Livestock Research for Rural Development 13 (6) 2001

Citation of this paper

Comparison of biodigester effluent and urea as fertilizer for water spinach vegetable

Kean Sophea and T R Preston

University of Tropical Agriculture, Royal University of Agriculture
Chamcar Daung, Dangkor District, Phnom Penh, Cambodia
kean.sophea@undp.org

 

Abstract

Two experiments were carried out from 22 July to 21 October 2001 in the University of Tropical Agriculture farm on the campus of the Royal University of Agriculture, Phnom Penh. The main objective was to measure the effect of different sources of fertilizer and the response in yield of water spinach (Ipomoea aquatica, var. reptans) to increasing levels of effluent from a biodigester charged with pig manure. In both studies the first fertilizer application was made one week after planting and the rest at weekly interval for 3 weeks. The total period from planting to harvest was 4 weeks.  

The first experiment was conducted to evaluate the effect of different fertilizing practices on water spinach (Ipomoea aquatica, var. reptans) yield. The crop was located on a sandy, poor soil derived from alluvial deposits (pH 5.45, N 0.13%). A completely randomized block design with four treatments was employed: no fertilization (control), 75 kg N/ha as urea, 75 kg total N/ha as biodigester effluent and 75 kg ammonia-N/ha as biodigester effluent. There was no difference in fresh biomass yield of water spinach between the two treatments with biodigester effluent (17.6 and 18.6 tonnes/ha, for total-N and ammonia-N, respectively), which were higher than the control (5.6 tonnes/ha) and tended to be higher than when the N source was from urea (15.5 tonnes/ha). 

In the second experiment the yield of water spinach was used as response criterion to different levels of N (0, 20, 40, 60, 80, 100, 120 and 140 kg N/ha) as effluent from a biodigester charged with pig manure. The fresh biomass yield was linearly related with the level of effluent N  (Y = 7.12 + 0.118X, R2= 0.96) reaching 23.6 tonnes/ha with140 kg N/ha. The yield response in the second harvest (re-growth), when the same levels of effluent N were applied, was much less reaching a maximum of 16 tonnes/ha of fresh biomass with 140 kg N/ha, and with a more variable response (Y = 7.03 + 0.0502X, R2= 0.64). 

It was concluded that: on the basis of total N content, biodigester effluent had a similar value as urea for fertilization of water spinach; and the yield response to effluent was linear over the range of 0 to 140 kg N/ha. In the first growth period, with 140 kg N/ha of effluent, 54% of the applied N was converted to N in the water spinach.

Key words: Biodigesters, effluent, urea, ammonia, nitrogen, fertilizer, water spinach, Ipomoea aquatica 
 

Introduction 

In Cambodia about 50% of all children aged 0-5 years are either stunted or underweight, which appears to be due to long term chronic under-nutrition rather than wasting from short-term, severe food shortages (UNDP 1997). Micronutrient deficiencies in the diet result in improper physical and mental development in children as well as adults, which is responsible for lower productivity and inferior quality of life. Leafy vegetables can contribute significant amounts of vitamins and minerals, and are especially excellent sources of protein, carotene (vitamin A), iron and ascorbic acid (vitamin C). The vegetable consumption in Cambodia is reported to be one of the lowest in the world, being only 35 kg per person per annum (DOA 1994). This is equivalent to only 30% of the recommended daily intake of vegetables (300g/day according to WHO). Therefore, a “ home garden “ for growing of vegetables is one way to improve the quality of life of the Cambodian people by increasing nutritional status and generating income as well. 

Cambodia’s economy is largely based on agriculture. Approximately 80% of the population live in rural areas and grow rice as their staple crop. Vegetable production is of secondary importance and is mainly in lowland areas, especially in the Mekong delta region where the soil is very fertile. The farm size in Cambodia is generally small especially concerning the vegetable garden and the soil is very poor. So the intensive cultivation and improved productivity per unit area is a key factor for rural economic and social development. 

