Research Findings: IV Reaching Towards Optimal Productivity

By R.J. Buresh, M.F. Pampolino, P.S. Tan, R. Rajendran, H.C. Gines, T.T. Son, and S. Ramanathan
print version pdf 687 kb

The workgroup "Reaching Toward Optimal Productivity" (RTOP) of the Irrigated Rice Research Consortium (IRRC) has been instrumental in the development, evaluation, and promotion of Site-Specific Nutrient Management (SSNM) as an approach for increasing farmers' profit through more efficient use of nutrients (IRRI 2006).

In January 2001, the RTOP workgroup was established with funding from the Swiss Agency for Development and Cooperation (SDC), the International Fertilizer Industry Association (IFA), The International Potash Institute (IPI), and the Potash & Phosphate Institute (PPIPPIC). The workgroup supported nutrient management-related research and the delivery of SSNM through partnerships with the National Agricultural Research and Extension Systems (NARES) in Bangladesh, China, India, Indonesia, Myanmar, the Philippines, Thailand, and Vietnam.

As a result of RTOP, the principles and practice of SSNM for rice have been well formulated and widely recognized across Asia. These SSNM principles have subsequently been utilized by partners across Asia to develop location-specific nutrient management practices for major rice-growing area.

Since 2001, the RTOP Workgroup of the IRRC has collaborated with NARES in eight Asian countries to systematically transform the initial SSNM concept (developed since 1994) into an inclusive, simplified framework for the dynamic plant-need based management of N, P, and K (IRRI 2006). The SSNM approach now enables: 1) dynamic adjustments in fertilizer N, P, and K management to accommodate field- and season-specific conditions; 2) effective use of indigenous nutrients; 3) efficient fertilizer N management through the use of the leaf color chart (LCC), which helps ensure N is applied at the time and in the amount needed by the rice crop; 4) use of the omission plot technique to determine the requirements for P and K fertilizer; and 5) use of micronutrients based on local recommendations.

Improved nitrogen management

The general principles for N fertilizer management with SSNM include: a) the application of only a moderate amount of N to young rice, within the first 14 days after transplanting (DAT) for transplanted rice or first 21 days after sowing (DAS) for direct-seeded rice, when the demand for N is small; and 2) the dynamic management of fertilizer N to ensure sufficient supply of N to the crop at the critical growth stages of mid-tillering and panicle initiation.

Parameter	India (Cauvery Delta)	Philippines (Nueva Ecija)	Vietnam
			(Mekong Delta)	(Red River Delta)
	(n = 61)	(n = 21)	(n = 56)	(n = 60)
Attainable yield with NPK (t/ha)	5.6 - 7.2	5.6 - 7.0	5.6 - 7.4	6.8 - 8.0
Yield without N fertilizer (t/ha)	3.2 - 4.8	3.4 - 4.8	3.5 - 4.9	4.3 - 5.9
Crop response to N fertilizer (t grain/ha)	1.6 - 3.2	1.4 - 3.2	1.6 - 3.0	1.6 - 3.0
Targeted agronomic efficiency of applied N (Δkg grain/kg N)	23	23	23	23
Approximate fertilizer N requirement (kg N/ha)	70 - 136	60 - 136	71 - 127	71 - 133
Estimated number of N applications during the season	3 - 4	3 - 4	3 - 4	3 - 4
N dose for each application of fertilizer N (kg N/ha)	20 - 45	20 - 45	20 - 35	10 - 40
Table 1. Attainable yield with NPK fertilization, crop response to N fertilizer, and N fertilizer requirements for the high-yielding season at four locations in 2001-04.

Table 1 presents for four locations an example of developing an N recommendation based on data collected during 2001-2004 high-yielding season. Yields obtained with sufficient N, P, K, (and Zn where needed) to eliminate deficiencies of these nutrients were considered as yield targets. The attainable yield with NPK was comparable among three sites (Cauvery Delta in India, Nueva Ecija in the Philippines, and Mekong Delta in Vietnam), which ranged from 5.6 to 7.4 t/ha. The Red River Delta (RRD) in Vietnam showed the highest attainable yield with NPK (6.8 - 8 t/ha). Yield without N fertilizer was also highest in the RRD, indicating the effect of heavy application of farm yard manure (about 10 t/ha/crop) by farmers. The N-limited yield gap, which is the response to N application, was similar across the four locations and ranged from 1.4 to 3.2 t/ha. Hence, the approximated fertilizer N requirement was also comparable among the four locations when a uniform agronomic efficiency of N (AEN, 23 kg increase in grain per kg of N applied) is targeted.

