1

EffEctivE fErtilizEr ManagEMEnt PracticEsfor HigH YiEld ricE Production of

granarY arEas in MalaYsia

Abd Razak H, Suhaimi O, and Theeba MStrategic Resource Research Centre, MARDI

G. P. O. Box 12301, 50774 Kuala Lumpur, Malaysia

AbstrAct

High yield targets in modern rice farming should be supported by effective and efficient fertilizer management practices. Overdependence on blanket subsidy fertilizer recommendation is unable to produce high yield. Site-specific balanced fertilizer recommendation option by FERTO package is able to enhance crop performance, subsequently achieving high yield target. The significant yield increment (up to 300 %) using FERTO package recommendation is accomplished because due consideration is given to four basic principles of fertilizer management, i.e. nutrient quantity, nutrient access, indigenous soil fertility status, and balanced crop requirement. Site-specific approach of the FERTO package recommendation is complex. Therefore, FERTO package recommendation is more applicable to large-scale commercialized rice production. Thus, for policy formulation, FERTO package recommendation has to be transformed into formulation according to cropping zones. Development of fertilizer formulation according to cropping zone is more practical and can easily be managed by fertilizer suppliers and rice estate managers. The formulation can also be adopted by progressive small-scale farmers. Fourteen fertilizer formulations for specific management zones within the granary areas were identified in this study. The formulation is able to enrich proportionate vegetative and reproductive growth for high rice-yield performance. However, the application of FERTO package recommendation rate can be further enhanced by using straight fertilizers in place of the present subsidy mixture or compound fertilizers. These straight fertilizers are much cheaper compared to the mixture or compound fertilizers in the market. Therefore, the adoptions of FERTO package using straight fertilizers will reduce cost of production and ultimately increase net farm returns.

Keywords: Rice, FERTO package, fertilizer recommendation, high yield, granary areas.

IntroductIon

Overdependence on fertilizer input to boost yield performance is a common practice in rice farming. Subsidy fertilizer was introduced in Peninsular Malaysia in 1979 (Wong and Jegatheesan, 1990). The subsidy fertilizer is a blanket recommendation rate. The recommendation was developed through fertilizer studies under transplanting cultural practices using varieties with low to medium yield performance (3.5-5.0 t/ha). As a result of rapid industrialization and rising labor costs in the late 80s, the labor-intensive transplanting technique was abandon and farmers opted for direct seeding (Hamzah and Zanil Abidin, 1990). Since then, production system in rice farming (post-harvest field management, land preparation, crop establishment, weed management, pest and disease control, and new high-yielding variety)

has greatly improved. The amount of nitrogen (N) in the subsidy fertilizer was also increased from around 63 kg/ha in early 1980s to 80 kg/ha in late 1980s, and further increased to 110 kg/ha in 1990s (Aminuddin et al., 2004). The blanket approach of fertilizer recommendation remains until now. However, the national and granary average yields only hover around 3.5 and 3.8 t/ha, respectively (Abd Razak et al., 2004). The National Agricultural Policy 3 (NAP3) placed significant importance on food security, improving farm return, and eradication of poverty among farmers (Anon, 1999a). Rice farming, dominated by poor farmers, is one of a targeted sector that was given the emphasis. Towards this, the Ministry of Agriculture (MOA) proposed a new yield target of 10 t/ha (Abd Razak et al., 2002), which subsequently would increase farmer’s income. With the introduction of modern high-

tb194.indd 1 5/4/2014 10:46:08 AM

2

yielding varieties as a result of intensive research and development program, the targeted yield will reasonably be achievable. However, the newly introduced variety is known to be more responsive to fertilizer (Kundu and Ladha, 1999). Experiences have shown that the blanket fertilizer recommendation, which lack consideration for balanced crop nutrient requirement and in-situ soil fertility status, is a major factor constraining high yield achievement of the new varieties. Therefore, this paper examines a new approach of fertilizer management technology towards approaching 10.0 t/ha yield target under a set crop for high production. It involves FERTO package application to calculate site-specific balanced fertilizer recommendation for set high-yield target.