The vegetable production in Cambodia is mainly dependent on external inputs in the form of  seeds, pesticides and chemical fertilizers. Most farmers cannot afford the chemical inputs and this leads to very low yields. Vegetables require many nutrient elements for good growth and production but N, P and K are the three elements of most concern. Leafy vegetables are especially heavy users of nitrogen.  

Animal manure is a potential replacement for chemical fertilizer and is traditionally used by poor farmers in Cambodia. However, it is not properly managed so that the efficiency of utilization of the manure is very low. The introduction of low-cost biodigesters in Southeast Asia (Bui Xuan An et al 1997) has made it possible for small-scale farmers to convert manure into biogas and a nutrient rich effluent. Le Ha Chau (1998a,b) showed that cassava and duckweed yielded more biomass of a higher protein content when effluent rather than manure was used as the fertilizer. 

Water spinach is used traditionally in Cambodia as a vegetable for consumption by people and animals.  It has a short growth period and is resistant to common insect pests. However, there appears to be no information in the literature on its response to fertilizer especially fertilizer of organic origin as is produced by the anaerobic digestion of livestock manure. 

The following experiments were carried out to evaluate the potential of water spinach to utilize the effluent from biodigesters charged with pig manure. The first experiment aimed to compare biodigester effluent with urea, which is the fertilizer most commonly used by vegetable farmers in Cambodia.  The second experiment was designed to derive the response curve of biomass yield to increasing levels of biodigester effluent as the only fertilizer. 

Hypotheses:


Materials and methods

Experiment 1: Comparison of biodigester effluent and urea
Location

The experiment was done in the  “ Ecological “ farm of the University of Tropical Agriculture situated in the campus of the Royal University of Agriculture, Chamcar Daung, Dangkor District, about 10 km from Phnom Penh. The experimental area was located on sandy soil derived from alluvial deposits. The pH was 5.45 and the soil had 0.13% of nitrogen (UTA  2001, unpublished data).

Design and treatments

The treatments were:

C:  control (no fertilizer)
N: Effluent from a biodigester charged with pig manure at rate of 75 kg N/ha
NH3: Effluent from a biodigester charged with pig manure applied at 75 kg ammonia-N/ha
Urea: Urea applied at 75 kg N/ha 

There were 8 replicates of each treatment arranged within two blocks in a completely randomized block design. The plot size was 1.5*1.00 m with a space of 50 cm between plots in each block, which were 10 m apart.

Land preparation

The soil was cultivated by hoe, two times and a raised bed prepared, which was 15 cm high.

Planting

Seeds of dry-land water spinach were planted on 22 July 2001, in rows across the bed, at spacing between seeds of 1-2 cm and at 2-3 cm depth. The distance between rows was 20cm. The distance between plots and rows was 50 cm.

Fertilizing

The fertilizers were applied 3 times in the growing period. Increasing quantities equivalent to 10, 40 and 50% of the total allowance were applied on days 7, 14 and 21, respectively (Table 1). For each application fresh quantities of effluent were brought from a biodigester charged with washings (manure, urine and water) from pigs fed broken rice and water spinach (Ipomoea aquatica). The effluent was analysed for total nitrogen (N) and ammonia-N (NH3-N). The quantities of effluent were calculated on the basis of the total amount of N (75 kg/ha) and the proportion of the total (10, 40 or 50%) to be applied.  

Table 1:  Quantities of effluent and urea applied per plot of 1.5 m2 during the experiment

Days

%  of total

Effluent (N)
Litres

Effluent (NH3-N)
Litres

Urea, g

7

10

4.89

5.02

2.45

14

40

13.6

11.4

9.78

21

50

16.1

26.8

12.2

Total

100

34.6

43.2

24.5

Irrigation

A watering can was used to apply water twice a day (morning and afternoon) at the rate of 3 to 4 litres/m). On rainy days no water was applied.

