Climate-smart rice (Oryza sativa L.) genotypes identification using stability analysis, multi-trait selection index, and genotype-environment interaction at different irrigation regimes with adaptation to universal warming

Abstract

Climate change has brought an alarming situation in the scarcity of fresh water for irrigation due to the present global water crisis, climate variability, drought, increasing demands of water from the industrial sectors, and contamination of water resources. Accurately evaluating the potential of future rice genotypes in large-scale, multi-environment experiments may be challenging. A key component of the accurate assessment is the examination of stability in growth contexts and genotype-environment interaction. Using a split-plot design with three replications, the study was carried out in nine locations with five genotypes under continuous flooding (CF) and alternate wet and dry (AWD) conditions. Utilizing the web-based warehouse inventory search tool (WIST), the water status was determined. To evaluate yield performance for stability and adaptability, AMMI and GGE biplots were used. The genotypes clearly reacted inversely to the various environments, and substantial interactions were identified. Out of all the environments, G3 (BRRI dhan29) had the greatest grain production, whereas G2 (Binadhan-8) had the lowest. The range between the greatest and lowest mean values of rice grain output (4.95 to 4.62 t ha^-1) was consistent across five distinct rice genotypes. The genotype means varied from 5.03 to 4.73 t ha^-1 depending on the environment. In AWD, all genotypes out performed in the CF system. With just a little interaction effect, the score was almost zero for several genotypes (E1, E2, E6, and E7 for the AWD technique, and E5, E6, E8, and E9 for the CF method) because they performed better in particular settings. The GGE biplot provided more evidence in support of the AMMI study results. The study's findings made it clear that the AMMI model provides a substantial amount of information when evaluating varietal performance across many environments. Out of the five accessions that were analyzed, one was found to be top-ranking by the multi-trait genotype ideotype distance index, meaning that it may be investigated for validation stability measures. The study's findings provide helpful information on the variety selection for the settings in which BRRI dhan47 and BRRI dhan29, respectively, performed effectively in AWD and CF systems. Plant breeders might use this knowledge to choose newer kinds and to design breeding initiatives. In conclusion, intermittent irrigation could be an effective adaptation technique for simultaneously saving water and mitigating GHG while maintaining high rice grain yields in rice cultivation systems.

Keywords: AMMI; GGE biplot model; MGIDI; Rice; Yield; web-based WIST.

Figure 1

The rice grain yield (t ha^-1) performance variation of studied five (5) rice genotypes across nine (9) environments in both AWD = Alternate wet and dry (A, C) and CF = continuous flooding (B, D) conditions.

Figure 2

AMMI1 and AMMI2 biplot for rice grain yield (t ha^-1) of five (5) genotypes under nine (9) diverse environments using GE IPAC scores in both AWD (A, C) and CF (B, D) conditions. G = genotype; E = environment; AMMI = Additive main effects and multiplicative interaction; PC = Principal component; AWD = Alternate wet and dry; CF = Continuous flooding.

Figure 3

AMMI Biplot nominal plot of five (5) rice genotypes with yield and stability in nine (9) different environments in both AWD (A) and CF (B) conditions. G = genotype; E = environment; IPCA 1 = Interaction principal component axis 1; G1 = Binadhan-10, G2 = Binadhan-8, G3 = BRRI dhan29, G4 = BRRI dhan47, G5 = Bina dhan-17.

Figure 4

Polygon of GGE Biplot for clustering environments in both AWD (A) and CF (B) conditions. G = genotype; E = environment; PC = Principal component). G1 = Binadhan-10, G2 = Binadhan-8, G3 = BRRI dhan29, G4 = BRRI dhan47, G5 = Bina dhan-17.

Figure 5

Assessment of five (5) inbred lines of rice according to their yield and stability in nine different environments in both AWD (A) and CF (B) conditions. G = genotype; E = environment; GGE = Genotype by environment biplot; PC = Principal component; G1 = Binadhan-10, G2 = Binadhan-8, G3 = BRRI dhan29, G4 = BRRI dhan47, G5 = Bina dhan-17.

Figure 6

GGE biplot graph based on genotype-focused scaling for comparison of genotypes with an ideal genotype in both AWD (A) and CF (B) conditions. G = genotype; E = environment; PC1 = Interaction principal component axis 1; G1 = Binadhan-10, G2 = Binadhan-8, G3 = BRRI dhan29, G4 = BRRI dhan47, G5 = Bina dhan-17.

Figure 7

GGE biplot graph based on environment-focused scaling for comparison of environments with an ideal environment in both AWD (A) and CF (B) conditions. G = genotype; E = environment; PC1 = Interaction principal component axis 1; G1 = Binadhan-10, G2 = Binadhan-8, G3 = BRRI dhan29, G4 = BRRI dhan47, G5 = Bina dhan-17.

Figure 8

Heatmap-based Speraman’s correlation coefficient among various stability parameters with yield data of 5 rice genotypes evaluated at 9 test environments. Parametric measures such as: ASV–AMMI stability value, RPGV: Relative performance of genotypes values; HMRPGV: Harmonic mean of RPGV, Ecoval: Wricke's ecovalence. Non-parametric measures such as: N1, N2, N3, N4: Thennarasu"s statistics, Sij: Deviations from the joint-regression analysis, S1: mean of the absolute rank differences of a genotype over the n environments, S2: variance among the ranks over the k environments, S3: sum of the absolute deviations, S6: relative sum of squares of rank for each genotype, Gai: Geometric adaptability index, Pi_a, Pi_f, Pi_u: Superiority indexes for all, favorable and unfavorable environments, respectively.

Figure 9

A Genotype ranking in ascending order for the MGIDI index in both AWD (A) and CF (B) conditions. The selected genotypes based on this index are shown in blue. The central blue circle represents the cut point according to the selection pressure. G = genotype; G1 = Binadhan-10, G2 = Binadhan-8, G3 = BRRI dhan29, G4 = BRRI dhan47, G5 = Bina dhan-17.

Figure 10

The strengths and weaknesses view of the selected rice genotype is shown as the proportion of each factor on the computed MGIDI values over two moisture regimes, namely, (A) AWD and (B) CF.

Figure 11

Water status of soil from October to April in Bangladesh. The data were collected from 2010 to 2019 by using the web-based warehouse inventory search tool (WIST).

Figure 12

Temperature, humidity, and rainfall variation during grain filling period in both AWD and CF methods at 9 environments in the rice crop season of 2019–2020.

Figure 13

Comparison of canopy of crop and soil wetness status under two different irrigation regimes i.e., continuous flooding and alternate wet and dry conditions during the rice tillering stage at Nachole during Boro season. (A) Continuous flooding, (B) Alternate wet and dry.