Indica-Nipponbare-Takanari (China, Japan, and Thailand)

Experimental treatments and design

All experiments at above experiment sites were conducted in 2001-02, except in Taoyuan, where they were done in 2002-03. Three rice varieties, representing major genotypes, were grown at all sites: indica variety IR72, japonica variety Nipponbare, and indica × japonica crossbred Takanari (only IR72 was not grown at Taoyuan in 2003).

All experiments were laid out in a randomized block design with three replicates and a plot size of 20 m2. At transplanting, 40 kg N ha–1 was applied, and 20 kg N ha–1 was topdressed at 20-day intervals, starting from 20 days after transplanting until 10 days after heading. Total N application varied from 100 to 160 kg ha–1, depending on actual growth duration of the crops. Control plots (0 N) were used that received no N fertilizer and were planted to IR72 only. All plots received 120 kg P2O5 and 120 kg K2O ha–1 as basal dressing.

Cultural practices

Rice seedlings with 4–5 leaves were transplanted at 15 × 30-cm spacing, with one seedling per hill in 2001 and two seedlings per hill in 2002-03. All fields were continuously submerged throughout the growing season. Weeds, pests, and diseases were adequately controlled by pesticides, fungicides, herbicides, and manual weeding.

Plant sampling and data collection

Dates of sowing, emergence, transplanting, panicle initiation, flowering, and maturity were recorded. For each variety, 20 plants were harvested at transplanting, and eight hills at 20 days after transplanting, panicle initiation, 2 weeks before and after flowering, and maturity. Green leaf area was measured and leaf area index (LAI) calculated. Weights of green leaves, dead/yellow leaves, stems (including leaf sheaths), and panicles were determined after oven-drying at 80°C till constant weight. N content of the plant organs was determined by near-infrared spectroscopic analysis (BRAN+LUEBBE, Infra-Alyzer500 equipped with IDAS software) and calibrated by the value obtained by the Kjeldahl method (except at Taoyuan in 2003). At maturity, grain yield was measured from 2 m2 and expressed at 14% moisture content.  Because of experimental errors, yields were not available for Nipponbare at Ubon in 2001-02, for IR72 at Iwate in 2001, and for all three varieties at Nagano in 2002. Final crop biomass was not available for Nipponbare at Ubon in 2001-02, and for IR72 at Iwate in 2001. Total N uptake was not available for any variety at Nagano in 2001-02, for Nipponbare at Iwate in 2001, and for Takanari at Ubon in 2002.

N uptake by the crop and its components were calculated from dry weight and N contents. Internal N-use efficiency (INUE) was calculated by dividing grain yield by total crop N uptake (following Witt et al 1999).

Climate data collection

Throughout the growing season, daily maximum and minimum temperature and solar radiation were recorded at weather stations installed at or near the sites.

Data analysis

GenStat for Windows, 8th Edition (www.vsn-intl.com/genstat/), was used in the analysis of variance of yield, N uptake, and INUE, using the method of residual maximum likelihood (REML; Welham and Thompson 1997).

We calculated the SE and CV of measured crop variables as averages over all experiments at all sites (per experiment, SE and CV are calculated from the three replicates) (Table 1).

Table 1. Standard deviation and error (SD and SE, the same unit as variable) and coefficient of variation (CV) for measured crop variables in the field experiments. N is the number of data in experiments.

Simulations and parameterisation

Following the procedure introduced in Section 2, we parameterized ORYZA2000 for each variety, using the experimental data from all sites in 2002 (calibration set) for special varietal parameters. All other common crop parameters were the same as in the original IR72 crop data file reported by Bouman et al (2001).

For each site, indigenous soil N supply was first estimated from N uptake by the crop in the N-fertilizer omission plots, and subsequently fine-tuned by matching simulated and measured values of crop N uptake.  Since N uptake in omission plots was measured only at Shimane, Kyoto, Ubon, Taoyuan, and Kyoto, we used the estimated values at Kyoto also for Iwate and Nagano, and the estimated values at Ubon for Chiangmai. Indigenous soil N supply at Nanjing was derived from a nearby experiment (Jing et al 2005, 2007).

Model evaluation

The performance of ORYZA2000 was evaluated separately for the calibration set of 2002 and for the validation data sets of 2001 and 2003 (for Taoyuan) following procedures developed by Bouman and van Laar (2006) and described in Section 2.5.

Variation in environmental conditions

Temperatures during the rice-growing periods at the different sites (Fig. 1) were relatively low at the beginning and end of the growing season at Iwate, Nagano, Kyoto, Shimane, and Nanjing, while they were relatively constant at Taoyuan, Chiangmai, and Ubon.  Temperatures were lowest at Iwate, with minima below 20°C and maxima not exceeding 30°C.  At Chiangmai and Ubon, minimum and maximum temperatures were above 20 and 30°C, respectively. Radiation during the rice-growing periods varied among locations (Fig. 2), with the lowest average values (14 MJ m–2 d–1) at Iwate and the highest values at Nagano (18 MJ m–2 d–1) and Taoyuan (17 MJ m–2 d–1).