Measurements

Plant height was measured and number of leaves counted weekly before applying the fertilizers. The water spinach was harvested 28 days after planting. All plants in individual plots were weighed. Leaf and stem were separated and analysed immediately after harvest to determine dry matter by microwave radiation (Undersander et al 1993) and nitrogen (Foss Tecator kjeldahl apparatus; AOAC 1990). Samples of soil were taken from each plot before planting and after harvest for analysis of dry matter, nitrogen, ash, organic matter and pH.

Maize biotest

Before planting and after harvesting the water spinach, two samples of soil (2 kg) were taken from the 0 – 20 cm layer in each plot and put into plastic bags. Three seeds of maize were planted in each plastic bag. The bags were watered every day. One week after planting the number of maize plants was reduced to one per bag for the growing test. After 30 days the height to the growing point was recorded and the biomass harvested and weighed fresh.

Statistical analysis

The data were analysed by ANOVA using the General Linear Model (GML) software of Minitab (Version Release 12.21). The variables were fertilizer treatments, blocks and error. 

Experiment 2: Response of water spinach to fertilization with increasing levels of biodigester effluent
Location

The experiment was carried out in the same location and with the same plots used for Experiment 1.

Treatments and design

The design was a production function with 8 levels of fertilization with biodigester effluent, equivalent to N levels of 0, 20, 40, 60, 80, 100, 120, 140 kg/ha.  Two harvests were made at intervals of 28 days and the fertilizer treatments were repeated in each of the two consecutive growth periods There were 4 replicates of each treatment. The plots used for each replicate were those used previously for each of the four fertilizer treatments applied in Experiment 1.

Land preparation

The plots were cultivated by hoe two times before re-making the raised beds. There was an interval of 7 days from the time of the harvest in Experiment 1 and planting the seed in Experiment 2, which was done on 25 August 2001.

Planting, fertilization, irrigation and measurements

The procedures were similar to those described in Experiment 1, except that no analyses were done on the soil.

Harvesting

Two harvests were made at intervals of 28 days.


Results and discussion

Experiment 1
Composition of the effluent

The original hypothesis was that the water spinach would respond more to the ammonia nitrogen present in the effluent rather than to the total nitrogen a proportion of which would be in organic form and thus less available to a rapidly growing plant such as water spinach.  Unfortunately there appeared to be considerable variation in the proportion of ammonia in the effluent applied on each occasion, with values ranging from 0.49 to 0.99 (Table 2). 

Table 2: Composition of the effluent and urea applied on the three occasions during the growth of the water spinach in the first growth period

Days

Effluent,
 mg N/litre

Effluent,
mg NH3/litre

Effluent, mg NH3-N/litre

Effluent,
NH3-N/N

Urea,
%N

7

230

224

184

0.802

46

14

330

395

325

0.986

46

21

350

210

173

0.494

46

The digested waste (effluent) is a high quality fertilizer. The digestion process converts a proportion of the nitrogen in the organic materials to ionic ammonia (NH4+), a form that becomes more stable when ploughed into the soil. The ammonia ion is readily “fixed” in the soil so that it can be absorbed by plants. However, there is little information on the proportion of the nitrogen that is converted to the ionized form. It is not known if the very high value recorded in the effluent applied on day 14 was a true value or the result of analytical error. Recent analyses of total nitrogen and ammonia nitrogen in the manure fed into the biodigester and the corresponding effluent indicated proportions of ammonia nitrogen of 0.11 and 0.06, respectively in fresh manure from cattle and pigs increasing to 0.62 and 0.67 in the effluent (Pok Samkol and Chan Noeng 2001, unpublished data). As the application of the effluent in the N and NH3-N treatment was based on the reported analyses (Table 2), the amounts of effluent applied on day 21 were much greater for the NH3-N treatment (26.8 litres) than for the N treatment (16.1 litres) with corresponding total amounts applied during the experiment being 43.2 and 34.6 litres (Table 1).

 Soil fertility changes

There were no differences among fertilizer treatments in the soil pH (P=0.28) or in the organic matter content (P=0.43) after harvesting of the water spinach (Tables 3 and 4 and Figures 1 and 2).  However, on all fertilizer treatments the soil pH was almost one unit higher after growing the water spinach (P=0.001).  By contrast, the organic matter content decreased when no fertilizer was applied and increased on the urea treatment. There were no changes for the two effluent treatments. 

