Analysing water productivity response to sowing window, irrigation levels and mulching using CERES-wheat model

Field experiments were carried out during rabi seasons of 2015-16 and 2016-17 at the Research Farm, Punjab Agricultural University, Ludhiana. Wheat variety PBW 621 was sown on three dates (D 1 : 4 th week of October, D 2 : 2 nd week of November and D 3 : 4 th week of November) with two irrigation levels (I 1 : IW/ CPE = 0.9, I 2 : At CRI, 5-6 weeks after 1 st irrigation, 3-4/5-6 weeks after 2 nd irrigation, 2/4 weeks after 3 rd irrigation as per dates of sowing) and mulch application (M 1 : without mulch, M 2 : straw mulch @ 5 t ha -1 ). Earlier sown mulch applied crop with four post-sowing irrigations produced highest (5312.5 kg ha -1 ) and late sown without mulch application crop with irrigation @IW/CPE = 0.9 produced lowest grain yield (3900.5 kg ha -1 ). Simulation results depicted -1.1 to 16.8 per cent deviation in crop yield, -1.4 to -21.0 per cent in water use and 12.7 to 45.5 per cent in water productivity. Increase in temperature from 1 o C to 3 o C decreased wheat yield by 6.3 to 27.0 per cent under D 1 and 3.3 to 17.6 per cent under D 2 , however, it increased from 8.1 to 16.2 per cent under D 3 , indicating D 3 as most appropriate under future warming scenarios. Increase in CO 2 concentration decreased water use and increased yield and water productivity.

farmers have increased. (Humphreys et al., 2010). Punjab is facing dual challenge of weather variability and overexploitation of its ground water resources with significant impact on crop productivity. Keeping this in view, the present study was conducted to simulate crop water requirement, grain yield and water productivity of wheat to variable temperature and CO 2 levels under different dates of sowing, irrigation levels and mulch application.

MATERIALS AND METHODS
The present investigation was carried out at the Research Farm of Punjab Agricultural University, Ludhiana situated at latitude of 30°54›N, longitude of 75°54›E and altitude of 247m above the sea level. Wheat variety PBW 621 was sown on three dates (D 1 : 4 th week of October, D 2 : 2 nd week of November and D 3 : 4 th week of November) with two irrigation levels (I 1 : IW/CPE = 0.9, I 2 : Irrigation at CRI stage, 5-6 weeks after 1 st irrigation, 3-4/5-6 weeks after 2 nd irrigation, 2/4 weeks after 3 rd irrigation as per date of sowing) and mulch application (M 1 : without mulch, M 2 : straw mulch @ 5 t ha -1 ). The experiment was laid out in a split plot design with dates of sowing in the main plots and irrigation and mulch application in combination as sub-plots. The soil of the experimental site was sandy loam. The soil properties used in the CERES-wheat model have been depicted in Table 1.
Crop yield under different treatments was recorded at the time of harvesting. Crop water use was recorded by periodic water depletion method and water productivity was computed from evapotranspiration and crop yield as mentioned below: WP= Water productivity (kg ha -1 mm of water); Y = marketable yield (kg ha -1 ) ; ET = E v a p o t r a n s p i r a t i o n (mm) Crop yield, crop water requirements and water productivity were simulated using CERES-wheat model. The genetic coefficients used for wheat variety PBW-621 have been given in Table 2. Observed and simulated yield, water use and water productivity were then analysed statistically by computing standard deviation, correlation and root mean squared error. Sensitivity analysis was carried out to evaluate water productivity of wheat under variable temperatures (+1 o C, +2 o C and +3 o C) and CO 2 levels (+200 ppm, +400 ppm and +600 ppm) for different treatments.

