Nutrient management for paddy rice grown in soils with different fertility in Guizhou province

Abstract: Most of nutrient management studies on paddy rice in Guizhou province in the past was focused on medium to high fertility soils and left the low to medium fertility soil untouched. In order to improve paddy rice yield and nutrient management practices, three field experiments were carried out to test effects of different rates of nitrogen (N), phosphorus (P) and potassium (K) on rice yield in three types of fertility soils. The field experiments were consisted of twelve treatments including five rates of N (0, 90, 120, 150 and 180 kg N/ha), four rates of P (0, 45, 90 and 135 kg P₂O₅/ha) and four rates of K (0, 90, 135 and 180 kg K₂O/ha). The treatments repeated three times. Nitrogen was used as urea (N 46%), P as single super phosphate (P₂O₅ 12 %) and K as KCl (K₂O 60 %). Nitrogen and K were split into two times as basal application at seedling transplanting and at tillering stage with proportions of 40:60, and P and fertilizer was used once at seedling transplanting.
Results showed that rice responded differently to different rates of fertilizer on three paddy soils. In the low productivity soil, rice had little response to added N, P or K fertilizers and produced the lowest rice yield. In the medium productivity soil, rice significantly responded to added N, P and K fertilizers and rice yields leveled off above an optimal nutrient rate. In the high productivity soil, however, rice yields significantly increased with an increase in N and K rates without leveling off. These results are of great importance in making soil test-based fertilizer recommendations, that is, any successful fertilizer recommendations cannot consider soil nutrient levels only but also other factors such as soil productivity, non-nutrient constraints of the soil and the like.

Introduction
Low yield paddy soils are common in China. This type of paddy soil is estimated to be 7.83 million hectares in China, accounting for 26% of nation’s paddy fields. Guizhou province has about 300 thousand hectares of low yield paddy soils, accounting for about 20% of the paddy fields in the province. In these fields, the rice yields are usually 3000-3750kg/ha，far below the yield levels of 7500-9000 kg/ha on the medium to high fertility soils. Though several yield limiting factors such as constraint soil profile, low redox potentials, imbalanced fertilization and so on, exist, imbalanced fertilization is no doubt the major cause. Thus, the objective of this study was to test the optimal N, P and K rates that promote rice yields.

Materials and Methods
Three field experiments were conducted, two on low fertility soils and one on medium fertility soil located in Machang town and Gaofeng town, Pingba county, Guzhou province. The first low fertility soil was a yellow paddy soil that had organic matter content 42.6 g/kg, available N 86.3 mg/kg, available P 16.8 mg/kg and available K 80.6 mg/kg. The other major non-nutrient constraints for this soil were low redox potentials and high contents of ferrous iron. The second low fertility soil was a alluvial paddy soil developed from limestone with soil organic matter content 38.8g/kg, available N 93.6 mg/kg, available P 21.4 mg/kg and available K 88.4 mg/kg. The other non-nutrient constraints were sandy texture and low soil temperatures. The reference medium fertility soil was a yellow earth paddy soil with organic matter content 25.3 g/kg, available N 97.5 mg/kg, available P 25.9 mg/kg and available K 97.5 mg/kg. There were no other major constraints in this soil.

Three field experiments used the same randomized block design consisting of twelve treatments including five N rates (0, 90, 120, 150 and 180 kg N/ha), four P rates (0, 45, 90 and 135 kg P₂O₅/ha) and four K rates (0, 90, 135 and 180 kg K₂O/ha). The treatments repeated three times. Urea (N 46%) was used for N, single super phosphate (P₂O₅ 12 %) for P and KCl (K₂O 60 %) for K. Nitrogen and K fertilizers were split into two times as basal application at seedling transplanting and at tillering stage with proportions of 40:60, and P and fertilizer was used once at seedling transplanting.