MAterIAls And Methods

FERTO package verification

The Fertilizer Recommendation Tool (FERTO) package was developed by MARDI. It provides site-specific fertilizer recommendation options. Before embarking on the exercise, the fitness of the model needs to be verified. For that purpose, various production-scale trials during main and off-seasons were conducted in MADA Kedah and KADA Kelantan. Soils from the study plots were sampled before land preparation. The soil was analyzed in the laboratory to determine its physico-chemical characteristics. Data was used to run the model, after which fertilizer recommendation for high yield production of the sites were produced for validation. Farm management activities started immediately after previous crop harvesting. The first activity was post-harvest field management involving slashing and burning of the above ground biomass. It was followed by first and second rotovations with raking under dried field condition. Final rotovation with land smoothing (nuplo) was carried out under wet condition. Immediately after that, broadcast seeding under water using motorized blower was undertaken. Germinated seed at rate 150 kg/ha of variety MR219 was used in this study. After seeding, in-field water was drained to saturation point for even seedling establishment. Other routine agronomic practices throughout the cropping cycle were strictly followed, such as infield water-depth control, weed management, and pest and disease control to ensure a better crop growth and to minimize yield loss. Fertilizer rate with four split applications, as

recommended by FERTO package model, was followed. Various growth performance indicators of crop at vegetative and reproductive stages such as number of tiller per square meter, number of leaf per tiller, and crop height were monitored to ensure crop achieve criteria of high production performance crop set. Finally, overall plot yield was measured during harvesting. Crop data will be used to indicate the performance of FERTO package recommendation in achieving high yield target. Yield from the study plot will also be compared with yield from the neighboring plots that were managed by farmer.

Mapping of fertilizer recommendation for granary areas

In Peninsular, rice is planted on 44 major soil series covering eight granary areas (Table 1). The physico-chemical characteristics of the soils were extracted from various soil survey reports prepared by Department of Agriculture. The data was used as input databases together with pre-set nutrient status for high-yielding crop and the set yield target to run the model. Only output from fertilizer recommendation window of the model that consists of N, P2O5, K2O, and processed chicken manure (CM) was used to develop fertilizer recommendation maps through integration with geographic information system (GIS). Each recommendation rate was spatially displayed by one pixel that represents the actual area sizes of 100 ha.

Development of fertilizer management zone

Using cluster analysis of GIS, fertilizer recommendation rates were summarized into fertilizer management zones. Dominant fertilizer recommendation rate of a growing area was used to represent the fertilizer management zones. Range of maximum and minimum fertilizer rate of each fertilizer element was also taken consideration.

results And dIscussIon

Fertilizer recommendation rate

Table 2 shows the fertilizer recommendation rates that were generated by FERTO package for plots in MADA and KADA areas, based on the physico-chemical properties of the soils and the set yield target. The recommendations had fertilizer rates

tb194.indd 2 5/4/2014 10:46:08 AM

3

ranging between 124 and 197 kg/ha for N, 45 and 95 kg/ha for P2O5, 129 and 222 kg/ha for K2O, and between 800 and 3400 kg/ha for chicken manure (CM). Generally, potassium fertilizer and chicken manure are recommended in higher amounts for crop in KADA, than in MADA. It is due to low potassium and cation exchange capacity (CEC) status of KADA soils, thus inability to enhance high yield. Soils originating from riverine alluvium with 1:1 parent material clay like KADA are generally of low inherent fertility status. MADA soil with marine 2:1 clay are always better in fertility status as indicated by their high organic matter content and CEC status. These two parameters are important to ensure that the fertilizer applied can be held by the soil before it can be taken up by the crop. Otherwise, the crop has less opportunity to absorb the applied fertilizer because most of it will be lost through leaching process. Normally, applied fertilizer is recommended to be split into various applications timing to

optimize crop nutrient uptake and minimize loss through leaching and evaporation processes. Therefore, FERTO package recommends four split applications, i.e. basal dressing, first top-dressing, second top-dressing, and final top-dressing during specific crop growing stages (Table 3). For basal application, compound fertilizer (15:15:15) at a blanket rate of 200 kg/ha was applied at seedling stage, about 15 days after seeding (DAS). Split applications of straight fertilizers for top-dressing were varied according to locality, depending on the rates recommended by FERTO package (Abd Razak et al., 2004). Chicken manure, if recommended, will be applied at early tillering stage, about 25 DAS. Fertilizer rate for each split application will be calculated by the model as formulated in the knowledge base. High N and K2O rates were recommended at vegetative and reproductive stages, respectively. This will ensure development of quality tillers and productive panicles to attain high yield performance.