Indigenous soil N supply was 0.29 kg ha–1 d–1 at Ubon, 0.79 at Shimane, 0.90 at Nanjing, 0.92 at Kyoto, and 1.46 at Taoyuan.

Average yields were lowest for Nipponbare and highest for Takanari (Table 2). Average yields of the three varieties ranged from 1,883 kg ha–1 at Ubon to 10,224 kg ha–1 at Taoyuan, illustrating the strong effect of environmental conditions on yield. The seasonal variation and the interaction of season by variety on yield were not significant, whereas the interaction of site by variety was significant. The differences in yield among the three varieties were reflected in the differences in N uptake and INUE. For these three variables, the highest values were obtained for Takanari and the lowest for Nipponbare (though not all differences among varieties were significant). The differences in N uptake and INUE among the sites again illustrate the significant effect of environmental conditions. The causes of variation for N uptake and INUE were the same as for yield.

Table 2. The means of yield, total N uptake, and internal N-use efficiency (INUE) of three cultivars and the source of variation at eight locations in Asia in two years.

Biomass

Simulated biomass of the whole crop and of the panicles during the growing season agreed well with the measured values in both the calibration and validation set (Fig. 3). Simulated values were also similar to measured values at the 95% confidence level by Student’s t-test (Table 3). The slope was usually close to 1 and the intercept was small in comparison with the measured mean values. RMSEn was 16% in the calibration set and 20–24% in the validation set. RMSEn for simulated crop biomass was 2.3 and 2.9 times the CV of the measurements in the calibration set and the validation set, respectively (compare Table 1). For panicle biomass, these values were 1.8 and 2.6, respectively.

Nitrogen uptake

The simulated and measured values of N uptake by the crop and by the panicles during the growing season were aligned well along the 1:1 line in both the calibration and the validation set (Fig. 3). Student’s t-test showed significant correlations between simulated and measured values, and the slopes of the linear relations were close to 1, with low values and high coefficients of correlation (Table 3).  RMSEnvalues were of the same order of magnitude as those for the biomass simulations (see above). RMSEn for simulated crop N uptake was 2 and 2.7 times the CV in the calibration set and the validation set, respectively.  For panicle N uptake, these values were 2.5 and 3.2, respectively.

Yield, final crop biomass, and nitrogen uptake

The simulated and measured values of yield, final crop biomass (at harvest), and final crop N uptake were well aligned around the 1:1 line in both the calibration and validation set (Fig. 4). In Ubon, both final crop biomass and yield were consistently overestimated by the simulations for all three varieties. The goodness-of-fit parameters confirm the satisfactory performance of the model for all three end-of-season variables (Table 3). RMSEn was 6–11% in the calibration set and 11–16% in the validation set.  Despite these low values, RMSEn for simulated yield, final crop biomass, and final crop N uptake was still 2.3, 1.5, and 0.9 times the CV in the calibration set, respectively, and 3.2, 1.9, and 1.9 times the CV in the validation set.

References

Bouman, B.A.M., & Van Laar, H. H. 2006. Description and evaluation of the rice growth model ORYZA2000 under nitrogen-limited conditions. Agricultural Systems 87, 249-273.     

Bouman BAM, Kropff MJ, Tuong TP, Woperies MCS, ten Berge HFM, van Laar HH. 2001. ORYZA2000: modeling lowland rice. Los Baños (Philippines): International Rice Research Institute, and Wageningen (Netherlands): Wageningen University and Research Centre. 235 p.

Jing Q, Bouman BAM, Hengsdijk H, Van Keulen H, Cao W. 2007. Exploring options to combine high yields with high nitrogen use efficiencies in irrigated rice in China. Eur. J. Agron. 26:166-177.

Jing Q, Van Keulen H, Cao W, Hengsdijk H, Dai T, Jiang D, You X. 2005. Quantifying N response and N use efficiency in rice-wheat (RW) cropping systems with different water management. In: Zhu Z, Minami K, Xing G, editors. The 3rd International Nitrogen Conference. Nanjing (China): Science Press USA Inc. p 81-89.

Van Keulen H. 1982. Graphical analysis of annual crop response to fertilizer application. Agric. Syst. 9:113-126.

Welham SJ, Thompson R. 1997. Likelihood ratio tests for fixed model terms using residual maximum likelihood. J. Royal Stat. Soc. Ser. B (Stat. Method.) 59:701-714.

Witt C, Dobermann A, Abdulrachman S, Gines H, Wang G, Nagarajan R, Satawatananont S, Son T, Tan P, Tiem L, Simbahan G, Olk D. 1999. Internal nutrient efficiencies of irrigated lowland rice in tropical and subtropical Asia. Field Crops Res. 63:113-138.