the
rest is sent to Thai traders who export the cassava to the
Chinese market. Therefore, the crop prices seem to
influence the usage of the land (crop change from maize
to cassava) in the study area.
At the same time, 63.3% of the farmers in this study
area had more than 5 years of experience in cassava
cultivation and they practice a crop rotation between
maize and cassava from year to year. The crop rotation is
a good way to improve soil properties such as soil
aggregate, soil fertilizer, and help to minimize pests and
diseases (Yuniwati et al. 2020). These processes
contribute to the development of soil structure (Ball et al.
2005).
16.51
41.29
28.44
11.01
2.75
0
5
10
15
20
25
30
35
40
45
Illiterate
Prima…
Seco…
High…
College
Percentage (%)

2020 Peuo et al.
Asian Journal of Agricultural and Environmental Safety
Vol. 2020, No. 1
20
Fig. 3: Land used and crop rotation for cassava farming in the both
provinces.
Cassava varieties
In Battambang province, the most popular variety is
Rayong 9 while variety 89 is the most popular in Pailin
province. More than 43.12% of the farmers planted
Rayong 9 variety and 17.43% planted the variety 89.
Moreover, 8.26% of the farmers planted an unknown
variety as shown in Figure 4. Among these cassava
varieties, Rayong 9, Huay Bong 60, and KU 50 are from
Thailand. Kromumyun variety is from Vietnam, while
the 89 and KorlTorl varieties are from unknown sources,
but the owner of a silo mentioned that those varieties
come from Thailand. Rayong 9, KorlTorl, Huay Bong 60,
and KU 50 have high yield potential and high starch
content and variety 89 has a very high yield potential.
Fig. 4: Varieties of cassava in the study area.
Fertilizer usage
In the study areas, there is great misunderstanding among
the farmers. In some cases, the information about cassava
production is based more on myths than on facts. Some
of those farmers often grow cassava with minimal or no
fertilizer at all. Also, they would apply fertilizer, not for
the cassava plant, but to improve the soil property only.
They apply NPK (15-15-15) fertilizers or Bio-fertilizer
only one time or 50 kg per hectare when they raise beds
before planting stems.
Table 2 showed that 58.33% of the cassava farmers in
Battambang province used fertilizer in their farming,
while in Pailin province, that figure is only 28.57%.
Likewise, liquid fertilizer (called Chy Tuek or Hormone
in the local language) is widely used. Cassava farmers
who were interviewed believe that this application
improves the cassava yield. One month after planting, the
farmers always spray liquid fertilizer on the cassava
leaves for the first time. The second time of spraying is
conducted 2 to 4 weeks after the first time. The rainfall
will have an influence on the frequency of spraying. In
Battambang 85% of the farmers sprayed the liquid
fertilizer while in Palin, only 4.08% of the farmers did
not spray as shown in Table 2. The majority of the
cassava farmers used the liquid fertilizers (85% -
95.92%) and the dry fertilizers (58.33% - 28.57%).
Continuous cropping, recycling, and reusing of
nutrients from organic sources may not be sufficient to
sustain crop yields. A study on the effect of fertilizer
application on continuous cropping of cassava from 2004
to 2007 in Indonesia revealed that without fertilizer
application, cassava yield decreased from more than 20
tons per ha in the first year to less than 10 tons per ha in
the third year, after which the yield remained constant at
about 9 tons per ha (Yuniwati et al. 2012).
Weed control
Weed control is a very important factor that can improve
the yield. A very good weed control could increase the
yield by 7 to 8% according to a previous study (Clair et
al. 2000). In the study area, the cassava farmers conduct
two types of weed control. For the first method, the weed
is removed by hand when the cassava is 1 month old and
then when trimming is done at 7-8 months old or 2-3
months before harvesting. The second method is by using
herbicides. Table 3 shows that the "hand weeding"
control in Battambang province was done once and it
represented 53.33% of their work time for the cassava
farming. In Pailin province, the" hand weeding" was
done 1 to 3 times and it represented 96% of their work
time. So, we can understand the reason why the cassava
farmers in these 2 provinces prefer to use herbicides for
weed control.
