Varied Responses of System of Rice Intensification to Medium and Short-aged Rice and Types of Planting Material in Central Plains of Thailand

Received 31 Oct, 2020

Accepted 18 Feb, 2021

Published 31 Mar, 2021

ABSTRACT
Background and Objectives: The System of Rice Intensification (SRI) as one of the methods of producing rice under controlled water management is known to contribute to the increasing demand for food and ensuring water security and ecological balance of the environment. The performance of short- aged rice under SRI has not tested adequately. Hence a pot experiment was conducted to investigate the performance of short and medium-aged rice varieties and different types of planting materials in SRI. Materials and Methods: Six treatments composed of 2 x 3-factor factorial combinations of two age groups of rice and three types of planting materials were tested in a Completely Randomized Design with three replicates each. Time for reaching different phenological stages, above and below ground growth parameters, and per plant yield parameters were assessed. Results: The medium-aged rice (MAR) variety (115-day-old) showed its adherence to its exact life duration and phenological stages, while the short-aged (SAR) variety (90-day-old) extended its all phenological stages beyond its stipulated durations under SRI practices. The shoot and root growth parameters were significantly different among the two age groups. A significantly higher per hill grain weight was observed in the medium-aged rice variety compared to the short-aged variety. Short-aged rice varieties irrespective of the type of planting material extended the duration of their phenological stages. Conclusions: Short-aged rice extends its phenological stages and performs poor growth and offers less yield under the System of Rice Intensification in the Central Plain of Thailand.

INTRODUCTION

The System of Rice Intensification (SRI), which was accidentally found in 1983 in Madagascar, is becoming popular currently for producing higher yields with lower consumption of water compared to conventional rice production suggesting an alternative to increasing water scarcity in the irrigated rice cultivation. This is an approach for increasing the yield of irrigated rice through integrated crop and resource management^1,2.

The SRI is one of the rice-producing methods which ultimately focus on food, economics, water securities, and ecological balance of the environment. Asian region pays greater concern in popularizing SRI, especially with its strength in acting against climatic change. The performance of different aged varieties of rice is an important criterion in SRI since there are rice varieties of different age groups adopted by farmers in different localities. However, the effects of the age of rice varieties for SRI have not been given adequate attention by many researchers in their SRI research work, except paying attention to the age of seedlings used for transplanting^3,4,5, alternate wetting and drying regimes⁶, spacing^7,8, and method of planting⁹ in the system. A few rice scientists have investigated mainly medium- and long-aged rice varieties, but not the short-aged rice varieties. Nissanka and Bandara¹⁰ have studied a 105-day rice variety and observed vigorous growth of plants in SRI producing a greater number of tillers and higher leaf area eventually giving high grain yields. Gupta¹¹ reported that the response of the SRI was better with 120-130-day-old rice cultivars in India. Vijaya kumar et al.¹² observed higher yields during two consecutive years with a 110-day variety when transplanted with 14-day-old seedlings in 25 x 25 cm spacing under the usual SRI water management compared to conventional production practices. Krishna and Biradarpatil⁴ reported that 12-day-old seedlings produced a greater number of productive tillers per plant at the harvest compared to 8-, 16-, and 25-day-old seedlings of a medium-duration rice variety, ES 18 (120-day-old). However, the literature on the performance of rice in SRI with short-aged varieties is deficient. Therefore, this study was conducted to explore the feasibility of growing short-aged rice varieties in SRI by evaluating their phenology, growth, and yield performances

MATERIALS AND METHODS

Study area: This study was conducted as a pot experiment in a plant house covered with a plastic roof (with a transmissivity of 93%) at the Agricultural Systems Experimental Station of the Asian Institute of Technology, Thailand (Latitude, 14^o 04’N, Longitude 100^o37’E and MASL, 2.27) from February to July 2009. The soils of the site belong to the Rangsit Series, which is deep, very fine clayey and extremely acid sulfate, with pH (soil: H₂O 1:1) of 4.5-4.9. The pH of the soil was raised to 6.2 by adding lime before filling the experimental pots.

Treatments: Six treatments were composed with 2 x 3 factorial combinations of two age groups of rice [viz. short-aged rice –SAR of 90-day-old and medium-aged rice MAR of 110-day-old] and three types of planting materials [viz. direct-seeded rice (DSR), transplanted rice with 8- (TPR₈), and 12-day-old seedlings – (TPR₁₂)], assigned to pots and arranged in a Completely Randomized Design (CRD) with three replicates for each treatment.

Research procedure: Pots were 0.6 m in diameter and0.8 m in height. A five (5) cm hole was made at the bottom of each pot and a PVC tube was fixed up with a removable cap to regulate water level during irrigation and to facilitate draining of water during the non-irrigation period. Pots were filled with a mixture of lime-treated, cow dung-added (at the rate of 10,000 kg ha^-1) and well-plowed soils up to 0.7 m of height leaving 10 cm from the top. During the initial two weeks, all pots were irrigated to help soils to settle down before commencing the study. After two weeks, two non-photoperiod sensitive rice varieties, namely Suphan buri 2 (SAR) and Pathum thani 1 (MAR) recommended to the Central plains of Thailand by the Department of Agricultural and Extension (DOAE), were established in the pots as per treatment schedule.

