Introduction

Rice (Oryza sativa L.) is a staple food crop for a large proportion of the world's population, and its demand is accelerating with the increasing population (Bezbaruha et al. 2011). Best practice nutrient management in cropping systems is characterized by the application of nutrients (no matter what the source: organic and/or inorganic) in the right quantities, in the right form, at the right time, and in the right place (Vaarst 2010). The mineralization of organic nitrogen (N) is a key process in determining the effectiveness of N nutrition for rice. Thus, improved understanding of both N dynamics in soil and its uptake by crops is necessary to improve nutrient use efficiency (Gastal and Lemaire 2002). Organic manures are usually applied during land preparation because their nutrient transformation/mineralization process is very slow. Crop response to organic manures depends on both the type and quality of manure (Tandon 1992). Vermicompost is rich in nutrients (Gandhi et al. 1997) and microbial activity (Edwards 1998) and contains growth hormones (Nielson 1985). All nutrients in vermicompost are in a readily available form, thereby enhancing uptake of nutrients by plants and, in turn, crop yield (Banik and Sharma 2009; Bejbaruha et al. 2009; Sreenivas et al. 2000; Jadhav et al. 1997). Vermicompost can be used alone for organic farming and as a component of integrated nutrient management for rice production (Banik and Sharma 2009; Bejbaruha et al. 2009, 2011). The nutrient use efficiency (NUE) of organic manures applied in basal dressings is quite low usually ranging from 5 to 20 % (Manivannam and Sriramchandrasekharan 2009; Myint et al. 2010). In order to increase crop yield and NUE, it has been recommended that fertilizer N is applied in split doses at different critical growth stages of a crop (Perez et al. 1996; Peng et al. 2010). Thus, this study aimed to test whether NUE and crop yield can be enhanced by split applications of vermicompost.

Materials and methods

Experimental site and soil

Experiments were conducted with rice (cv. Pankaj) under rainfed conditions during the wet season (June–September) in 2008 and 2009 at the Agricultural Experimental Farm of the Indian Statistical Institute, Giridih, India. The farm is located in the eastern plateau region (24° 1′ 30″ N latitude; 80° 31′ 15″ E longitude) at an elevation of 276 m above mean sea level with an average slope of 2 %. The climate is subhumid with an average annual rainfall of 1,343 mm and average maximum and minimum temperatures of 23.8 and 16.8 °C, respectively. The relative humidity ranges from 78 to 95 %, and the mean annual evapotranspiration is 1,293 mm. The soil is an Oxisol with a loam texture (26.3 % clay, 42 % sand, and 31.7 % silt). The physicochemical properties of the soil (0–20 cm) determined by standard methods (Jackson 1973) were the following: pH 5.7, 0.43 % organic C, 0.04 % total N, 13.8 kg ha−1 available P, and 138 kg ha−1 available K.

Experimental treatments and field operations

Five treatments were compared in plots measuring 5 × 5 m arranged in a randomized block design with each treatment replicated four times. The treatments, except the absolute control (T1), were supplemented with vermicompost (VC) to supply 120 kg N ha−1 as follows: full basal (T2), 1/2 basal + 1/2 at maximum tillering (MT) stage (T3), 1/3 basal + 1/3 at MT stage + 1/3 at panicle initiation (PI) stage (T4), and 1/3 basal + 1/3 at MT stage + 1/6 at PI stage + 1/6 at flowering stage (T5). Vermicompost contained 1.7 % N, 0.7 % P2O5, and 0.86 % K2O. Three seedlings (20–23 days old) were transplanted per hill with a plant-to-plant spacing of 15 cm and row-to-row spacing of 20 cm. Transplanting was done on 14 and 18 July and harvesting on 13 and 16 November in 2008 and 2009, respectively. The crop was weeded twice manually. There was no pest control used as major insect pest incidence was not noticed.

