The field trials were executed during 2013–14 and 2014–15 at the Agricultural Research Farm of Banaras Hindu University, Varanasi, India situated at 25°15′19.7″ N latitude, 82°59′34.2″ E longitude, 75 m above mean sea level. The experimental location was characterized by a semi-arid to sub-humid climate with a mean annual rainfall of 1,150 mm and potential evapotranspiration 1525 mm. The weekly weather condition was collected and is presented in Figure 1. During the cropping season, a total of 952.7 mm in 2013–14 and 834.7 mm in 2014–15 rainfall was received with the distribution being normal during the experimental period. The mean weekly maximum temperature was 28.03°C during 2013–14, and 28.65°C during 2014–15. The mean weekly sunshine length varied from 0.6 to 9.1 h in 2013–14 and 1.0 to 9.5 h in 2014–15.

The factors for experimentation were fitted in the split-plot design and replicated thrice. The main plots were allocated to fertilizer doses [60% of RDF (F₁), 80% RDF (F₂), 100% RDF (F₃), and 120% RDF (F₄)] while the panchagavya application {control (D₀); 3% foliar sprays at 15, 30, and 45 days after transplanting (DAT) (D₁); seedling root dip + 3% foliar spray at 30 DAT + application with irrigation water at 60 DAT (D₂); 6% foliar sprays at 15, 30 and 45 DAT (D₃) and seedling root dip + 6% foliar spray at 30 DAT + application with irrigation water at 60 DAT (D₄)}to sub-plots. Rice hybrid PRH 10 - a basmati type was the test crop. The RDF for rice crops was 150, 75, 75, and 5 kg/ha of N, P₂O₅, K₂O, and Zn respectively. One-fourth dose of nitrogen (N) and a full dose of phosphorus (P), potassium (K), and zinc (Zn) were applied 7 days after transplanting and the remaining three-fourths in two splits i.e., two-fourths at tillering stage (30 DAT), and one-fourths at panicle initiation stages (55 DAT).

Seedlings for the system of rice intensification (SRI) were raised in 100 m² area for transplanting in one hectare. Good quality seeds (5 kg ha⁻¹) of the hybrid “PRH 10” were sown by spreading on the nursery bed. For raising nursery, beds of 1.0 m width, 10 m length along with 40 cm furrow were made for seed sowing. Raised nursery beds were made with a combination of FYM and soil at a 1:2 ratio. Carbendazim (@1 g dissolved in one liter water) and streptocycline (@ 6 g in 20 L of water) were used for seed treatment before sowing. To ensure quick and uniform germination, the paddy seeds were drenched in water for 2 days (48 h) and then allowed to shade dry before broadcasting on a raised nursery bed. Well sprouted seeds (20 g m⁻¹) were sown unvaryingly on nursery bed in such a way that every seed had to be isolated from other very easily. Seeds were covered on seedbed with straw of wheat for at least 48 h following sowing to protect them from birds. Watering was done twice a day (early morning and evening). For transplanting, seedlings of 14 days of age (1 seedling hill⁻¹) with a spacing of 25 × 25 cm were used. The net plots (2 rows from all side was removed as a border and 1 row as sampling area) were harvested with sickles and dried on a cemented floor on a sunny day for 5 days and then observation of biomass yield was noted. Grain yield (14% moisture using moisture meter) was noted while completing the process of threshing, cleaning, and drying. The biological yield was calculated using the following formula:

The comprehensive data of all the inputs (panchagavya, seeds, fertilizers, fuel, agro-chemicals, human as well as implements and machine) and outputs (by-product and main) were arranged for the study of energy input and output. Inputs were translated from the physical units to energy units (Yadava et al., 2017) by multiplying with conversion co-efficiency and energy input (Binning et al., 1983) and energy productivity (Tuti et al., 2012) were computed. The energy equivalents (Devasenapathy et al., 2009) of all inputs were added for the estimation of energy inputs (Devasenapathy et al., 2009) to provide an estimate for total energy input (Datta et al., 2014). Similarly, by multiplying the amount of production by its corresponding energy equivalent (Devasenapathy et al., 2009) the energy output (Chaudhary et al., 2006) was calculated. The energy output of the by-product (Mandal et al., 2015) was calculated by multiplying it with its corresponding equivalent (Chaudhary et al., 2009). The other energy parameters were calculated as described below.