Table 1. Major soil series planted with rice in the granary areas

No Soil series No Soil series No Soil series No Soil series

1 Akob 12 Hutan 23 Linau 34 Selangor2 Bakau 13 Idris 24 Lubuk Itik 35 Sembrin3 Batu Hitam 14 Jabil 25 Lubok Sendong 36 Serong4 Bernam 15 Jawa 26 Lunas 37 Sintok5 Binjai 16 Kangkong 27 Lundang 38 Sungai Amin6 Briah 17 Kecai 28 Manik 39 Tebengau7 Brown Clay 18 Kranji 29 Organic Clay 40 Telemong8 Chempaka 19 Kuala Kedah 30 Pasir Putih 41 Telok9 Chengai 20 Kuala Perlis 31 Sabrang 42 Tepus10 Cherang Hangus 21 Kundur 32 Sedaka 43 Tok Yong11 Guar 22 Lating 33 Sedu 44 Tualang

Table 2. FERTO package and farmer fertilizer rates used in the study locations

Granary Location (Season) FERTO Rate Farmer Rate N P2O5 K2O CM N P2O5 K2O (kg/ha)

MADA Kobah (1/2000) 197 95 150 1568 135 75 55KADA Ketereh (1/2001) 137 49 129 3400 149 89 69 Meranti (1/2002) 188 85 195 3400 192 110 124 Meranti (2/2002) 124 94 221 2393 192 110 124 Meranti (1/2003) 124 94 221 800 192 110 124 Senor (1/2003) 149 95 222 2418 150 90 70

tb194.indd 3 5/4/2014 10:46:09 AM

4

Crop yield performance

The yields obtained from the study plots with FERTO package recommendation rates and from the farmer plots are summarized in Table 4. The yields ranged from 7.1 to 10.1 t/ha with the FERTO package rate and from 1.7 to 5.5 t/ha with farmers’ management. The yield was increased significantly in Senor (KADA) exceeding 300 %, while in Kobah (MADA) yield improvement exceeded 80 %. In Ketereh and Meranti, yield increase between 125 and 200% was recorded. The yields from FERTO package plot were higher compared to the farmer’s plot, indicating that the application of FERTO package rate has positive impact on improving rice yield performance. The success of the recommendation is attributed to

the incorporation of four principles of effective fertilizer management to ensure proportionate vegetative growth and reproductive development (Suhaimi and Abd Razak, 2004). These principles are:

i. Nutrient quantity (amount of fertilizer required for set yield target),

ii. Nutrient access (continuous nutrient availability throughout a cropping cycle),

iii. Indigenous soil fertility status , andiv. Balanced crop nutrient requirement

(appropriate nutrient ratio at specific crop stages).

Based on the achievements, the recommended FERTO package rates are able to improve rice yield performance towards achieving the set yield target.

Table 3. Split applications of FERTO package fertilizer recommendation rate.

Location: Senor (KADA) Sampling date : 15/04/03 Season : I Seeding date : 15/05/03 Recommendation N P2O5 K2O CMrate (kg/ha)

149 95 222 2418 Split Crop NPK UREA MOP TSP CMApplication stage (15:15:15) (46) (60) (46) (kg/ha)

Basal dressing Seedling 200 Organic fertilizer Early tillering 24181st Top-dressing Active tillering 42 72 110 2nd Top-dressing Panicle initiation 132 124 65 Final top-dressing Initial heading 84 124

Table 4. Yield performance based on FERTO package and farmer fertilizer rates

Granary Location (Season) FERTO FERTO yield Farmer yield Plot size (ha) (t/ha) (t/ha)

MADA Kobah (1/2000) 3.1 10.1 5.5KADA Ketereh (1/2001) 0.4 8.1 3.5 Meranti (1/2002) 1.2 7.6 2.5 Meranti (2/2002) 1.2 7.9 3.5 Meranti (1/2003) 1.2 8.7 3.5 Senor (1/2003) 1.3 7.1 1.7