Table 3:  Least square means for pH of the soil before and at the end of the first growth period for the control, “N” , “NH3” and urea fertilizer treatments.

 

Control

N

NH3

Urea

Mean

SEM/Prob

Before

4.64

4.73

4.80

4.49

4.67

0.039/0.001

After

5.52

5.41

5.42

5.40

5.44

Mean

5.08

5.07

5.11

4.94

 

 

SEM/Prob

0.055/0.17

 

 

 

  

Table 4:  Least square means for organic matter of the soil before and at the end of the experiment for the control, “N” , “NH3” and urea fertilizer treatments

 

Control

N

NH3

Urea

Mean

SEM/Prob

Before

2.76

1.97

2.15

1.92

2.20

0.184/0.852

After

2.23

2.04

2.19

2.54

2.25

Mean

2.49

2.01

2.17

2.23

 

 

SEM/Prob

0.26/0.62

 

 

  

Figure 1. Effect of different sources of fertilizer on soil pH before and after growing of water spinach


F
igure 2. Effect of different sources of fertilizer on soil organic matter before and after growing of water spinach

 There was reasonable uniformity among treatment plots in overall soil fertility before starting the experiment, as measured by the maize test (P=0.311) but a tendency for it to be less on the urea treatment (Tables 4 and 5; Figure 3).  On all treatments there was an apparent decrease (P=0.091) in soil fertility after growing the water spinach, however, as the maize growth is subject to the prevailing climatic conditions during the test, caution is needed in interpreting differences between tests done at different times.  The use of sand as a reference medium, as demonstrated by Chamnanwit Promkot (2001), would have given more validity to the interpretation of differences between tests at different times.

Figure 3. Effect of different sources of fertilizer on growth rates of maize
(soil biotest) before and after planting water spinach

 

Table 5:  Least square means for the biotest (weight of maize plant, g)  of the soil before and at the end of the experiment

 

C

N

NH3

Urea

Mean

SEM/Prob

Before

16.7

13.4

20.1

18.3

17.1

1.59/0.311

After

8.91

10.6

11.9

7.83

9.83

Mean

12.8

12.0

16.0

13.0

 

 

SEM/Prob

1.27/0.001

 

 

 Water spinach yield

There were no differences between the two levels of effluent (based on total N and ammonia-N) in the effects on the height and green biomass yield of the water spinach. Both these treatments tended to support higher yields than urea alone, which in turn was superior to the control (Table 6; Figures 4 and 5). 

Table 6:  Least square means for height and yield of water spinach in plots of 1.5m according to fertilizer treatment

 

Control

N

NH3

Urea

SEM/Prob.

Height, cm

19.7c

34.5a

34.8a

30.7b

0.71/0.001

Biomass yield, g fresh weight per plot

Green material

6.58c

17.6ab

18.6a

15.5b

1.23/0.001

Roots

1.50b

2.24a

2.16a

2.12a

0.14/0.001

abc Means within rows without letter in common are different (P< 0.05)

Figure 4. Effect of different sources of fertilizer on height of water spinach after 28 days growth

Figure 5. Effect of different sources of fertilizer on green biomass yield of water spinach at 28 days

The plots receiving the ammonia-N treatment received 25% more effluent than those where the effluent  was applied according to the total N content, yet the yields differed by only 6%. This implies that the total N content is as good a guide to the fertiliser value of biodigester effluent as is the content of N in the form of NH4+ and that there was little benefit from the extra 25% of N on the “NH3” treatment.  There is no obvious explanation for the latter result,  other than that the uptake of N was less when the highest  quantities were applied on day 21 of the growth period. That effluent should be superior to urea is not surprising as the effluent supplies a variety of other nutrients, including phosphorus, potassium and other minor elements all of which are needed for plant growth. In an experiment in southern Minnesota, comparing pig manure and nitrate-N fertilizer, maize yields were greater for the pig manure than for the commercial fertilizer at six of the seven sites (Randall 1998). Cushman et al (1999) compared commercial nitrate fertilizer and swine effluent for production of vegetable crops in Mississippi and showed that all treatments receiving swine effluent always yielded equal to or greater than treatments that received commercial fertilizers. These results also showed that differences in the predominating nitrogen form, that is nitrate nitrogen (NO3) from commercial fertilizer sources or ammonia-N  from the effluent source, did not affect yield or quality of the vegetables (Cushman et al 1999). 