Wheat yield
The results revealed that early sown mulch applied crop with four post-sowing irrigations (D 1 I 2 M 2 ) produced highest grain yield (5313 kg ha -1 ) followed by (5096 kg ha -1 ) mid November sown mulch applied crop (D 2 I 2 M 2 ), whereas the lowest yield (3901 kg ha -1 ) was observed in the late sown without mulch application crop with irrigation @IW/CPE = 0.9 (D 3 I 1 M 1 ) ( Table  3). The simulation analysis also depicted highest yield (5257 kg ha -1 ) under earlier sown mulch applied crop with four recommended post-sowing irrigations (D 1 I 2 M 2 ) followed by (5186 kg ha -1 ) under mid November sown mulch applied crop (D 2 I 2 M 2 ). Although less crop yield was observed under late sown conditions, but simulation analysis depicted the scope of improvement in wheat yield (5091 kg ha -1 ) with irrigation management and mulch application (D 3 I 2 M 2 ). Simulation results depicted deviation in crop yield in the range of -1.1 to 16.8 per cent in different treatments.

Crop water use
Among different treatments, crop water depletion was observed to be highest (408 mm) under earlier sown without mulch applied crop with four recommended post-sowing irrigations (D 1 I 2 M 1 ) ( Table 3) as compared with rest of the treatments. In general, water depletion decreased with delay in sowing, which might be due to reduction in crop duration under late sown conditions. However, irrigation management and mulch application seemed to be quite effective in checking water depletion under all the dates of sowing. Simulation analysis also depicted highest crop water use (362 mm) under earlier sown without mulch crop with four recommended postsowing irrigations (D 1 I 2 M 1 ) and lowest (290 mm) under late November sown mulch applied crop with four post-sowing irrigations (D 3 I 2 M 2 ). Simulation analysis under-estimated crop water use in all the treatments, the deviation ranging from -1.4 to -21.0 per cent.

Water productivity
Late October and mid November sown mulch applied crop with four post-sowing irrigations (D 2 I 2 M 2 and D 1 I 2 M 2 ) reported highest water productivity (13.5 and 13.4 kg ha -1 mm -1 , respectively) followed by earlier sown mulch applied crop (13.0 kg ha -1 mm -1 ) with irrigation application @IW/CPE = 0.9 (D 1 I 1 M 2 ). Among all the dates of sowing and irrigation levels, mulch application improved crop water productivity. Simulation analysis also depicted highest water productivity (17.6 and 16.7 kg ha -1 mm -1 ) under late and mid November sown mulch applied crop with four post-sowing irrigations (D 3 I 2 M 2 and D 2 I 2 M 2 , respectively). Over-estimation of simulated water productivity was observed, the deviation being in the range of 12.8 -45.0 per cent under different treatments. However, the deviation ranged from 14.5 -16.2 per cent for late October sown crop, 12.7 -23.7 per cent for mid November and 25.5 -45.5 per cent for end November sown crop, depicting ample scope of improvements in wheat productivity with microclimatic modifications especially under late sown conditions (Table 3).

Relation between observed and simulated parameters
A good correlation was found in observed and simulated yield. Average wheat yield of 4552.4+390.3 and 4718.3+445.8 kg ha -1 was observed during 2015-16 and 2016-17 seasons, respectively, whereas simulated yield for the corresponding years was 4721.9+237.8 and 5185.0+246.7 kg ha -1 , respectively with correlation coefficient of 0.84 and 0.81 and RMSE of 277.9 and 540.0

Effect of increased temperature
Increase in temperature from 1 o C to 3 o C decreased wheat yield by -6.3 to -27.0 per cent under D 1 and -3.3 to -17.6 per cent under D 2 , however, increase in yield from 8.1 to 16.2 per cent was observed under D 3 , which indicates that sowing of wheat during 4 th week of November is most appropriate sowing window under future warming scenarios (Fig. 1a). Similarly, increase in temperature resulted in reduced grain yield under irrigation and mulch treatments. Simulation results also depicted decrease in water uptake by the crop with increase in temperature under all the dates of sowing, which might be due to enhanced maturity under warming scenarios. I 2 irrigation regime and mulch application resulted in reduced water depletion, thus improving crop water productivity. The simulation analysis also revealed that under I 1 and M 1 treatments crop water productivity decreased beyond 2 o C increase in temperature, however, with irrigation management (I 2 ) and mulch application (M 2 ), the crop water productivity could be managed even upto 3 o C increase in temperature.
A critical appraisal of simulation analyses clearly depicted the effect of sowing time, irrigation and mulching in sustaining wheat productivity under warming scenarios. The analysis indicated that with increase in temperature by 3 o C, the wheat yield could be improved from 3614 kg ha -1 to 4568 kg ha -1 by shifting sowing time from 4 th week of October to 4 th week of November as crop water requirement decreases and water productivity increases. As in the 4 th November sown crop, water depletion decreased by 5-10 per cent and water productivity improved by 15-30 per cent with increase in temperature by 1 to 3 o C (Fig.1a).