Where

The economic crop yield was converted into the equivalent value of carbohydrate (kg ha⁻¹) (Gopalan et al., 2004). The carbon output was calculated based on the biomass of the plant which contains on an average 44% carbon (Lal, 2004).

Plant samples (grain and straw) were dried at 70⁰C for 48 h, grounded with the help of a Macro-wiley mill, and passed through a 40 mesh sieve for determination of N, P, and K content. The total N and P content in the plant was determined by using the Kjeldahl and vanado molybdate method, respectively. The K content of the plant sample was estimated by using a flame photometer. Standard methodologies for the determination of N, P, and K content in rice were used as suggested by (Jackson, 1958).

The following formula as suggested by (Black, 1967) was used for the calculation of nutrient removal by grain and straw of rice:

Soil samples (0–15 cm soil profile) were collected at 30 and 60 DAT from each experimental plot. A soil auger having a diameter of 5 cm was used for the collection of composite soils from four different places. Just after, the dry shade soil samples were grounded with a pestle and mortar. After the removal of inert matter with the help of a 2 mm sieve, the soil samples were stored in a sterile polypropylene bag. Before the analysis of soil microbial biomass carbon (SMBC), enzymatic activities viz. dehydrogenase (DHA), urease (UA), alkaline phosphatase activity (APA), and microbial population (actinomycetes and bacteria) the soil samples were stored at 4°C. The soil microbial population, UA, DHA, APA, and SMBC were estimated as per the method as described by (Chhonkar et al., 2007).

The initial and post-harvest soil samples were collected (0–15 cm soil profile) from four different locations with the help of a core sampler having 5 cm diameter from each experimental plot for chemical analysis. The soil samples were ground with the help of pestle-mortar and sieved in a 2-mm sieve and analyzed for pH (Jackson, 1958) and electrical conductivity (Richards, 1954) following standard analytical procedures (Table 3). The processed samples were also analyzed for their available N content (Subbiah and Asija, 1973), organic carbon (Walkley and Black, 1934), Olsen P (Olsen et al., 1954), available K (Jackson, 1958), and DTPA-extractable Zn (Lindsay and Norvell, 1978) at depths of 0–15 cm.

The data recorded were analyzed using a standard statistical procedure to draw a valid conclusion. For computing, standard ANOVA and comparing treatment means the SPSS 17.0 (SPSS, 2008, New York, USA) statistical package was used. For comparing treatment means, critical difference (CD)/least significant difference (LSD) at p < 0.05 significance was used.

Rice grain yield (mean of 2 years) was significantly (p < 0.05) influenced by including panchagavya with RDF (Table 4). The application of RDF (100%) along with the D₄ level of panchagavya produced an additional (2.12 t ha⁻¹) grain yield over 60% RDF and no panchagavya. Interactions of RDF × panchagavya (p = 0.0001) were significant for grain and biological yield.

The application of 100% RDF (150, 75, 75, and 25 kg N, P₂O₅, K₂O, and ZnSO₄.7H₂O ha⁻¹) recorded significantly higher grain yield, which was 23.8, 6.1, and 3.31% higher over 60, 80, and 120% RDF respectively (Table 5), likewise, panchagavya level D₄ (seedling root dip + 6% spray at 30 DAT + application with irrigation water at 60 DAT) resulted in the highest grain yield (5.93 t ha⁻¹). The interaction effect between RDF and panchagavya was significant and it was found that the F₃ along with D₄ treatments recorded maximum grain yield (6.34 t ha⁻¹) over RDF along with control (4.22 t ha⁻¹).