The yield on the best treatment was equivalent to 18.6 tonnes fresh biomass/ha in one month of growth.  On a yearly basis, assuming irrigation is available, this is equivalent to 223 tonnes fresh biomass/ha, which with a dry matter content of 8% is 18 tonnes of dry matter. This indicates there is a very high potential for the use of water spinach as an appropriate crop to exploit the use of biodigester effluent in integrated farming systems. 

Experiment 2
First harvest

There was an interaction (P=0.001) between growth periods and effects of the level of effluent  in the first harvest (Figures 6 and 7). The water spinach had a very similar height on all effluent levels following the first application of effluent, which was 20% of the total amount. At the second measurement on day 14, differences due to effluent level became apparent, and these were magnified considerably by day 21 (Table 7). 

Figure 6. Effect of effluent level on height of water spinach

Figure 7. Effect of level of effluent on number of leaves of water spinach.

There were no differences among levels of effluent in the effects on the number of leaves (Table 8), but there were big differences between periods (P=0.001). 

Table 7:  Least square means for height (cm) of water spinach in plots of 1.5m according to level of effluent N

Growth
 stage

Level of effluent, kg N/ha

SEM/Prob

0

20

40

60

80

100

120

140

14 days

8.07

8.01

6.70

7.74

8.52

9.02

8.40

11.1

0.37/0.001

21 days

15.3

18.15

16.4

18.9

21.3

23.5

22.8

26.4

0.66/0.001

28 days

23.9

31.0

30.24

37.3

39.3

45.4

42.6

48.8

1.18/0.001

 

Table 8:  Least square means for numbers of leaves of water spinach in plots of 1.5m according to level of effluent

Periods

Level of effluent kg N/ha

SEM/Prob

0

20

40

60

80

100

120

140

14 days

5.90

5.95

5.50

5.65

5.97

5.87

5.95

6.17

0.14/0.001

21 days

8.95

9.27

8.47

8.82

9.42

9.42

9.05

9.80

0.19/0.001

28 days

11.3

12.0

11.5

12.0

12.4

12.9

12.1

12.8

0.22/0.001

  

Figure 8. Effect of level of effluent on dry matter and crude protein content
of whole water spinach (stem and leaves).

There were tendencies for the dry matter content to decrease slightly (Y = 9.16 – 0.0497X0.0026) and for crude protein to increase (Y = 15.6 + 0.0492X0.016) as the effluent level was increased (Figure 8).

 

Figure 9. Effect of effluent on green biomass yield of water spinach (first harvest after planting 28 days) 

The response of biomass yield to effluent level was linear over the range of 0 to 140 kg N/ha (Figure 9; Table 9). 

Table 9:  Least square means for fresh biomass yield of water spinach according to level of effluent (first harvest)

 

Level of effluent kg N/ha

SEM/Prob

0

20

40

60

80

100

120

140

Biomass,  tonnes/ha

6.66

10.5

10.6

14.1

17.4

20.7

20.0

23.6

1.44/0.001

The fact that the yield response was linear up to the maximum level applied (140 kg N/ha) implies that there might have been responses to even higher levels of application of effluent N. Although climatic conditions and soils are very different, it is interesting to compare the levels of N used in the present study with the recommendations for leafy vegetables, typically recommended in the UK (MAFF 1994) (eg: 300 kg N/ha for sprouts, 250 kg N/ha for cauliflowers and 200 kg N/ha for lettuce).