The herbicides used are produced in Thailand: 48%
of Glyphosates and 28% of Paraquate which are mostly
used in the study area to control the weeds. However,
because the study area is close to the Thai border, the
cassava farmers mostly bought herbicides or chemical
pesticides from Thailand.
6.11 5.86
30.00
58.03
43.66
9.05 2.67 2.64
0
10
20
30
40
50
60
70
Rice
Fruit trees
Maize
Cassava
Maize
Soybean
Mungbea
n
Sesame
Percentage (%)
43.21%
17.43%
11.01%
8.26%
6.42%
5.50%
8.26% Rayong 9
89
KorlTorl
HuayBong 60
KU 50
Kromomyun
Others

2020 Peuo et al.
Asian Journal of Agricultural and Environmental Safety
Vol. 2020, No. 1
21
Table 2: Both types of fertilizer application of household samples in the study area.
Both Types of fertilizer application
Battambang
Pailin
Frequency
Percentage
Frequency
Percentage
Dry fertilizer
Not used
25
41.67
35
71.43
Bio-fertilizer
13
21.67
11
22.45
Chemical
17
28.33
1
2.04
Both
5
8.33
2
4.08
Total
60
100
49
100
Liquid fertilizer
Not used
9
15.00
2
4.08
1 Time
42
70.00
46
93.88
2 Times
9
15.00
1
2.04
Total
60
100
49
100
Table 3: The ways of controlling weed by cassava farmers in study area.
Ways of controlling weed
Battambang
Pailin
Frequency
Percentage
Frequency
Percentage
By hand
Not used
4
6.67
0
0
1 Time
35
58.33
15
30.61
2 Times
6
10.00
13
26.53
3 Times
13
21.67
19
38.78
4 Times
2
3.33
2
4.08
Total
60
100
49
100
By herbicide
2 Times
9
15.00
11
22.45
3 Times
20
33.33
36
73.47
4 Times
18
30.00
2
4.08
5 Times
9
15.00
0
0
6 Times
4
6.67
0
0
Total
60
100
49
100
Pesticide usage
Figure 5 show that 50% of cassava farmers in
Battambang have used insecticides while only 24.48%
did in Pailin. The majority of invasive insect species are
mites and mealybugs in the study area. This situation
agrees with the study of MAFF (2015) and Ignazio et al
(2016), which shows that the main pests in South East
Asia are the cassava mealybugs, cassava mites, and
whiteflies.
In Battambang province, the highest percentage of the
farmers who used insecticides 2 times is 21.66% while in
Pailin province; it is 10.20% of the farmers who have
used insecticides 1 and 2 times (Fig. 5).

2020 Peuo et al.
Asian Journal of Agricultural and Environmental Safety
Vol. 2020, No. 1
22
Fig. 5: Frequency and percentage of pesticides usage for cassava farmers.
Cassava yield gap
In the study area, the yield of fresh root cassava showed
a great variation, ranging from 10 to 34 tons per hectare,
with an average of 22 tons per hectare. The yield gap
between the maximum and the average yield was 12 tons
per hectare, while the gap between the average and the
minimum yield was 12 tons per hectare. The fields with
a low to moderate level of yield accounted for 63% of the
fields surveyed in this study. Therefore a total gap
between the maximum and minimum yield is 24 tons per
hectare. These data are similar to the study of Sopheap et
al (2012) which showed a large difference in the yield of
cassava in Kampong Cham province where the highest
yield was 37.26 tons per hectare and the lowest yield was
only 12.8 tons per hectare.
In the study area of Battambang and Pailin provinces,
the group with the highest frequency experienced
moderately low yields with frequency declining towards
both the higher and the lower ends. However, the yield
groups of moderately high, moderate, and moderately
low represented the smallest variations among groups
(Table 4). The mean yield of the moderately high to the
low yield group ranged from 82.35% to 29.41% of the
highest yield recorded, while the corresponding yield gap
ranged from 6 to 24 tons per hectare. The maximum yield
of 34 tons per hectare was considered to be representative
of the maximum potential farm yield.
