Hydroponic Salinity Screening by Deep Flow Technique on All Paddy Growing Phases

Salinity screening under hydroponic Deep Flow Technique (DFT) has not been widely studied, especially on the nature of rice tolerance to salinity stress. According to previous screening studies, this method was effective in distinguishing the nature of rice tolerance to salinity stress. However, they were tested only at generative phase. Therefore, evaluation on screening method with hydroponic DFT at all phases of paddy growth is essential. The objective of this study is to evaluate the filtering under hydroponic DFT at all paddy phase and to determine secondary character that support productivity which can be utilized as selection character in this screening process. The experiment was arranged with a complete nested group design with nested replication is the NaCl stress. There were 5 (Five) tested rice varieties and the stress environment consisted of three levels: 0, 60, and 120 mM of NaCl, all with 3 (three) replications. The nutrient culture screening was adapted to the modified Egdane method. The results showed that screening under hydroponic DFT was effective at the concentration of 60 nM of NaCl. The best selection character was yields per clump, number of productive tillers and total chlorophyll. The variety of Jeliteng, Ciherang and Inpari 34 of Salin Agritan, were classified as tolerant group. This hydroponic DFT Filtering method could be recommended as one salinity screening method for all paddy growing phases


Research Design and Procedures
The experiment was arranged in a nested and randomized complete group design, where the nested replication was on the NaCl environment. The tested rice varieties were Inpari 34 Salin Agritan, Ciherang, IR 29, Inpari 29, and Jeliteng. The salinity environments in the experiment were 0 mM of NaCl (normal), 60 mM of NaCl and 120 mM of NaCl.
The seedling was grown on rockwool media for 15 days then transferred to a DFT hydroponics system running on PVC pipes. Each pipe has holes with a diameter of 5 cm each and the distance between the holes was 20 cm. The seedlings on the rockwool were placed in a netpot container by ensuring the seedling's root are in contact with the nutrient culture. The nutrient culture media was AB mix with a concentration of 5 ml per liter of water. The volume of nutrients in each DFT installation was 120 L. Induction of salinity stress was applied at the age of 7 days in the DFT. Stress induction was applied gradually to avoid osmotic shock. The first application of NaCl was 50% of the prescribed concentration, it is then increased to 60 mM and 120 mM (as the treatments) in the following three days. The stress condition was sustained for 14 days and then transferred to normal environmental conditions. The nutrient solution was replaced after 7 days under normal conditions. Administration of HCL or NaOH was to maintain the pH of the solution in the range of 5.5-6.5.
The observed characters of this experiment were plant height, root length, number of leaves, number of tillers, number of productive tillers, flowering age, chlorophyll A, chlorophyll B, total chlorophyll, stomata density, wet weight of shoots, dry weight of shoots, dry weight of roots, length of panicles, length of flag leaves, percentage of filled seeds, percentage of hollow seeds, weight of 100 seeds, and production per clump.

Data Analysis
The significant different characters in the interactions of each character under the variance analysis were further analyzed with pearson correlation tests, cross-prints, and key component analysis based on the value of each character's stress tolerance index. Heatmap cluster analysis was further performed to visualize kinship patterns of complex variable whish is more simple through color gradation.
The Index of Stress Tolerance was calculated by equation (Fernandez, 1992): ITC =