Rice was transplanted manually in all the pots. Pre-soaked and pre-germinated seeds were directly seeded in pots assigned for DSR. To produce 8- and 12-day-old seedlings a separate nursery was used. Direct seeding was done first. Eight-day-old seedlings were then taken from the nursery and transplanted in the corresponding pots assigned for TPR₈, while TPR12 treatment pots seedlings were transplanted when the seedlings in the nursery were12-days old. This practice enabled all plants to be in the same age and growth stage and to expose to the same environmental conditions at the same time. A plant density of two seedlings per hill was maintained with 25 x 25 cm spacing within each pot. Accordingly, four planting hills were maintained in each pot. Pots were irrigated daily until the crop became well-established.

To assure and monitor the depth of water, indicator scales were placed in all pots. Then water from the nearby irrigation canal was pumped into the pots as per treatment to maintain a 3-5 cm deep water layer during the periods of irrigation. Re-watering was done when the water level depleted below 3 cm in each pot within the period of inundation. Two-week irrigation and two-week non-irrigation period were strictly practiced as water management cycles, which continued until flowering. After flowering, the water level in the pots was maintained to a depth of 5cm until grains reached the hard dough stage, i.e., two weeks before harvesting, and then drained out to facilitate grain drying. Pots were regularly hand weeded. Integrated pest management (IPM) practices were adopted for pest management.

The number of days taken for 50% of the plants to reach maximum tillering (MT), flowering, and harvesting was recorded. The number of tillers per hill was counted at MT, flowering, and harvesting. The plant height of each hill was measured and recorded at MT and flowering and the mean height was calculated. Leaf area per hill was estimated by measuring the length and breadth of each leaf at flowering as described by Yoshida¹³.

Yield data were obtained from all four hills in each pot. The number of productive tillers per hill and the number of filled and unfilled grains per panicle were counted and recorded. One hundred (100) grain weight was recorded after drying the grains in the sun followed by an oven drying at 80^oC until a constant weight was reached. The moisture content of grains of all hills was recorded using Grain Moisture Meter (Grainer II PM 300, Made in Japan). Later, grain yield was computed to a 14% moisture content.

After harvesting, all pots were turned upside down on the ground and the depths of the root systems were recorded using a meter ruler. Thereafter, complete root systems of all four plant hills in each pot were carefully separated and washed with running water on a mesh (6.4 squares/cm) until all soil particles disappeared from the root mass. Then roots were sun-dried and dry weights were recorded. A representative sample from each root system was taken, its weight was recorded and the total length measured using a meter ruler and the gridline intersect method¹⁴. Then the root weight to root length ratio was calculated. The total length of the root system was estimated using the root weight to root length ratio. Roots were then dried in an oven at 80^oC until a constant weight was reached, and dry weights of whole root systems were recorded. The shoot to root ratio was calculated using shoot and root dry weights. The solar radiation and temperature inside and outside the greenhouse were also recorded daily throughout the cropping period.

Statistical analysis: Analysis of variance (ANOVA) was performed for data as per Complete Randomized Design (CRD)and Fisher’s Protected Least Significant Difference (LSD) was used for mean separation¹⁵.

RESULTS

Weather: Weather data during the study period showed the temperature and solar radiation ranged from 26^oC to 40^oC and from 172.88-732.63 Wm^-2 inside, respectively, and from 27 to 37^oC and 221.16- 1671.77 Wm^-2 outside the plant house, respectively (Fig. 1).

Fig. 1: Variation of temperature and solar radiation in and out of the greenhouse

Growth Performances
Plant Phenology : The age of the rice variety, the type of planting material, and the interaction between the two factors did not significantly influence the time to commence and 50% flowering as shown in Table 1. Similarly, the time of harvesting between SAR and MAR varieties also did not differ significantly against their 20-day age difference of the SAR variety. Flowering commenced between 72 and 77 days (i.e., a delay of 30-32 days), and between 79 and 81 days for MAR variety with a delay of 13-15 days from expected days. Fifty percent flowering occurred in 83-87 days for 90-day variety with a delay of 28-32 days, and in 86-90 days for MAR variety with a delay of 10-14 days from expected days as per Table 2.