Fig. 1 and Fig. 2 show examples of N recommendations that indicate amount and timing of N applications as developed through the SSNM approach (IRRI, 2006). Early N applications are small and vary depending on the amount and quality of applied farmyard manure (FYM) or organic materials, as shown by the low rate of basal N (10-15 kg N/ha) in northern Vietnam where FYM application is common and the higher rate of early N (30 kg N/ha) in the Philippines where application of organic source of N is minimal. Rates of N application during the critical growth stages of tillering and panicle initiation are adjusted based on plant need for nitrogen as indicated by leaf color.

Improved P and K management

Through the process of developing and evaluating SSNM, the nutrient omission plot technique was conducted on numerous farmers' fields in the different RTOP sites from 2001-2004. Yields for fully fertilized plots (no N, P, K, and Zn constraints to crop growth), yields for Pomission plots, and yields for K-omission plots varied among years and farmers' fields depending upon climate and crop management practices.

Crop responses to P and K (i.e., the differences between yields in NPK plots and yields in nutrient omission plots) varied within a range of 1 t/ha for each of the five locations as shown in Table 2. More than 1 t/ha response to P was obtained in the Mekong Delta (MD) and the RRD Vietnam, and in some cases also in Nueva Ecija, Philippines. Crop response to K was generally higher (> 0.5 t/ha) in the RRD and the New Delta (Southern India) than the other three locations (Table 2).

Parameter	India (Cauvery Delta)		Philippines	Vietnam
	Old Delta	New Delta	Nueva Ecija	Mekong Delta	Red River Delta
	(n = 29)	(n = 32)	(n = 21)	(n = 56)	(n = 60)
Attainable grain yield with NPK (t/ha) †	5.4 - 7.0	5.7 - 7.3	5.6 - 7.0	5.6 - 7.4	6.8 - 8.0
Crop response to P fertilizer (t grain/ha) †	0 - 0.6	0.2 - 0.8	0.1 - 1.1	0.7 - 1.7	0.4 - 1.2
Crop response to K fertilizer (t grain/ha) †	0.2 - 0.8	0.5 - 1.1	0 - 0.9	-	0.6 - 1.6
Estimated amount of straw returned (t/ha)	1.0	1.0	2.5	0.5	-
Farmyard manure application (t/ha)	-	-	-	-	10
Estimated P requirement (kg P2O5/ha) †	25 - 33	26 - 36	25 - 33	26 - 40	9 - 25
Estimated K requirement (kg K2O/ha) †	50 - 78	57 - 87	29 - 57	-	17 - 47
Table 2. Attainable yield, P and K deficits, and estimated P and K requirements during the high-yielding season at five locations in three countries in 2001-04.

Fertilizer P and K requirements — estimated using NuDSS software (Nutrient Decision Support System for rice, Witt et al., 2005), which maintains the scientific principles of the underlying QUEFTS model for rice (Witt et al., 1999, Janssen et al., 1990) — varied depending on yield target (yield with NPK), crop response to P or K, and the amount of organic inputs (e.g. rice straw, FYM). Fertilizer P requirements for the high-yielding season in Southern India (Old and New Delta) and the Philippines were comparable and ranged from 25 to 36 kg P₂O₅/ha (Table 2). MD had the highest P requirement (26-40 kg P₂O₅/ha) because of the very low organic input (0.5 t straw/ha/crop) and the relatively low P supply as indicated by a high crop response to P (0.7-1.7 t/ha), while RRD had the lowest fertilizer P requirement (9-25 kg P₂O₅/ha) due to the heavy application of FYM (about 10 t/ha/crop) (Table 2). Since large applications of FYM are common in the RRD, the estimated fertilizer K requirements for that location were low (17-47 kg K₂O) despite the high crop response to K (0.6-1.6 t/ha). Estimated fertilizer K requirements were highest in Southern India due to low straw input, especially in the New Delta (57-87 kg K₂O/ha) where a higher crop response to K (0.5-1.1 t/ha) was indicated (Table 2).