The higher biological yield (14.02 t ha⁻¹) was recorded with the application of 120% RDF which produced 30.2, 13.8, and 6.1% higher biological yield than 60, 80, and 100% RDF, respectively. Among the panchgavya treatments, the maximum biological yield (13.36 t ha⁻¹) was produced under D₄, which was significantly higher than control, D₁, and D₂, while remaining at par with D₃. The interaction effect of NPK levels and panchagavya on the biological yield of rice indicated that the highest biological yield (15.08 t ha⁻¹) was noticed with the application of 120% RDF with D₄.

Total N, P, and K uptake was significantly higher (p < 0.05) under 120% RDF along with the D₄ level of panchagvya application over 60% RDF without panchagavya (Table 4). An increase of 78.6, 19.0, and 94.39 kg ha⁻¹, respectively have been recorded in the total N, P, and K uptake with the D₄ level of panchagvya application over 60% RDF with no panchagavya after 2 years of experimentation.

120% RDF application observed significantly more N uptake (132.27 kg ha⁻¹) over 60, 80, and 100% RDF. Amongst the panchagavya levels, maximum N uptake (123.25 kg ha⁻¹) by grain + straw was recorded with D₄, followed by D₃. The interaction effect of RDF (120%) and panchagavya (D₄) was also found significant in N uptake (150.1 kg ha⁻¹) by grain + straw.

The significantly higher total P uptake by grain + straw (26.73 kg ha⁻¹) was recorded with 120% RDF application. The total P uptake by grain and straw under different levels of RDF was in the order as 120% RDF > 100% RDF > 80% RDF > 60% RDF. The application of panchagavya caused significant variation in the total P removal by the crop. Significantly higher P removal by rice grain + straw (23.92 kg ha⁻¹) was noted with D₄ than the remaining treatments. The conjoint application RDF (120%) with panchagavya (D₄) exhibited the higher total P removal (31.07 kg ha⁻¹) by the crop (Table 6).

Significantly higher K removal by rice (163.31 kg ha⁻¹) was recorded with RDF (120%) as compared with 60, 80, and 100% RDF, and among the panchagavya treatments, D₄ resulted in the higher total potassium removal by grain + straw (142.58 kg ha⁻¹) while remaining at par with D₃. The highest K removal (185.89) was recorded due to the interaction between the 120% RDF and with D₄, however, the lowest K removal (91.5 kg ha⁻¹) was noticed with 60% RDF without panchagavya (Table 6).

Higher available N (245 kg ha⁻¹), P (26 kg ha⁻¹), and K (205 kg ha⁻¹) contents were observed with 120% RDF in combination with the D₄ level of panchagavya (Figures 2A–C), which was 16.66, 44.44, and 9.1% higher, respectively, over the application of 60% RDF without panchagavya.

The different levels of RDF had a significant result on the soil bacterial population. The maximum bacterial population (51.3 cfu × 10⁵ g⁻¹ soil at 30 DAT and 72.1 cfu × 10⁵ g⁻¹ soil at 60 DAT) was recorded with 120% RDF, which remained on par with 100% RDF, but significantly superior over 60 and 80% RDF. Amongst the panchagavya levels, D₄ noticed a significantly higher population of bacteria (54.8 cfu × 10⁵ g⁻¹ soil at 30 DAT and 79.8 cfu × 10⁵ g⁻¹ soil at 60 DAT) than the rest of the treatments at 30 and 60 DAT. The interaction effect between the application of 120% RDF along with D₄ significantly increased the bacterial population (55.7 at 30 DAT and 86.3 at 60 DAT) over other treatment combinations but remained at par with 100% RDF + D₄ at 30 and 60 DAT (Table 7).