Figure 6 indicates the difference of yield in those
communes. For example, in PhnomPhrek the percentage
of “moderately low ” and “low’’ yields is large, while
in Salakrav, the percentage of “High” and “Moderately
high”is also large. However, the percentage of average
yield per hectare is low for SereiMeanChey PhnomPhrek
commune in Battambang Province and OuAndoung
Salakrav in Pailin Province. The low average yields in
these communes seem to be caused by drought, weed
competition, pests, disease, and inappropriate agro-
practice of the cassava production.
Factors affecting cassava yield
To find the factors affecting cassava yield, we calculated
the correlation coefficient between cassava yield per
hectare and the variables representing the farmer’s
attributes and his agro-practice. As shown in Table 5, the
frequency of liquid fertilizer application has a high and
significant correlation with the cassava yield (r=0.60,
P<0.01), the usage of herbicides represented a negligible
correlation (r=0.27, P<0.01), and the weed controlling
has the highest and significant correlation (r=0.90,
P<0.01). So the agro-practices might influence the
cassava yield in the study area, but we cannot call it an
unequivocal proof.
Table 4: Distribution of yield group and yield gap for fresh root cassava.
Yield group
Mean
(ton/ha)
Range (ton/ha)
No. of
fields
% of maximum yield
Yield gap range
(ton/ha)
High
34
> 30.99
18
100.00
-
Moderately high
28
26.00-30.99
22
82.35
6
Moderate
22
20.00-25.99
25
64.70
12
Moderately low
17
14.00-19.99
26
50.00
17
Low
10
<14.00
18
29.41
24
7
13 10
30
552
37
0
5
10
15
20
25
30
35
40
1 Time 2 Times 3 Times 1 Time 2 Times 3 Times
Used Not used Used Not used
Battambang Pailin
0
10
20
30
40
50
60
70
80
Frequency
%
Frequency Percentage
11.67
21.66
50
11.67 10.20 10.20 4.08
75.51

2020 Peuo et al.
Asian Journal of Agricultural and Environmental Safety
Vol. 2020, No. 1
23
Fig. 6: Percentage of mean yields per hectare in each commune.
Table 5: Correlations between yield and agro-practice factors influencing the yield.
Variables
1
2
3
4
5
6
7
8
9
10
1Yield/H
1
2Farm size
-.002
1
3Farm
Experience
-.104
-.030
1
4Education
.031
.263**
.080
1
5Chemical
.045
.055
-.013
-.011
1
6Bio
Fertilizer
-.163
.035
-.023
.004
-.280**
1
7Liquid
Fertilizers
.600**
.162
-.174
.148
-.096
-.020
1
8Pesticides
-.027
.089
.007
-.013
.301**
.123
-.210*
1
9Herbicides
.275**
.016
-.041
-.089
.026
.031
.056
.378**
1
10Weed
Controlling
.908**
.047
-.134
.059
.048
-.159
.588**
-.017
.177
1
**: Correlation is significant at the 0.01 level (2-tailed). *: Correlation is significant at the 0.05 level (2-tailed).
The planting season is a more appropriate term to use
here. It was not included in the computation of
correlation because the data did not allow a meaningful
analysis, as most of the farmers planted cassava from
February to March and only a small number of farmers
planted in May. Planting interval is also not included in
the analysis because all the sample farmers planted 3-4
stems per meter. So, these factors were not the cause of
the yield variation in the present study.
To control the effect of confounding factors, a
regression analysis was also conducted. We first
estimated a model using all the independent variables,
and then we removed the variables with VIF values larger
than 10 in models 1 and 2 to avoid multicollinearity. The
variables that we removed are all those indicating the
cassava variety such as (Rayong 9, 89, KorTorl,
HauyBoung 60, and unknown).
Model 1: The analysis was carried out to determine if
cassava is affected by the knowledge of the farmers
(education level and farm experience) as well as the agro-
practices such as varieties, agro-chemical usage and
weed control (See Table 6, Model 1). The results
indicated that weed control, herbicides, and liquid
fertilizers were significant factors affecting the yield.