Preparation of vermicompost

Six biodegradable organic wastes (BOWs) were used for the preparation of vermicompost, viz., cow dung, municipal solid waste, aquatic weeds such as common duckweed (Lemna minor L.), water hyacinth (Eichhornia crassipes), Bermuda grass (Cynodon dactylon), fly ash (effluent from thermal power plant), and kitchen waste. Adult earthworms (Eisenia fetida), suitable for this subtropical climate, were inoculated at the rate of 15 worms per kg of biodegradable waste (Edwards 1998). The materials were incubated for 50 days. During vermicomposting, watering was done with the help of a sprinkler as and when required to moisten the BOWs. The feed material was converted into loose granular mounds as a result of passing through the alimentary canal of the earthworms. All material was collected, and two pyramidal heaps were made. After a day, a heap was slowly brushed aside, and adult earthworms were collected. The compost was dried in the shade for 2 days and sieved (4 mm mesh) to separate cocoons. Samples (100 g) were collected for chemical analysis of nutrients in the laboratory. Organic carbon was estimated using the method described by Walkley and Black (1934). Total nitrogen, total phosphorus, total potassium, and micronutrients were estimated by adapting a method described by Jackson (1973).

Yield and yield components

Plant height (10 randomly selected hills) and panicles per square meter area were recorded at harvest. From the harvested area, 10 panicles were randomly selected to record numbers of spikelets and filled grains per panicle. Straw and grain yields were recorded on a plot basis by leaving two border rows. Grain yield is reported at 14 % moisture content.

Plant and soil chemical analyses

Grain and straw were separated, oven dried (at 60 °C for 48 h), ground, and stored in a desiccator. To determine total N, samples were digested in a triacid mixture (H2SO4/HClO4/HNO3 at a ratio of 1:3:10) with CuSO4 and distilled in the semi-micro Kjeldahl apparatus (Jackson 1973). Inorganic N (NH4–N and NO3–N) was determined using the direct distillation method (Keeney and Bremner 1966) from soil samples (0–20 cm depth) collected at 25, 50, 75, and 90 days after transplanting (DAT).

N use efficiency

Taking into account the grain yield and total N uptake, the apparent N recovery (ANR) (Westerman and Kurtz 1974) and agronomic N efficiency (ANE) (Dobermann and Fairhurst 2000) were calculated using the following equations:

$$ \begin{array}{l}\mathrm{ANE}\left(\mathrm{kg}\;\mathrm{grain}/\mathrm{kg}\;\mathrm{N}\;\mathrm{applied}\right)=\left(\mathrm{Yt}-\mathrm{Yo}\right)/\mathrm{Na}\hfill \\ {}\mathrm{ANR}\left(\%\right)=\left[\left(\mathrm{Ut}-\mathrm{Uo}\right)/\times 100\right]/\mathrm{Na}\hfill \end{array} $$

where Yt is the grain yield in the test treatment (kilogram per hectare), Yo is the grain yield in the control plot (kilogram per hectare), Ut is the uptake of N in the test treatment (kilogram per hectare), Uo is the uptake of N in the control plot (kilogram per hectare), and Na is the nitrogen applied.

Statistical analysis

Data were analyzed using SPSS statistical software (ver 11.0; StatSoft, USA). Analysis of variance (ANOVA) was used to determine treatment effects at 5 % confidence level for their significance by Duncan's multiple range test (DMRT) (Cochran and Cox 1992).

Results

Yield and yield components of rice

The yield components, except plant height, were significantly increased by application of VC (Table 1). Split application of VC increased the number of panicles compared with the basal application (T2) with no difference between the different split applications (T3, T4, T5). A similar trend was noted for filled grains per panicle, but there were no significant differences between the VC treatments in the total spikelets per panicle, but all VC treatments were significantly higher than the control (Table 1). The VC-treated plots had significantly higher yield of rice straw than the control plots (Table 2). Application of VC increased the grain yield of rice compared with the control; yield was significantly higher in T5 compared with the basal application, although it was not significantly different from T4 and T3 (Table 2). A four-split application (T5) had 142 and 32 % more grain yield than T1 and T2, respectively.