Re-growth period

Biomass yields were lower for the second harvest (Figure 10; Table 10) compared with the first harvest (Figure 9; Table 9) and the response was less consistent, with apparently lowest yields for the 40 and 60 kg N/ha levels. The variation may have been caused by the exceptionally heavy rain which fell during the re-growth phase. The yield of fresh biomass with 140 kg N/ha was 16 tonnes/ha compared with  24 tonnes/ha  in the first harvest. 

Table 10:  Least square means for fresh biomass yield of water spinach according to level of effluent (second harvest)

 

Level of effluent, kg N/ha

SEM/Prob

0

20

40

60

80

100

120

140

Fresh biomass, tonnes/ha

9.25

9.16

7.41

7.00

11.0

11.0

13.2

16.3

1.92/0.04

 

Figure 10. Effect of different levels of effluent Non biomass yield
 of water spinach (second harvest)
 

The dry matter content of the leaf tended to be higher than in the stem but there was no apparent effect of the level of effluent (Figure 11).

Figure 11.  Effect of different levels of effluent N on the dry matter content
of the leaves and stem of water spinach (second harvest)
 

The protein content of the leaf of water spinach was higher than in the stem but, as in the case of the dry matter content, there was no relationship with the level of effluent N that was applied (Figure 12).

 

Figure 12. Effect of different levels of effluent N on the crude protein content
of the stems and leaves of water spinach
 

The efficiency with which nutrients are recycled is an important indicator of the sustainability of the farming system. For the first harvest, the transfer of nitrogen into the foliage of the water spinach was 54% of the quantity of nitrogen applied in the effluent (Table 11).  The comparable figure for cassava managed as a forage crop and fertilized with effluent from a biodigester charged with pig manure was 67%. 

Table 11.  Recovery of nitrogen in the foliage of water spinach and cassava when fertilized with biodigester effluent

Crops

N applied,
kg/ha

Biomass yield,
 kg/ha

DM,
%

N in DM,
 %

N in biomass, kg/ha

Recovery of N
%

Water spinach

140

24000

  9

3.50

75.6

54.0

Cassava#

100

  8700

22

3.52

67.4

67.4

# Le Ha Chau 1998a

 

Conclusions

 

References

AOAC  1990  Official Methods of Analysis. Association of Official Analytical Chemists. 15th edition (K Helrick editor) Arlington pp 1230 

Bui Xuan An, Preston T R and Dolberg F 1997 The introduction of low-cost polyethylene tube biodigesters on small scale farms in Vietnam. Livestock Research for Rural Development  (9) 2: http://www.cipav.org.co/lrrd/lrrd9/2/an92.htm 

Chamnanwit Promkot  2001 Study of the use of maize and water spinach in a biotest for evaluation of soil fertility. Mini-projects in MEKARN MSc course (http://www.forum.org.kh/~ mekarn) 

Cushman K, Bufogle A, and Horgan T 1999 Proceedings of the 28th National Agricultural Plastics Congress, Tallahassee, Florida, p.168.

Le Ha Chau 1998a Biodigester effluent versus manure from pigs or cattle as fertilizer for production of cassava foliage (Manihot esculenta). Livestock Research for Rural Development  (10) 3: http://www.cipav.org.co/lrrd/lrrd10/3/chau1.htm

Le Ha Chau 1998b Biodigester effluent versus manure from pigs or cattle as fertilizer for duckweed (Lemna spp.) Livestock Research for Rural Development  (10) 3: http://www.cipav.org.co/lrrd/lrrd10/3/chau2.htm  

MAFF 1994. Fertilizer Recommendations for agricultural and horticultural crops. Reference Book 209, 6th edition .London. HMSO.  

Randall G 1998 Hog manure boosts corn yield more than fertilizer in U of M study. University of Minnesota Extension Service. http://www.extension.umn.edu/extensionnews/1998/JO1013.html 

Undersander D, Mertens D R and Thiex N 1993 Forage analysis procedures. National Forage Testing Association. Omaha pp:154 

UNDP 1997  Annual Report

Back to top