Among these agro-practices, weed control by hand has
the largest positive effect on cassava yield. If cassava
farmers increase the frequency of hand weed control by
one, the cassava yield will increase by 6.6 tons per
hectare. This is a factor of most importance for the agro-
practice in cassava farming, in the study area.
The second factor affecting yield is the usage of liquid
fertilizers. If the farmers increase the frequency of
spraying liquid fertilizers by one, the yield will increase
by 2.35 tons per hectare. For the herbicides, there will be
an increase of 1.19 tons per hectare if the farmers increase
the frequency of application by one. All variables
together accounted for 83% of the total yield variability
(R2 = 0.83) as shown in Table 6.
0
5
10
15
20
Low Moderately low Moderate Moderately high High
Percentage (%)
Serei MeanChey PhnomPhewk Salakrav OuAndoung

2020 Peuo et al.
Asian Journal of Agricultural and Environmental Safety
Vol. 2020, No. 1
24
Table 6: The coefficients analysis of the determinants of cassava yield.
Variable
Model 1
Model 2
Model 3
Definition
Coefficients
t-value
Coefficients
t-value
Coefficients
t-value
(Constant)
6.885
5.326
7.281
6.990
22.061
10.689
-
Farm Size
-.128
-1.353
-.136
-1.511
-.004
-.018
Planted area of cassava (ha)
Farm Experience
.043
.662
-
-
-.169
-1.071
Years of cassava farming
1Gender
.195
.192
-
-
3.786
1.549
Gender of head household
2Education
-.075
-.215
-
-
.610
.717
Level of Education
KU50
-1.192
-.887
-1.284
-.973
-
-
Dummy variable
(Types of variety)
Kromomyun
-.230
-.145
-.287
-.184
-
-
Chemical Fertilizers
.631
.637
.630
.644
-
-
Categorical variable
(1= Used, 0= Don’t use)
Bio-Fertilizers
-.345
-.441
-.346
-.450
-
-
Liquid Fertilizers
2.359**
2.350
2.265**
2.321
-
-
Liquid fertilizers usage (Frequency)
Pesticides
-.681
-.844
-.696
-.874
-
-
Pest controlling by pesticides
(Frequency)
Herbicides
1.190 **
3.174
1.208**
3.308
-
-
Weed controlling by herbicides
(Frequency)
Weed Controlling
6.609**
15.959
6.603**
16.194
-
-
Weed controlling by hand (Frequency)
Adjusted R2
0.83
0.84
-0.002
-
**. Statistically significant at 1% level; Sample (N) = 109 cassava farmers. 1(Male=1, Female=2); 2(Score: 1= Illiterate, 2= Primary school, 3= Secondary
school, 4= High school, 5= College.
Model 2: Includes only the variables representing
agro-practices and farm size. The data showed that the
adjusted R2 does not change much from the model 1 (If
the knowledge of the farmers affects cassava yield by a
large degree, the adjusted R2 of model 2 must be much
smaller than that of model 1) this means that agro-
practices are the most important factors to determine the
cassava yield, while the attributes of farmers including
their knowledge does not have a significant influence on
cassava yield.
Model 3: The knowledge of the farmers might affect
agro-practices that they adopt and hence can have an
indirect effect on cassava yield. To examine this
possibility, Model 3, which includes only the farm size
and the attributes of the farmers, was also estimated.
But the estimation result showed that the variables such
as education level and farming experience do not have
significant association with cassava yield. This result
implies that education and farming experience do not
have a significant correlation with agro-practices that
affect cassava yield (Table 6).
Constraints of cassava farmers before planting
According to data analysis, only 49.54% of the farmers
had no problem with the previous planting, while 50.46%
have encountered some problems as shown in the left pie
chart of Figure 7. Problems mentioned by the farmers
include lack of rain (15.60%), a lack of cut stem (9.17%),
the high cost of the cut stem (4.59%), lack of capital
(8.26%), and lack of labor (12.84%), as shown in the
right pie chart of Figure 7.
