Optimization of liquid bio-fertilizer and coco peat application on nutrient uptake and losses of pak choi (Brassica chinensis L.) via RSM approach

Optimal liquid bio-fertilizer application rate and frequency with coco peat as growth media were estimated using Response Surface Methodology (RSM) to maximize Nitrogen Uptake Efficiency (NUE) and minimize Nitrogen loss (N loss) with favourable Pak Choi yield. A central composite design (CCD) was conducted to determine the experimental design of Pak Choi into 20 treatments based on 5 level codes (-α, -1, 0, 1, α) for three factors, as assigned coco peat (CP) (0%, 25%, 50%, 75%, 100%), bio-fertilizer rate (BFR) (0, 50, 100, 150, 200 kg N/ha) and application frequency (AF) (subsequent 1, 2, 3, 4, 5 day), respectively. The growth performance of plant height, girth, and fresh weight was measured at harvesting day and analysed the nutrients composition. These field data were used to determine the optimum value of plant yield, N loss and NUE by full quadratic polynomial model generated from RSM. The RSM analysis optimum output indicated combination of 25% CP with 150 kg N/ha of BFR at subsequent 4 days that produced the best growth performance on plant height, girth and yield of 11.88 cm, 9.31 cm, and 18.29 kg, respectively. No significant interaction of three factors found on the nutritive compositions (N, P, K, Ca and Mg) by RSM. Within RSM optimization process, an ANOVA output indicated significant terms that fitted two- factor interaction models on yield and NUE as well as a quadratic model on N loss which described their relationships found in this study. These two models enabled to predict response optimization range based on maximizing yield and NUE while minimizing N loss. The models result was predicted Pak Choi maximum yield (≥ 10378.43 kg/ha) were obtained at optimum range: 0 - 20% of CP with 155 - 200 kg N/ha of BFR and AF at subsequent day 4 to day 5. The maximum NUE (≥ 47.75 kg yield/ kg N) were obtained at the range of 0 - 20% CP, 175 - 200 kg N/ha BFR and subsequent 4 to 5 day of AF. While, 0 - 50% of CP, 175 – 200 kg N/ha of BFR and subsequent 3 - 5 day of AF were predicted to minimize N loss (≤ 13.86 kg N/ha). In line with these models optimum output ranges, the models could be used to predict the optimum value by setting the goal of yield, NUE and N loss align with adopting a numerical optimization function. Hence for example, this study was targeted to maximize the NUE (++++), minimize the N loss (++++) and maximize the yield (+++) based on the degree of importance (+, the most “+”, the greater of desire). The output of predication shows the combination of 19% CP, 190 kg N/ha of BFR and AF at subsequent 4 day was given the best optimum value at maximum of NUE of 49.60 kg yield/kg N, with minimum N loss of 1.05 kg N/ha, ultimately achieved better Pak Choi yield of 10739.02 kg/ha. These results were used to conduct a validation experiment on field to validate the model prediction accuracy. Validation result proven a small deviation between the expected and observed value for yield (10725.47 kg/ha), NUE (47.07 kg yield/kg N), and N losses (1.09 kg N/ha) at 0.12%, 5.10%, 3.38%, respectively. This expresses that the models are valid and fit to be used for prediction on Pak Choi yield, NUE and N loss. In conclusion, this study proven that RSM model simulation result shows a promising Pak Choi yield, along with better NUE, while reduces in N losses at the value of CP < 20 %, together with BFR value < 200 kg N/ha at subsequent 4 days of AF. This also indicated that reduced in the cost of fertilizer and significant to environmental friendly (less pollution) by applied lower rate of fertilizer than common farming applied (200 kg N/ha). In addition, farmer able to achieve their targeted yield and reduce the fertilizer cost by using the model obtained from the RSM optimization. The research finding highlighted utmost effect of liquid bio-fertilizer and coco peat on Pak Choi yield, NUE and N loss.

Text FP 2021 47 IR.pdf Download (935kB)

Actions (login required)

View Item