Benefits of SSNM

Yield increase and economic benefits with SSNM

Fig. 3. Grain yield obtained with farmers' fertilizer practice (FFP) and site-specific nutrient management (SSNM) conducted in farmers' fields at four locations in 2001-04. The asterisk (*) indicates a significant difference between the two treatments at P<0.05. The total number of farmers per location participating in these evaluation trials ranged from 44 to 90 in the highyielding season and 47-112 in the low-yielding season.

Grain yield obtained in on-farm evaluation of SSNM from 2001-2004 showed an increased yield with SSNM as compared to farmers' fertilizer practice (FFP), consistently across four locations and for both high- and low-yielding seasons (Fig. 3). A comparison of the economic benefits in terms of gross return above fertilizer cost (GRF) derived using SSNM and FFP for two years (2003-2004) in the Red River Delta (Northern Vietnam) revealed a higher GRF for SSNM than for FFP (Table 3). The added benefit from using SSNM amounted to US$ 147 for two crops per year (Table 3).

Parameter	Unit	Spring rice		Summer rice		Annual total
		FFP	SSNM	FFP	SSNM	FFP	SSNM
Grain yield	t/ha	6.95	7.44	5.21	5.51	12.16	12.94
Gross returns  (GR)†	US$/ha	960	1027	784	829	1744	1856
Fertilizer N application	kg N/ha	114	87	117	85	231	172
Fertilizer P application	kg P/ha	31	26	34	25	65	51
Fertilizer K application	kg K/ha	63	76	63	76	126	152
Total cost of fertilizers (FC)	US$/ha	117	103	130	107	246	210
Gross return above fertilizer cost (GRF) ‡	US$/ha	844	924	655	722	1498	1645
Added benefit from SSNM §	US$/ha		80		67		147
Table 3. Economic benefit derived with using improved N, P, and K management with site-specific nutrient management (SSNM), averaged over two years (2003-04) and 60 farms across five soil types in the Red River Delta, northern Vietnam. Actual costs of produce and inputs were used.
† Gross returns (GR) = grain yield x price of grain	‡ Gross returns above fertilizer cost (GRF) = GR - FC			§ Added benefit from SSNM = GRFSSNM - GRFFFP

Fertilizer N use efficiency and its impact in the environment

The use of SSNM resulted to increased yields and higher N use efficiency (i.e., partial factor productivity, PFPN, expressed as kg grain per kg applied N) as compared to farmers' fertilizer practice (FFP) without necessarily reducing application rates of fertilizer N (Table 4).

Country	Location	Nutrient management practice†	Applied fertilizer	Annual yield	Partial factor productivity	GWP‡ expressed as carbon dioxide equivalent (CDE)
			(kg N ha-1)	(t ha-1)	(kg grain kg-1 applied N)	(kg t-1 grain yield)	(kg kg-1 fertilizer N)
India	Old Delta (n = 5)	FFP	258	12.4	48	61	2.8
		SSNM	256	12.9	51	58	2.9
		Prob > t §	0.75	0.004	0.23	0.66	0.85
	New Delta (n = 5)	FFP	201	11.6	58	81	4.7
		SSNM	241	12.8	53	89	4.7
		Prob > t	<0.001	<0.001	0.01	0.31	0.87
Philippines ¶	Nueva Ecija (n = 10)	FFP	249	9.7	39	415	15.8
		SSNM	225	10.4	47	366	16.7
		Prob > t	0.07	0.004	0.02	0.22	0.38
Vietnam #	An Giang (n = 5)	FFP	197	9.3	47	212	9.6
		SSNM	174	9.6	55	160	8.5
		Prob > t	0.006	0.24	<0.001	<0.001	<0.001
	Cantho (n = 5)	FFP	224	9.4	42	180	7.5
		SSNM	189	10.3	55	116	6.3
		Prob > t	<0.001	0.002	<0.001	<0.001	<0.001
	Tien Giang (n = 5)	FFP	218	8.8	41	201	8.0
		SSNM	188	9.6	51	130	6.6
		Prob > t	0.001	0.003	<0.001	<0.001	0.003
Table 4. The contribution of simulated N2O emission from soil and fertilizer to annual global warming potential (GWP) as influenced by nutrient management practices based on measured on-farm annual rice yields and fertilizer N application rates in three countries for two irrigated rice crops in 2002-03.
† FFP is farmers' fertilizer practice and SSNM is site-specific nutrient management ‡ GWP based on N2O emissions from soil and fertilizer as simulated by DNDC (Li 2000; Li et al. 2004) § Probability level of the difference between FFP and SSNM. ¶ Two crop establishment methods with five farms per method; there was no crop establishment x treatment interaction, hence values shown are means of the two crop establishment methods. # Sites included are only for southern Vietnam (Mekong Delta)