Fig. 7: Representation of the percentage of constraints before cassava
planting.
Constraints of cassava farmers during cassava growing
Forty-six percent (45.87%) of the respondents answered
that they faced problems during the growing period; the
other 54.13% said that they had encountered some
problems as shown in the left pie chart in Figure 8.
Those problems, as shown in the right pie chart of
Figure 8 were associated with pests and diseases
(16.52%), weed competition (10.09%), increased
drought (12.84%), lack of capital (5.50%) and a lack of
labor (9.17%).
Hence, pests and diseases represented around 16%
and a big percentage of problems in the study area. The
farmers showed the invasive pests such as mealybugs,
cassava mites and Cassava witches broom (CWB)
diseases in their plantation. According to another study,
CWB has affected 64% of the fields in several prime
49.54%
15.60%
9.17%
4.59%
8.26%
12.84%
50.46%
No problem Lack of rain
Lack of cut stem High cost of cut stem
Lack of capital Lack of labor

2020 Peuo et al.
Asian Journal of Agricultural and Environmental Safety
Vol. 2020, No. 1
25
cassava-growing areas, and was especially problematic
in Cambodia where 78% of the cassava fields were
impacted by CWB (Ignazio et al. 2016).
Fig. 8: Representation of the percentage of constraints during cassava
farming.
Discussion
The farmers had, in general, a low education; 41% of
them had only a primary school education diploma. So, it
was difficult for them to understand information on new
techniques, the weather factors, the pest mitigation, and
appropriate agro-practices. The knowledge of all those
factors could improve the yield and narrow the cassava
yield gap. A major effect of education on agriculture is
the cognitive effect, whereby a farmer acquires basic
literacy and numeric ability to read the instructions on
fertilizer, pesticides, and herbicides containers and to
calculate the number of multiple inputs correctly to
enhance productivity (Appleton and Balihuta 1996).
The farmers use different cassava varieties because
they want to achieve higher yields. According to our
interview, they don't know which variety is suitable for
their specific agro-ecological condition. Our survey also
revealed that for planting a new crop, most farmers used
cassava stems which they bought or kept from their
previous cultivation, for their convenience.
Unfortunately, this technique provides an easy way for
the diseases to spread to the next season especially when
cassava stems affected by viruses are kept. Presently, no
cassava breeding program has been either established or
carried out in Cambodia and only the testing of some
varieties from cassava breeding centers of Thailand,
Vietnam, and China have been done in Cambodia
(MAFF 2015). It is undeniable that the cassava farmers
have a difficulty to find or to get healthy and high-quality
planting material.
The rainfall conditions and good management in
agro-practices have effects on the yield of cassava
production in the study area. The study Luar et al (2018)
revealed that the optimal nutrient management is the key
to closing wide yield gaps and to attain a sustainable
intensification in the cassava production. Continuous
cropping of cassava without balancing the fertilizer
application can lead to soil nutrient depletion and yield
decline over time. Our findings showed that cassava
farmers did not use fertilizer based on cassava needs. The
concept of the 4R Nutrient Stewardship is a framework
for promoting the right application of nutrients sources
(or products) at the right rate, right time, and right place
(IFA 2009). 4R Nutrient Stewardship Framework as a
means of linking science to practice, and supporting
effective communications with all stakeholders
(Johnston and Bruulsema 2014). Thus, the judicious use
of chemical fertilizers is essential to maintain soil
fertility. Fertilizer usage is closely associated with the
growth phases of cassava. Apply N, P, and K fertilizer 2
to 4 weeks after planting to ensure that the crop has
enough nutrients to support its early growth (IPNI 2012).
On the soils that are moderately deficient in P and K, a
general recommendation is to use a fertilizer with an N:
P: K ratio of roughly 1:1:2, e.g. 40-80 kg N, 40-80 kg P,
and 80 -160 kg K per hectare (MAFF 2015).