At 30 and 60 DAT, the actinomycetes population was significantly higher with 120% RDF (47.8 cfu × 10⁴ g⁻¹ soil at 30 DAT and 63.0 cfu × 10⁴ g⁻¹ soil at 60 DAT) over 60 and 80% RDF, however, remained on par with 100% RDF. The actinomycetes population at different growth stages (30 and 60 DAT) under D₄ remained 55.1 cfu × 10⁴ g⁻¹ soil and 67.3 cfu × 10⁴ g⁻¹ soil at 30 and 60 DAT, respectively over the rest of the treatments. The interaction effect of the application of 120% RDF with D₄ treatment significantly increased the actinomycetes population as 58.4 cfu × 10⁴ g⁻¹ soil and 74.9 cfu × 10⁴ g⁻¹ soil at 30 and 60 DAT, respectively, but remained at par with 100% RDF at D₄ (Table 7).

The interaction effect of 120% RDF with D₄ significantly increased the SMBC (135.8 and 199.2 μg C g⁻¹ soil at 30 DAT and 60 DAT, respectively) over other treatment combinations but remained on par with 100% RDF under D₄ (Table 8).

On the other side, 100% RDF and D₄ exhibited significantly higher dehydrogenase activity (DHA) (139.6 and 158.2 μg TPF /g soil/24 h at 30 and 60 DAT, respectively), followed by 100% RDF with D₃. Among the other interactions, the application of 120% RDF with D₄ significantly increased the alkaline phosphatase activity (83.8 and 97.7 μg p-NP g⁻¹ soil h⁻¹ at 30 DAT and 60 DAT, respectively) (Table 9). Further, 120% RDF and D₄ registered significantly higher urease activity (292 and 343.7 μg UH g⁻¹ soil h⁻¹ at 30 and 60 DAT, respectively) (Table 9).

Based on the energy equivalent (Table 10) parameters, like energy output (187,867 MJ ha⁻¹), net energy returns (164,319 MJ ha⁻¹), and energy intensity in economic terms (5.08 MJ ) were calculated and found significantly higher with 120% RDF than control (Table 11), while energy ratio (8.68), energy productivity (0.922 kg MJ⁻¹), and energy profitability (7.68) were highest with 80% RDF followed by 60 and 100%, respectively.

Significantly higher energy output (180,004 MJ ha⁻¹) and net energy returns (159,746 MJ ha⁻¹) were recorded under D₄, D₃ remaining at par with it. But, the energy ratio (8.91), energy productivity (0.293 kg MJ⁻¹), energy intensity in economic terms (4.97 MJ ), and energy profitability were significantly highest with D₄. Among the fertilizer doses and panchagavya applications, the highest energy input was recorded with F₄ (23,548 MJ ha⁻¹) and D₃ (20,259 MJ ha⁻¹), respectively.

The carbohydrate equivalent yield (4,641 kg ha⁻¹) was recorded highest with 100% RDF, however, carbon output was recorded higher (5,608 kg CO₂ eq ha⁻¹) with 120% RDF (Table 11). Among panchagavya applications, D₄ resulted in the highest carbohydrate equivalent yield (4,611 kg ha⁻¹) and carbon output (5,345 kg CO₂ eq ha⁻¹) over control.

The application of panchagavya increased rice yield during both years of experimentation. Panchagavya contains an appreciable amount of IAA and GA₃ (Somasundaram et al., 2003), when sprayed twice along with root dip stimulates plants for increased production of growth regulators in the cell system (Yadav and Lourduraj, 2006). Besides, the beneficial effects of panchagavya could also be attributed to its micronutrient content, higher biological activity, and plant growth-promoting substances (Yadav and Lourduraj, 2006). Improvements in the yield attributes of sunflower, maize, green gram, French bean, and okra have already been reported for a foliar spray of panchagavya (Boomiraj, 2003; Somasundaram et al., 2003; Selvaraj et al., 2007). Panchagavya in different formulations as well as in combinations with various fertility levels brought about significant improvement in both grain and straw yields of rice. D₄ treatment (seedling root dip + 6% spray at 30 DAT + application with water at 60 DAT) along with 100% RDF produced higher grain yield and straw yield 60% RDF plus D₄ (Table 4). All the vegetative and reproductive characters followed a similar trend.