tb194.indd 4 5/4/2014 10:46:09 AM

5

Spatial fertilizer recommendation

Fertilizer recommendation rates for high rice yield performance of granary areas are spatially displayed in different N, P2O5, K2O, and CM maps (Fig. 1-4). The recommended rate is based on single pixel that represents 100 ha on the ground. The nitrogen, phosphorus and potassium fertilizer rates range between 109 and 198, 79 and 97, and 135 and 233 kg/ha, respectively. The chicken manure (media with capacity to continuously supply available nutrient in minute quantities throughout the crop cycle) ranges between 0 (no application) and 3.2 t/ha. Soils with CEC exceeding 25 meq/100 g soil do

not require chicken manure application due to their ability to hold applied nutrients and are less prone to leaching process. Based on FERTO package yield performance, it can be concluded that the blanket subsidy fertilizer rate (120N: 60P2O5: 40K2O kg/ha) is deficient in nutrients to achieve the yield target. Therefore, to achieve the high yield target set by MOA, additional fertilizer, as recommended by FERTO package, should be included in the recommendation. The additional fertilizer can be incorporated into the present policy of subsidy fertilizer to help poor small-scale farmers realize the high yield target.

Fig. 1. Nitrogen fertilizer rate (kg/ha) for high yield performance of rice in the granary areas

MADA I

MADA II

MADA III

MADA IV

IADP Penang

Krian Darat Perak

KADA

Semerak--Kemasin Kelantan

KETARA Terengganu

PBLS South Selangor

FELCRA Seb. Perak

Krian Laut Perak

Sungai Manik

PBLS North Selangor

tb194.indd 5 5/4/2014 10:46:10 AM

6

Fertilizer management zone

The use of site-specific FERTO package recommendation as a new subsidy fertilizer is complicated. The recommendation rate may hinder producers or suppliers to formulate the fertilizer. To overcome the problem, re-grouping the recommendation based on dominant recommendation rate through cluster analysis was done. Each recommendation rate is specifically targeted for a cropping zone. A total of 14 cropping zones were identified. Table 5 shows the fertilizer recommendation zones of N, P2O5, K2O and CM that can be utilized to achieve 10 t/ha yield target. The N, P2O5, K2O and CM range between 140 and 170, 80 and 95, 190 and 230, and 800 and 2500 kg/ha, respectively. The FERTO package recommendation zoning essentially is the delineation of the current

eight granary areas into 14 zones with location-specific fertilizer rate (Table 5). On top of efficient fertilizer management, other agronomic practices must come together towards achieving sufficient crop population for set high-yield target. Insufficient crop population per unit area is one of the major constraints that hinder high yield performance. The problem will exist due to existence of low and high spots in the plot due to poor land preparation (Suhaimi and Abd Razak, 2004). Low spot will cause water ponding, subsequently resulting in low seedling establishment. High spot will expose the soil surface to drying, causing tiller death and heavy weed infestations. High spot with shallow water depth will stimulate the crops to produce excess tiller numbers. This phenomenon will result in the development of non-uniform tiller quality that

Fig. 2. Phosphorus fertilizer rate (kg/ha) for high yield performance of rice in the granary areas.

MADA I

MADA II

MADA III

MADA IV

IADP Penang

Krian Darat Perak

KADA

Semerak--Kemasin Kelantan

KETARA Terengganu

PBLS South Selangor

FELCRA Seb. Perak

Krian Laut Perak

Sungai Manik

PBLS North Selangor

tb194.indd 6 5/4/2014 10:46:11 AM

7

Fig. 3. Potassium fertilizer rate (kg/ha) for high yield performance of rice in the granary areas

MADA I

MADA II

MADA III

MADA IV

IADP Penang

Krian Darat Perak

KADA

Semerak--Kemasin Kelantan

KETARA Terengganu

PBLS South Selangor

FELCRA Seb. Perak

Krian Laut Perak

Sungai Manik

PBLS North Selangor

Fig. 4. Chicken manure rate (kg/ha) for high yield performance of rice in the granary areas

MADA I

MADA II

MADA III

MADA IV

IADP Penang

Krian Darat Perak

KADA

Semerak--Kemasin Kelantan

KETARA Terengganu

PBLS South Selangor

FELCRA Seb. Perak

Krian Laut Perak

Sungai Manik

PBLS North Selangor

tb194.indd 7 5/4/2014 10:46:13 AM

8

definitely cannot produce high yield performance. Proper management of infield water depth together with application of chemicals will prevent weed infestation effectively. Other yield reduction factors affecting high yield performance are pest and disease attack (Abdul Latif et al., 1990), and harvesting and field handling losses. Pest and disease attack can be minimized through calendar sprays as practiced by many progressive farmers. Harvesting and field handling losses can be reduced through proper supervision during operation by respective growers.