Simulations with DNDC model (DeNitrification-DeComposition); (Li, 2000; Li et al., 2004) showed possible benefits of SSNM on reduced global warming potential (GWP) associated with reduced emissions of N₂O. The GWP expressed per unit of grain yield and unit of fertilizer N was significantly reduced by SSNM at all three sites in Southern Vietnam (Table 4). Treatment differences in Southern India and the Philippines were not significant, although a reduction in GWP with SSNM was also indicated in the Philippines. The simulated reduction in GWP with use of SSNM averaged 49 kg carbon dioxide equivalent (CDE) t^-1 grain in the Philippines and 62 kg CDE t^-1 grain in Vietnam (Table 4). This corresponded to an average reduction of 56 kg CDE t^-1 grain across the two countries or a 22% reduction in CDE per unit of rice produced. There was no reduction in GWP due to SSNM in Southern India probably because the efficiency of fertilizer N (e.g. PFPN) and yields were already relatively high with FFP (Table 4). The simulated N₂O emissions were low (data not shown), perhaps as a reflection of the already relatively high efficiency of fertilizer N use. In such cases, SSNM was able to increase yield with the same (Old Delta) or increased (New Delta) rate of fertilizer N without additional N₂O emission per unit of grain yield or fertilizer used.

	Grain yield in N omission plots (-N, +P, +K) is used to estimate the indigenous supply of N from soil and other nonfertilizer sources.
	Grain yield in P omission plots (-P, +N, +K) is used to estimate the indigenous supply of P from soil and other nonfertilizer sources.
	Grain yield in K omission plots (-K, +N, +P) is used to estimate the indigenous supply of K from soil and other nonfertilizer sources.

References
International Rice Research Institute (IRRI). 2006. Site-specific nutrient management. http://www.irri.org/irrc/ssnm/ Accessed 23 Oct 2006.

Janssen, B.H., Guiking, F.C.T, van der Eijk, D., Smaling, E.M.A., Wolf, J., and H. van Reuler. 1990. A system for quantitative evaluation of the fertility of tropical soils (QUEFTS). Geoderma 46:299-318.

Li, C. 2000. Modelling trace gas emissions from agricultural systems. Nutr. Cycling Agroecosyst. 58: 259-276.

Li, C., Mosier, A., Wassman, R., Cai, Z., Zheng, X., Huang, Y., Tsuruta, H., Boonjawat, J. and R. Lantin. 2004. Modelling greenhouse gas emissions from rice-based production systems: Sensitivity and upscaling. Global Biogeochem. Cycles 18: GB 1043, doi: 10.1029 / 2003 GB 002045.

Pampolino, M.F., Manguiat, I.J., Ramanathan, S., Gines, H.C., Tan, P.S., Chi, T.N., Rajendran, R. and R.J. Buresh. 2006. Environmental impact and economic benefits of site-specific nutrient management (SSNM) in irrigated rice systems. Agricultural Systems (in press).

Witt, C., Dobermann, A., Abdulrachman, S., Gines, G.C., Wang, G.H., Nagarajan, R., Satawathananont, S., Son, T.T., Tan, P.S., Tiem, L.V., Simbahan, G.C. and D.C. Olk. 1999. Internal nutrient efficiencies of irrigated lowland rice in tropical and subtropical Asia. Field Crops Res. 63: 113-138.

Witt, C. and Dobermann, A. 2004. Toward a decision support system for site-specific nutrient management. In pp 359-395 Dobermann A, Witt C, Dawe D (eds). Increasing the productivity of intensive rice systems through sitespecific nutrient management. Enfield, NH (USA) and Los Baños (Philippines): Science Publishers, Inc., and International Rice Research Institute (IRRI). p 1-420.

Witt, C., Fairhurst, T.H., Sheehy, J.E., Dobermann, A. and A. Gfroerer Kerstan. 2005. A nutrient decision support system (NuDSS) for irrigated rice (online). Available at www.irri.org and www.seap.sg (accessed 28 Apr 2006). Los Baños, Philippines: International Rice Research Institute (IRRI) and Singapore: Potash & Phosphate Institute/Potash & Phosphate Institute of Canada (PPI/PPIC) and International Potash Institute (IPI).

print version pdf 687 kb