In India to increase the rice production, increasing the rice productivity of alkali soils is an option which are prevalent to an extent of 2.4 million ha .The productivity of rice in alkali soil is low (2.5 to 3.0 t / ha) with low nutrient use efficiencies. Hence the Site Specific Nutrient Management (SSNM) approach was used to increase the rice yields with optimum fertilizer inputs.

With the SSNM approach, fertilizers are applied to meet the deficit between rice demand for nutrients and the indigenous nutrient supply of nutrients from soils and organic inputs for a targeted yield of rice. Hence, to know the indigenous nutrient supply from reclaimed alkali soils, on-farm trials (OFT) were conducted during 2003 -04 and 2004-05 in Manikandam block of Tiruchirappalli district, Tamil Nadu, India.

Also, one on-station experiment was conducted in alkali soils of A.D. Agricultural College and Research Institute farm (pH 9.2 and  ESP 26) in split plot design with three replications and rice (BPT 5204) as test crop during samba season (October 2004 to February 2005).  The main plot treatments comprised of commonly used amendments for reclamation of sodic soils viz., No amendment (M₁), Gypsum (M₂) and Distillery spentwash (DSW) (M₃). The subplot treatments included six nutrient omission plots viz.,   (-) F (T₁), PK (T₂), NK (T₃), NP (T₄), NPK (T₅) and N alone (T₆). The amendments gypsum @ 50% GR and DSW @ 5 lakh liters ha^-1 were applied one week and one month before transplanting respectively. The recommended dose of nutrients (RDN) was 150:50:50 kg/ ha of N, P2O5 and K2O was applied as per the treatment schedule viz., N in four splits, full dose of P and ZnSO₄ as basal and K in two splits.   In all the experiments, grain and straw yields were recorded, nutrient uptake, nutrient requirement to produce one tonne of rice and indigenous nutrient supply viz., N, P and K were computed (Table 1).

            Among the amendments, distillery spentwash (DSW) recorded higher rice grain yield.  The rice grain yield of 6.8 t / ha under no amendment gave scope for increasing rice yields in moderate   alkali soils with tolerant cultivars. In gypsum amended alkali soils, rice grain yield ranged between 4.25 and 7.5 t/ha. Compared to RDN with FFP, the yield gap varied between 0.06 to 0.88, 0.04 to 1.95 and 1.36 to 1.72 t /ha in no amendment, gypsum and DSW amended soils respectively. The average NPK requirement to produce one tonne of rice grain was reported as 17:3: 17 kg /ha in normal rice growing soils of Asia (Doberman and Fairhurst, 1999). But the nutrient requirement in both DSW and gypsum amended soils is higher indicating poor internal use efficiency under stress condition induced by alkali environment. The K requirement of rice was above 33 kg/ tonne in both no amendment and DSW amended soils while it ranged between 16 to 29 kg / tonne in gypsum amended soils. The indigenous nutrient supply was highest in DSW amended soils followed by gypsum amended alkali soils. Hence by growing alkali tolerant cultivars and reclaiming alkali soils either with DSW or gypsum and  following recommended dose of nutrients and splits, rice productivity of alkali soils can be improved.

Table 1.Nutrient requirement of rice  and indigenous nutrient supply in alkali soils

S.No.	Location	Year	Rice  grain yield (t / a)	Nutrient requirement (kg/ tonne)			Indigenous nutrient supply
I. Distillery spentwash amended				N	P	K	N	P	K
1.	ADAC&RI Farm	2003-04	7.11	22	4	33	103	20	209
2.	ADAC&RI Farm	2004-05	7.14	25	5	56	104	28	379
II Gypsum amended
3	Chinnadurai	2003-04	6.25	23	4	19	79	14	107
4	Rangaswamy	2003-04	7.07	21	4	21	76	18	129
5	Veeramuthu	2003-04	4.25	22	4	16	46	10	60
6	Arokiyasamy	2004-05	6.30	21	4	19	83	30	115
7	Deivegarajan	2004-05	5.50	16	3	21	54	19	92
8	Arumairaj	2004-05	6.30	22	6	25	78	30	119
9	ADAC&RI 1	2004-05	7.50	19	5	19	66	25	190
10	ADAC&RI 2	2004-05	6.40	21	5	26	80	24	164
11	ADAC&RI Farm	2004-05	7.00	19	4	29	74	25	203
III No Amendment
12	ADAC&RI Farm	2004-05	6.80	17	4	33	69	17	173