conclusIon

From the present study, it can be concluded that adoption of proposed FERTO package fertilizer management practices is able to achieve the 10 t/ha yield target. The practice will enable rice farmers to achieve high and sustainable rice production at farm level. Subsequently, national rice production can be increased, self-sufficiency level can be met, and import of rice commodity can be reduced for the purpose of balance of trade. The complexity to follow FERTO package recommendation as mentioned earlier can be overcome if growers are willing to accept a new subsidy fertilizer approach using straight fertilizers. Under such circumstances, the FERTO package recommendation can be adopted by growers with additional benefit of cost saving because straight fertilizers are cheaper than

the compound fertilizer currently used. This will increase crop performance and yield achievement, reduce cost of production, and increase net farm return.

references

Abd Razak H, Aminuddin BY, Ramlan MF, Muhammad H, and Abdul Rahman AB. 2002. Nutrient management in rice cultivation, Paper presented at the Seminar on Application of remote sensing, GIS and related technologies for precision farming, MACRES, September 9-11, 2002, Kuala Lumpur.

Abd Razak H, Suhaimi O, Aminuddin BY, Ramlan MF, and Mohd Zain M. 2004. FERTO: A computerized fertilizer management tool for high yield rice production, Zain ZZ, et al. editors, Proc. Soils 2004 Conference, April 2004, Penang, Malaysia, MSSS, p. 13-17.

Abdul Latif AZ, Azmi M, Saad A, and Nik Mohd Noor S. 1990. Management of pesticide, chemical and herb in rice crop, Othman O, et al. editors, Improving effectiveness of input used in rice production., MARDI Seberang Perai, Penang.

Aminuddin BY, Abd Razak H, Mohammud CH, Mohd Zain M, and Normah M. 2004. An assessment of paddy soil degradation and its impacts on sustainable rice production, Malaysia Journal of Soil Science, Vol. 7.

Table 5. Fertilizer recommendation zones by FERTO package for high rice yield performance

No. Zone name N P2O5 K2O CM (kg/ha)

1 MADA I 170 80 200 8002 MADA II 160 80 200 5003 MADA III 160 85 200 5004 MADA IV 170 85 200 8005 IADP Penang 140 85 190 10006 Krian Laut 140 95 190 5007 Krian Darat 140 85 190 10008 FELCRA Seberang Perak 140 85 190 15009 Sungai Manik 140 95 230 250010 PBLS North 160 85 230 100011 PBLS South 150 80 200 50012 KETARA 160 95 230 180013 Semerak-Kemasin 160 95 230 250014 KADA 170 95 230 2000

tb194.indd 8 5/4/2014 10:46:13 AM

9

Anon. 1999a. Third National Agricultural Policy 1998-2010., Ministry of Agriculture, Kuala Lumpur.

Hamzah C, and Zanil Abidin Y. 1990. Farm organization system with management and input used for rice in Projek Barat Laut Selangor (PBLS), Othman O, et al. editors, Improving effectiveness of the input used in rice production., MARDI Seberang Perai, Penang.

Kam SP. Ariffin T, Hoanh CT, Abd Razak H, Rala AH, and Villano L. 2002. Rice supply and demand scenarios for Malaysia., Sombilla M, et al. editors, Development in the Asian Rice Economy, IRRI, The Philippines.

Kundu DK, and Ladha JK. 1999. Sustaining productivity of lowland rice soils: issues and options related to N availability, Balasubramaniam V, et al. editors, Resource management in rice systems: Nutrients, pp. 27-44, Kluwer Academic Publishers, The Netherlands.

Suhaimi O, and Abd Razak H. 2004. The adoption of CREST and FERTO technology package for high yield rice production, Paper presented at the meeting of MARDI Scientific Council No. 118, June 17, 2004, Cameron Highlands, Malaysia.

Wong CY, and Jegatheesan S. 1990. Impact and Implication of the Fertilizer Subsidy Scheme: A view from the fields, Othman O, et al. editors, Improving effectiveness of the input used in rice production, MARDI Seberang Perai, Penang.

tb194.indd 9 5/4/2014 10:46:13 AM