Effect of increased CO 2 concentration
Increase in CO 2 concentration from 200 to 600 ppm increased grain yield under all the treatments, in the range of 21.7 to 49.0%, 26.4 to 51.5 per cent and 26.3 to 57.7 per cent under D 1 , D 2 and D 3 , respectively. Among the irrigation levels, increase in yield was 27.7 to 53.4 per cent and 22.0 to 43.4 per cent under I 1 and I 2 , respectively and under mulch treatment, it was improved to the tune of 26.2 to 50.3 per cent without mulch application and 22.9 to 45.8 per cent under mulch (Fig. 1b). Simulation results also depicted decrease in water depletion and increase in water productivity with increase in CO 2 concentration in all the treatments. Among the dates of sowing, the crop sown in fourth week of November depicted highest decrease in water depletion ranging from 13.7 to 14.0 per cent. Similarly, irrigation management and mulch application significantly reduced water depletion and improved water productivity. I 2 irrigation level decreased water depletion by 14.8 to 15.4 per cent and improved water productivity by 43.7 to 69.8 per cent. Similarly, mulch application reduced water depletion by 14.2 to 14.8 per cent and improved water productivity by 43.4 to 71.3 per cent with increased CO 2 concentration from 200 to 600 ppm. Among all the treatments, mulch application depicted maximum water productivity (22.1 kg ha -1 mm -1 ) with increased CO 2 concentration upto 600 ppm.

Interaction effect of increased temperature and CO 2
Analysis of interactive effect of temperature and CO 2 had revealed that reduction in yield and water productivity due to increase in temperature could be negated by the positive effect of increase in CO 2 concentration. Almost comparable grain yield was observed at ambient temperature with increase in CO 2 by 200 ppm (5780 kg ha -1 ), with 2 o C increase in temperature and CO 2 concentration increment by 400 ppm (5758 kg ha -1 ) and 3 o C increase in temperature with increase in CO 2 by 600 ppm (5784 kg ha -1 ) (Fig. 2a&b). Similar effect was observed for water productivity at ambient temperature with 200 ppm increase in CO 2 concentration, 1 o C increase in temperature with CO 2 increment by 400 ppm and 2-3 o C increase in temperature with CO 2 concentration increment of 600 ppm (18 kg ha -1 mm -1 each). These results indicate that increase in CO 2 concentration by 600 ppm can counter balance the negative effect of global warming scenarios by 2-3 o C on wheat grain yield and water productivity.

CONCLUSION
The study concluded that adoption of appropriate sowing time, irrigation management and mulch application have a significant effect on sustaining crop yield and water productivity even under late sown conditions in wheat. Simulation analysis indicated sowing of wheat during fourth week of November to be most appropriate in view of future global warming scenarios. Sensitivity analysis indicated that increase in CO 2 concentration by 600 ppm can counterbalance the negative effect of global warming scenarios by 2-3 o C on wheat grain yield and water productivity in the region.

Conflict of Interest Statement :
The author(s) declare (s) that there is no conflict of interest.

Disclaimer :
The contents, opinions, and views expressed in the research article published in the Journal of Agrometeorology are the views of the authors and do not necessarily reflect the views of the organizations they belong to.
Publisher's Note : The periodical remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.