The performance of any crop in terms of yield depends entirely on its genetic ability (Elizabeth et al., 2007) and the capacity to assimilate the applied nutrients (Chen and Liao, 2017). Integration of an organic nutrient source with chemical fertilizers improves soil ambiance for optimum plant growth (Upadhyay et al., 2018). Thus, the application of panchagavya improves soil health and quality, thereby facilitating plants to extract more nutrients from the soil for transformation in the sink. As a result, a 23.7% yield increase was noticed over control due to the panchagavya application. The stimulatory effect in plant growth due to the release of phytohormones like IAA, GA₃, cytokinin with the use of panchagavya in rice (Xu, 2001; Somasundaram and Amanullah, 2007) and several other crops (have been reported in several studies in the past also (Hossain et al., 2007).

The combination of panchagavya and RDF followed quadratic response function for rice grain yield with the successive increment in RDF up to 100% (150, 75, 75, and 5.25 of N, P₂O₅, K₂O, and Zn kg ha⁻¹, respectively) in the present study (Upadhyay et al., 2019), a positive response in seed yield. The optimum economic doses of fertilizer estimated for highest production were 107.3 kg N, 53.7 kg P₂O₅, and K₂O, and 3.8 kg Zn per ha along with panchagavya. The enhanced microbial count, higher microbial biomass carbon, and improved microbial activities were recorded under D₄ and higher soil availability of applied nutrients for better uptake and more efficient assimilation along with F₃.

A linear relationship was recorded with panchagavya application between RDF and straw yield (Upadhyay et al., 2019). The higher vegetative growth and tillers with 120% RDF and panchagavya application (D₄) led to higher straw yield, however, the harvest index remained highest under 100% RDF due to more effective tillers.

The application of graded levels of panchagavya enhanced the N, P, and K concentrations in both grain and straw. The application of panchagavya viz. D₄ (seedling root dip + 6% spray at 30 DAT + application with irrigation water at 60 DAT) was found significantly superior to control and D₁ for total NPK removal. In addition to major (N, P, and K and calcium) and minor (zinc, iron, copper, and magnesium) nutrients, a high amount of total reducing sugars was also recorded in panchagavya (Yadav and Lourduraj, 2006). Yadav and Kumar (2009) have also observed that various ammonifiers and nitrifiers in the form of chemo-lithotrops and autotrophic nitrifiers in panchagavya inhabit the leaves and increase the ammonia uptake and enhanced total N supply. N supply through panchagavya and recommended fertilizer nutrients ensure a steady and continuous supply of N in the plant. These results are in agreement with the views of Singh et al. (2019). The higher nutrient removal by plants with panchagavya is corroborated with prolonged nutrient availability in soil (Vasanthi and Kumuraswamy, 1999).

Soil available residual N significantly increased by different levels of panchagavya with D₄ exhibiting the highest soil available N. In the present study, D₄ along with 120% RDF maintained higher N availability in the soil (Figure 2). A steady decomposition of inherent organic nutrients from panchagavya containing chemolithotrops and autotrophic nitrifiers might have enhanced the total N supply in the soil (Papen et al., 2002; Yadav and Kumar, 2009).

Similarly, a higher available soil P was observed with the combined use of panchagavya (D₄) and RDF. Various organic acids released by the micro-organisms from panchagavya favor the solubilization of insoluble soil phosphate. The dominant ionic organic compounds in panchagavya compete with the phosphate ion adsorbed by colloidal sites (Pavinato et al., 2008) and clay mineral lattice (Fink et al., 2016) reduce phosphate-fixation in soil. Organic nutrient sources when applied in combination with inorganic fertilizer enhance labile P in the soil by forming a complex with cations like Ca²⁺ and Mg²⁺ (Urkurkar et al., 2010). The maximum potassium status in soil was also observed in D₄ along with 120% RDF. The application of panchagavya through seedling root dip reduced K-fixation and enhanced K concentration due to positive organic matter and clay interface (Prasad et al., 1997).