C. Result and Discussion
Results of analysis of the variance for the entire growth phase of rice is shown in Table 1. It shows that the stressed environment affected the entire observed character, whereas the different rice varieties affected almost all of the observed characters except the plant height, percentage of filled seeds, and the percentage of hollow seeds. The interaction of varieties and salinity stress had a significant to very significant effect on the character of root length, number of leaves, number of tillers, number of productive tillers, flowering age, chlorophyll A, chlorophyll B, total chlorophyll, stomata density, shoots wet weight, shoots dry weight, roots wet weight, roots dry weight, length of flag leaves, weight of 100 seeds and production per clump.
Significant interactions shown by the variance analysis are early indicators in stress screening, this has been reported by Anshori (2019); Farid BDR M, Nasaruddin, Anshori MF, Chaerunnisa ANJ. (2020) that characters which have significant interactions have different response patterns among genotypes in normal and stressed environments. Based on the analysis, the characters of root length, number of leaves, number of tillers, number of productive tillers, flowering age, chlorophyll A, chlorophyll B, total chlorophyll, stomata density, shoots wet weight, shoots dry weight, roots wet weight, roots dry weight, length of flag leaves, weight of 100 seeds and production per clump, all can be used as candidates of rice selection criteria for salinity stress on DFT hydroponic screening. However, it is necessary to do more in-depth analysis with several other multivariate analysis series.
The assessment of the variety's response to the stress condition should operate a tolerance index. It has been reported by Mau YS, Ndiwa ASS, and Arsa AGBA.  Table 2. ITC1 is the value of the tolerance index at 60 mM of NaCl, and ITC2 is the value of the tolerance index at 120 mM of NaCl. The value of this index will be used as the basis for the next multivariate analysis in determining the character of selection. Correlation analysis is the most common analysis in identifying the best selection characters. Correlation analysis in this experiment focused on production per clump as its main character. The relationship between other characters to production per clump can be used as the best selection character against salinity stress. The use of this analysis has been reported by  (Table 3) shows that the root length character (0.78), the number of leaves (0.91), chlorophyll A (0.93), chlorophyll B (0.95), total chlorophyll (0.94), stomata density (0.83), shoots wet weight (0.88), shoots dry weight (0.88), roots wet weight (0.93), roots dry weight (0.78), length of flag leaves (0.89) flowering age (0.87), The number of tillers (0.81), the number of productive tillers (0.97), and the weight of 100 seeds (0.93), all have a significant correlation to production per clump. Based on this correlation analysis results, it is not possible to distinguish the direct and indirect effects of each character on the production character. Therefore, it is necessary to continue with cross-print analysis to understand the magnitude of both direct and indirect effects on production (Singh and Chaudhary 2007;Rohaeni and Permadi 2012;Anshori et al.2018).  Correlation analysis is an overview of the level of kinship between one character to the others, but the value of correlation cannot explain the causal relationship of the kinship level among characters. Therefore, in order to elaborate the correlation coefficient to be more useful, come the role of cross-print analysis. The results of cross-print analysis could describe how significant the direct and indirect effects of a character to the main character (Rohaeni and Permadi, 2012). The use of correlation analysis and cross-print analysis in determining the character of selection has also been performed by many researchers including Milligan SB, Gravois KA, Bischoff KP, Martin FA. (1990), Akhmadi (2016), Anshori et al., (2019), Fadhli et al.,(2020), Farid et al.,(2020) and Farid et al., (2021). The cross-print of IST of the production per clump character shows a representative result on the determinant coefficient value of 0.80 (Table 4). Based on the crossprint analysis, the character of total chlorophyll (5.77) had the highest and significant direct effect on the character of production per clump, followed by the character of the number of productive tillers (1.08). Consequently, based on these results the total chlorophyll character and number of productive tillers can be recommended as the selection characters in screening for salinity tolerance throughout the rice growing phase on the DFT hydroponic system. However, these findings need to be supported by other multivariate analyses such as major component analysis and heat map cluster analysis.      Major Component Analysis is one of the multivariate analyses that analyzes the data of several interrelated variables. The purpose of this analysis is to extract important information from the data and describe it as a new set of orthogonal variables called the main components, and to display patterns of observational similarity in other words to summarize the data with a smaller number of variables (Ilmaniati and Putro, 2018). The results of the main component analysis (AKU) produced one main component that could describe the character of productivity (Table 5) i.e. the first major component (KU1) with a total proportion of 98% and eigenvalue of 2.93. Based on main component 1 (KU1), the production character (0.58) in stressed condition is within same direction with total chlorophyll character (0.57) and the number of productive tillers (0.58). Cluster analysis is a common method in plant breeding. There are two main functions of cluster analysis applications: measurement to identify outliers and classifying sample subtypes (Zhao S, Guo Y, Sheng Q, Shyr Y., 2014). However, in the current development, cluster analysis is often combined with heatmap analysis (Virga G, Licata M, Consentino BB, Tuttolomondo T, Sabatino L, Leto C, Bella SL., 2020; Anshori et al. 2020). Based on the GROUPING OF ITC characters, productivity is separate from other character groups, while the other group consists of the number of productive tillers and chlorophyll A. Based on the grouping of varieties, the first group consists of Jeliteng ITC1, Ciherang ITC1, Inpari 34 Salin Agritan ITC1 and Inpari 29 ITC1 while the second group are IR 29 ITC1 and ITC2 for all varieties (Figure 1).

D. Conclusion
The environment at a concentration of 60 mM of NaCl was the best stressed environment in screening rice salinity at all growing phases. A good selection character in this screening procedure was the production per clump, the number of productive tillers and the total chlorophyll. Varieties of Jeliteng, Ciherang, Inpari 34 Salin Agritan were those that considered tolerant to salinity in the screening procedures at all growing phases on DFT hydroponic system. The DFT hydroponic screening method at all growing phases can be recommended as one of the artificial salinity screening methods.