Table 1 Average yield attributing characters of rice under different treatments during 2 years (2008–2009)
Full size table
Table 2 Average rice straw and grain yields and N use efficiencies under different treatments during 2 years (2008–2009)
Full size table

Nitrogen uptake, apparent N recovery, and agronomic efficiency

As nitrogen uptake by a crop is a function of total biomass production and percent N content in the biomass (straw and grain), N uptake followed a similar pattern to that observed with grain yield (Table 2) with the highest N uptake in T5 (69 kg ha−1). ANE and ANR increased as the number of splits in VC application increased (Table 2). With the basal application of VC, an ANE 22.5 kg of grain per kg of applied nitrogen was found, and this increased maximum ANE of 38.3 kg of grain per kg of applied nitrogen was recorded due to a four-split application of VC (Table 2).

NH4–N concentrations in the soil were higher than NO3–N concentrations in all sampling dates (Table 3). Both forms of mineral N were significantly higher in all the VC-treated plots than in the control. When the N was applied in a basal application of VC (T2), mineral nitrogen was higher at 25 DAT and then declined over the period of time. Where VC was applied as split applications, mineral N content in soil increased up to 50 DAT and declined thereafter, but still maintained higher mineral N concentrations at 75 and 90 DAT than those in T2 (Table 3).

Table 3 Nitrogen (milligram per kilogram) dynamics at different days after transplanting (DAT) due to vermicompost application (2 years pooled data)
Full size table

Discussion

Increasing the availability of N through split application of VC seems to have increased the synchrony of N supply and demand and therefore increased the N uptake (Bhattacharyya et al. 2006; Dobermann and Fairhurst 2000). Following repeated applications of VC, mineral N concentrations in the soil can be maintained through sustained mineralization of organic N (Manivannan and Sriramachandrasekharan 2009; Kale et al. 1992). Lower ANE and ANR with the basal application of VC may be a result of greater nutrient losses or inefficient utilization of nutrient (Bhattacharyya et al. 2006; Kale et al. 1992). Rice plants take up nitrogen in the NH4 form (Myint et al. 2010); when nitrification occurs, the NO3 form of N is more prone to losses by leaching (Dobermann and Fairhurst 2000).

By splitting applications of VC, nutrient supply is better matched with plant demand throughout the crop growth period, increasing the yield components and grain yield (San-oh et al. 2008; Bhattacharyya et al. 2006). In rice, the main yield determinants have been identified as follows: (a) leaf area and shape to maximize photosynthetic efficiency, (b) a well-developed deep root system, (c) high leaf area index at flowering, and (d) higher crop growth rate during PI. Yield components such as the number of productive tillers, spikelets per panicle, and grain weight are also considered to contribute largely to the yield (Bhandari et al. 2002). Rice requires a sustained nutrient supply until its reproductive stage (Murali et al. 2007). Consequently, synchronized nutrient supply increases the efficiency of the translocation of photosynthates to sink (grains), resulting in more filled grains as indicated by the data collected here. Grain weight is a genetically controlled trait; therefore, agronomic manipulation of this component is limited. Crop yield (grain and straw) is a function of several yield determinants. There is some indication of increased rice yield with the split application of vermicompost, though the impact was less than that observed by Dobermann et al. (2000) with split application of N fertilizers.

Conclusion

The nutrient use efficiency (NUE) of organic manures applied in basal dressings may be able to be improved by split applications during the crop-growing period. However, the effectiveness of such approaches will depend on whether the manure can be readily incorporated in the presence of a growing crop and whether it contains some readily available nitrogen which is released immediately on application. More work is needed to develop appropriate locally adapted strategies for nutrient management in organic farming systems. The labor requirements of split applications should also be taken into account in planning any nutrient management strategy. Where high NUE is achieved with high labor requirements, overall system efficiency and profitability may not be improved.