Based on our interviews, the cassava farmers often
used fertilizers, pesticides and herbicides based on their
experience and the suggestion of other cassava farmers
or agrochemical sellers rather than the manufacturer’s
recommendations or agricultural experts, and the farmers
would often mix pesticide with chemical fertilizer spay.
If the farmers followed the experts’ recommendations,
they would benefit from it and would not need to spend
extra money and time with herbicide spraying.
Conclusions
The results of this study have shown substantial
disparities in cassava yields at the Cambodia-Thai border
region. The maximum yield obtained, 34 tons per
hectare, was considered to be representative of the
maximum potential yield under the rainfall conditions in
this region with proper agro-practice. However, large
gaps relative to the maximum yield were found for most
fields, and fields with a low to moderate level of yield
accounted for 63% of the fields surveyed in this study.
The regression analysis revealed that weed control by
hand, application of liquid fertilizers and herbicides
significantly affect the cassava yield. Another possible
constraint to higher yield is the usage of cassava stems
from a previous crop to plant their new crop, because of
the risk of carrying diseases from a previous cassava
generation. The knowledge of all those factors could
improve the yield and narrow the cassava yield gap.
Climate change (drought or rainfall) would also affect
cassava yield, as many of the farmers surveyed
mentioned the climate as a problem in their cassava
cultivation. But this study could not examine whether or
not and to what extent the climatic factors affect the
cassava production. Further research is needed to
investigate the effects of the climate factor.
Acknowledgements
We would like to offer particular thanks to Her Royal
Highness Princess Maha Chakri Siridhorn scholarship for
45.87%
8.26%
8.26%
10.09%
12.84%
5.50%
9.17%
54.13%
No problem Pests Diseases
Weeds Increased drought Lack of capital
Lack of labor

2020 Peuo et al.
Asian Journal of Agricultural and Environmental Safety
Vol. 2020, No. 1
26
supporting this study. We are grateful to the farmers, and
authorities of all levels that actively collaborated in this
study and taught us about the multiple roles of cassava
within their farms. We are also indebted to Dr.Sarawut
Saengsotchot, Assist. Prof. Dr.Yardrung Suwannarat for
advising and supporting the field survey in this study.
This study resulted from a grant from the Rambhai Barni
Rajabhat University Thailand and the University of
Battambang, Cambodia, from the support of students and
coordinators with authority for field surveys. Special
thanks are also extended to Mr. Luc Payant, a Canadian
retired nurse, volunteer at Peaceful Children Home in
Cambodia, for his kind help in the English editing of this
manuscript.
References
AFSIS. 2017. Report on ASEAN Agricultural
Commodity Outlook. Eleased by AFSIS Secretariat
Office of Agricultural Economics (OAE) Ministry of
Agriculture and Cooperatives Bangkok, Thailand.
http://www.aptfsis.org.
Appleton S. and Balihuta A. 1996. Education and
Agricultural Productivity Evidence from Uganda.
Journal of International Development, 8 (3): 415-
444.
Asadullah M. N. and Rahman S. 2005. Farm productivity
and efficiency in rural Bangladesh: The role of
education revisited. Applied Economics, Taylor and
Francis Journals, 41(1): 17-33.
Ball B. C., Bingham I., Rees R. M., Watson C. A. and
Litterick A. 2005. The role of crop rotations in
determining soil structure and crop growth
conditions. Canada journal of Soil Science, 85(5):
557-577.
Calvino P. and Sandras V. 2002. On-farm assessment of
constraints to wheat yield in the south-eastern
Pampas. Field Crops Research, 74: 1-11.
El-Sharkawy M.A. 2004. Cassava biology and
physiology. Plant Molecular Biology, 56(4): 481-
501.
FAOSTAT. 2017. FAOSTAT (database), Food and
Agriculture Organization of the United Nations
Statistics Division, Rome, http://faostat3.fao.org/
home/E.
Howeler R.H. and Cadavid L.F. 1983. Accumulation and
distribution of dry matter and nutrients during a 12-
months growth cycle of cassava. Field Crops
Research, 7: 123-139.