The highest microbial activity in terms of bacterial and actinomycetes population was recorded with D₄ (Table 6). Panchagavya has abundant N-fixers and P-solubilizers (Sreenivasa et al., 2009) which produce various useful metabolites like, organic acids, and antibiotics (Sangeetha and Thevanathan, 2010). The optimum microbial activity lowers the electrical conductivity of panchagavya and promotes the uptake of salts and ions by soil microbes. All panchagavya preparations increased soil microbial populations and plant growth due to the presence of beneficial rhizospheric microbes in several crops (Radha and Rao, 2014; Shubha et al., 2014). The fluorescent pseudomonas in panchagavya (Bhat et al., 2005) facilitate the synthesis of phytohormones and several other growth-promoting amalgams (Vennila and Jayanthi, 2008). Xu (2001) has also concluded that the soil quality by the application of panchagavya could increase the growth and productivity of crops due to the presence of several beneficial microbes in the rhizosphere.

A significant effect on SMBC was observed with the application of different levels and methods of panchagavya. The highest MBC was recorded with D₄ (seedling root dip + 6% spray at 30 DAT + application with irrigation water at 60 DAT), but it remained at par with D₃ (three sprays at 15, 30, and 45 DAT @ 6%) and D₂ (seedling root dip + 3% spray at 30 DAT + application with irrigation water at 60 DAT). Panchagavya acts as a fine substrate for microbial growth and thus increased MBC (Fraser et al., 1994; Cerny et al., 2008).

Likewise, urease, a vital extracellular enzyme for catalyzing the hydrolysis of urea to ammonia (NH₃) and subsequently transforming it to ammonium (NH4+ and nitrate (NO3-) ions (Byrnes and Amberger, 1989). Its activity was recorded higher in D₄ in which also improved the nitrogen fertilizer use efficiency.

The dehydrogenase activity was evaluated and was found to be consistently influenced by panchagavay. D₄ (seedling root dip + 6% spray at 30 DAT + application with irrigation water at 60 DAT) treatment resulted in the highest dehydrogenase activity. The lowest activity was registered under control among all other treatments indicated the beneficial effect of panchagavya on enzymatic activity in the soil.

Dehydrogenase is a vital enzyme in all viable microbes and its activity is a measure of their vigorous metabolic state (Watts et al., 2010). Dehydrogenase activity (DHA) is thus an important bio-indicators for soil fertility (Wolinska and Stepniewska, 2012). Its activity depends on the microorganisms' abundance and dynamism (Järvan et al., 2014).

Panchagavya as applied in D₄ caused a significant improvement in alkaline phosphatase activity which plays a critical role in P cycles (Speir and Ross, 1978) in soil ecosystems. The application of panchagvya also helps in providing P to the crop as it contains a good amount of P. The combined applications of 100% RDF and D₄ registered maximum DHA which was higher than the only application of either 100% RDF or D₄. The significant interaction of RDF and panchagavya were found in enhancing the activity of urease and alkaline phosphatase.

The energy inflow was primarily influenced by both fertilizer and panchgavya, but more so with panchagavya. Among energy sources, fertilization accounts for the major share of the energy input that directly influences the system's net energy output and energy efficiency (Devasenapathy et al., 2009). The application of panchgavya helps in improving the system's net energy returns and system energy ratio. The decrease in energy input is primarily achieved by avoiding mineral fertilization which is the main component of energy consumption in a conventional production system (Pratibha et al., 2015). The system sustainability and resilience can be further improved with agronomic interventions on crop/cultivars in combination with nutrient management practices.