Results and Discussion

The rice yield responded quite differently to different fertilizer treatments in different soils (Table 1-3). Among the three soils, the rice yields were the lowest in the yellow paddy soil1 but the highest in the yellow paddy soil2. Rice responded poorly to added N in the yellow paddy soil1 but highly significantly in the other two soils (Table 1). As a result, rice yield only increased by 4.81% at 90 kg N/ha but decreased with further increase in N levels in the yellow paddy soil1. Rice yield increased with an increase in N rates but leveled off at 150 kg N/ha in the alluvial soil. In the yellow paddy soil2, however, rice yield increased with an increase in N rates without leveling off. The results imply that the higher the productivity of the soil, the higher N rates needed.

Table 1 Effects of different nitrogen rates on rice yields on three soils with different fertility in Guizhou province

Treatment	Yellow paddy soil1		Alluvial paddy soil		Yellow paddy soil2
	Yield	Yield increase	Yield	Yield increase	Yield	Yield increase
	(kg/ha)	(%)	(kg/ha)	(%)	(kg/ha)	(%)
N0P2K2	7556 b	-	8059 c	-	8295d	-
N1P2K2	7902 a	4.82	8437 b	4.69	9051c	9.11
N2P2K2	7807a	3.32	8594 b	6.64	8972 c	8.16
N3P2K2	7461 b	-1.35	9224 a	27.95	9366 b	12.91
N4P2K2	7398 b	-2.10	7807 c	11.46	9932 a	19.73

The responsive trend of the rice to added P fertilizer in different soils was generally similar to N fertilizer (Table 2). A poor response of the rice to P was observed in yellow paddy soil1 but significant responses observed in the other two soils. Unlike the response to N fertilizer, there was virtually no P effect on rice yield in yellow paddy soil1 but rice yield dropped at P rate above 90 kg P₂O₅/ha in the other two soil. This implies that there is no need to apply P to the yellow paddy soil1 and the optimal P rates fall in 45-90 kg P₂O₅/ha in the other two soils.

Table 2 Effects of different phosphorus rates on rice yields on three soils with different fertility in Guizhou province

Treatment	Yellow earth paddy soil1		Alluvial paddy soil		Yellow earth paddy soil2
	Yield	Yield increase	Yield	Yield increase	Yield	Yield increase
	(kg/ha)	(%)	(kg/ha)	(%)	(kg/ha)	(%)
N3P0K2	7556 ab	-	7713 c	-	8531 b	-
N3P1K2	7713 a	2.08	8909 b	15.51	9161 a	7.38
N3P2K2	7461 b	-1.13	9224 a	19.59	9366 a	9.79
N3P3K2	6926 c	-8.34	8091 c	4.9	8579 b	0.56

In terms of K fertilization, rice had little response to added K in the yellow paddy soil1 and significantly responded to added K only at 90 kg K₂O/ha in the alluvial paddy soil (Table 3). In the yellow paddy soil2, however, rice yield significantly responded to added K rates without leveling off. The highest rice yield was achieved at the highest rate of K (180 kg K₂O/ha).
Table 3 Effects of different potassium rates on rice yields on three soils with different fertility in Guizhou province

Treatment	Yellow earth paddy soil1		Alluvial paddy soil		Yellow earth paddy soil2
	Yield	Yield increase	Yield	Yield increase	Yield	Yield increase
	(kg/ha)	(%)	(kg/ha)	(%)	(kg/ha)	(%)
N3P2K0	7241 a	-	8846 c	-	8075 c	-
N3P2K1	7083 b	-2.19	9948 a	12.46	8972 b	11.11
N3P2K2	7461 a	3.04	9224 b	4.27	9366 b	15.99
N3P2K3	7398 a	2.17	8122 d	-8.19	10987 a	36.06

The results of these field experiments reveal that when fertilizer recommendations are made on basis of soil testing, one should not consider soil nutrient levels only. The other factors such as soil productivity, non-nutrient constraints of the soil and the like are even more important in making successful fertilizer recommendations.