IFA. 2009. The global “4R” nutrient stewardship
framework-developing fertilizer best management
practices for delivering economic, social and
environmental benefits, International Fertilizer
Industry Association, Paris, France.
Ignazio G., Nami M., Elizabeth A., Dung T. N., Trinh X.
H., Tin M. A., Juan M.P., Prapit W. and Kris AG.W.
2016. A comprehensive evaluation of geographic
priorities, management options and research needs:
Emerging pests and diseases of South-east Asian
cassava. Society of Chemical Industry, DOI:
10.1002/ps.4250.
Inthavong S., Fukai S. and Tsubo M. 2011. Spatial
variations in water availability, soil fertility and grain
yield in rainfed T lowland rice: A case study from
Savannakhet province, Lao PDR. Plant Production
and Sciences, 14(2): 184-195.
IPNI. 2012. 4R Plant Nutrition Manual. A Manual for
Improving the Management of Plant Nutrition,
Bruulsema TW et al., (ed.). Peachtree Corners, GA,
USA.
Johnston A.M., and Bruulsema T.W. 2014. 4R Nutrient
Stewardship for Improved Nutrient Use Efficiency.
“SYMPHOS 2013”, 2nd International Symposium on
Innovation and Technology in the Phosphate Industry
Procedia Engineering, 83: 365-370.
Luar L., Pampolino M., Ocampo A., Valdez A., Cordora
D. F. and Oberthür T. 2018. Cassava Response to
Fertilizer Application, Southeast Asia. Better Crops
with Plant Food, 102 (2): 11-13.
MAFF. 2017. Annual Report of Agriculture, Forestry and
Fisheries 2016-2017 and Directions 2017-2018.
MAFF Conference, April 2017.
MAFF. 2015. Cassava Handbook. Phnom Penh,
Cambodia. China-Cambodia-UNDP Trilateral
Cooperation Cassava Project Phase II, 2015.
Neumann K., Verberg P.H., Stehfest E. and Muller C.
2010. A yield gap of global grain production: A
spatial analysis. Agricultural System, 103(5): 316-
326.
Oduro O.E., Aboagye A.P. and Acquaye N.A.E. 2014.
Effects of education on the agricultural productivity
of farmers in the OFFINSO municipality.
International Journal of Development Research,
4(9): 1951-1960.
Onwueme C. and Sinha T.D. 1999. Field Crop
Production in Tropical Africa. CTA, Wageningen,
Netheflands pp401-411.
Sopheap U., Patanothaib A. and Aye T.M. 2012.
Unerveiling constraints to cassava production in
Cambodia: An nalysis from farmers’ yield variations.
International Journal of plant Production, 6(4): 409-
428.
Sopheap U., Makara O., Howeler R.H. and Aye T.M.
2008. Enhancing cassava production and utilization
through Nippon Foundation project in Cambodia, in:
Howeler, R.H. (eds.), A New Future for Cassava in
Asia: Its Use as Food and Fuel to Benefit the Poor.
Proceeding of 8th Regional Cassava Workshop. 20-24
Oct. 2008, Vientiane, Lao PDR. CIAT, Cali,
Colombia, pp. 460-469.
Yamane T. 1967. Statistics: An Introductory Analysis, 1st
Edition, New York: Harper and Row.
Yamane T. 1973. Statistics: an introduction analysis, 2nd
Edition, New York: Harper and Row
Yuniwati E.D., Basuki N., Wisnubroto E.W. and Utomo
W.H. 2012. Combating land degradation in cassava

2020 Peuo et al.
Asian Journal of Agricultural and Environmental Safety
Vol. 2020, No. 1
27
field by crop yield improvement. Asian Journal of
Crop Science, 2: 4975-4982.
Yuniwati E.D., Wani Hadi Utomo W.H. and Wisnubroto
E.I. 2020. Effect of Biochar Application on the
Growth of Cassava and Maize, Maize Yield and Soil
Erosion in Cassava+Maize Intercropping System.
Asian Journal of Crop Science, 12 (2): 109-114.