Table 1: Effect of age of the variety and type of planting material and their interaction on reaching different phenological stages of rice
Analysis of variance
Source of variation 1/	DF		Mean squares
			Beginning flowering	50% flowering	Harvest
A	1		117.55	18.00	2.72
P	2		15.50	46.33	122.38*** 2/
A*P	2		11.05	14.33	3.39
Error	12		13.44	7.44	1.17
Results
Treatments			Phenological stage, days
			Beginning Flowering	50% flowering	Harvest
Age of rice variety (A)
SAR		74.4 ± 3.9 3/		85.3 ± 2.9	109.0 ± 4.9
MAR		79.6 ± 3.4		87.3 ± 3.2	109.8 ± 3.8
LSD (p=0.05)		ns		ns	ns
Type of planting material (P)
DSR			78.2± 3.9	84.7± 39	113.7±1.4
TPR8			75.2± 4.9	88.5±1.9	109.8±1.5
TPR12			77.7± 4.6	85.8±3.3	107.7±0.8
LSD (p=0.05)			ns	ns	1.5
CV%			4.7	3.2	1.1
1/A - Age of the rice variety; P-Type of planting material; DSR – Direct- seeded rice; TPR8 –Transplanted rice with 8-day-old seedlings; TPR12– Transplanted rice with 12-day-old seedlings. 2/*- F value is significant at p=0.05; ** - p=0.01; and *** - p=0.001; ns – not significant at p = 0.05; 3/ Standard deviation of means

Both varieties reached the harvesting stage in 109 days from the date of establishment. However, the time of harvesting significantly varied due to the type of planting material: direct-seeded rice (DSR) reached the harvesting stage in 113 days from the establishment, which was significantly longer than that of transplanted rice of 8 and 12-days old. Similarly, 8-day-old rice also had a significantly longer time (almost 2 days) than12 day-old rice. There was a delay in reaching the harvesting stage compared to the expected days. When transplanted at the age of 12-days, the SAR variety took more than 14 days to reach the maturity stage when transplanted at the age of 12-days, while MAR took an extra six days as per Table 2.

Table 2: Expected and observed durations for reaching specific phenological stages of two rice varieties as influenced by age group and type of planting material in the System of Rice Intensification (SRI).
Age of Rice Variety	Type of planting material 1/	Commencement of flowering		50% Flowering		Harvesting
		Expected	Observed	Expected	Observed	Expected	Observed
		--------------------------------- Number of days ---------------------------
SAR	DSR	45	75	55	83	90	114
	TPR8	45	72	55	87	90	109
	TPR12	45	77	55	86	90	104
MAR	DSR	66	81	76	87	110	113
	TPR8	66	79	76	90	110	111
	TPR12	66	79	76	86	110	106
1/DSR – Direct seeded rice; TPR8 - Transplanted rice with 8-day-old seedlings;  TPR12 - Transplanted rice with 12-day-old seedlings;

Shoot Growth Parameters
Tiller number per hill: The tiller number per hill was significantly influenced only by the type of planting material and that was only at maximum tillering, and by the age of rice variety at harvest (Table 3). Direct seeded rice had the highest tiller number (45 per hill) and was significantly greater (p<0.01) than rice established with TPR₈ and TPR₁₂ at maximum tillering. Rice transplanted with12-day-old seedlings had the lowest tiller number per hill, which was only 60.1% from the DSR. Transplanting at 8-day-old rice had a moderate tiller number per hill (41 tillers – a 9% reduction) but was not significantly different from DSR.

Table 3: Effect of age of the rice variety and type of planting material and their interaction on selected shoot growth parameters
Analysis of variance
Source of variation1/	DF	Mean squares
		Tiller number (No./hill)			Plant height (cm)		Leaf area at flowering (m2/hill)	Shoot biomass (g/hill)
		Maximum tillering	Flowering	Harvest	Maximum tillering	Flowering
A	1	17.60	33.35	361.81**/2	567.85***	373.56**	0.29***	190.71
P	2	562.71**	220.79	53.01	1013.57***	177.18*	0.29***	572.08**
A*P	2	79.74	91.76	43.39	13.31	47.76	0.05***	699.80
Error	12	37.06	76.26	20.48	16.11	38.81	0.001	210.40
Results
Age of rice variety (A)
SAR		38.9±9.33/	57.3±11.4	35.1±6.3	66.4±11.3	102.2±7.8	0.8±0.1	130.2±27.1
MAR		37.0±11.4	54.6±7.9	44.1±3.9	77.6±12.4	111.3±7.3	1.0±0.3	136.7±12.2
LSD (p=0.05)		ns	ns	4.9	4.0	6.4	0.09	ns
Type of planting material (P)
DSR		45.6±6.8	58.4±7.5	42.9±5.1	83.8±7.7	112.9±7.5	1.1±0.3	144.7±16.2
TPR8		41.3±7.7	60.3±10.3	38.8±9.0	74.1±6.1	104.8±7.8	0.8±0.1	140.7±16.4
TPR12		27.1±4.5	49.0±7.8	37.1±5.6	58.1±7.3	102.6±8.3	0.7±0.1	114.9±17.1
LSD (p=0.05)		8.1	ns	ns	4.9	7.8	0.1	20.7
CV%		16.02	15.62	11.43	5.58	5.83	3.69	10.80
1/A - Age of the rice variety; P- Type of planting material ; DSR – Direct- seeded rice; TPR8–Transplanted rice with 8-day-old seedlings; TPR12– Transplanted rice with 12-day-old seedlings; 2/*- F values are significant at p=0.05; ** - p=0.01; and *** - p=0.001; ns – not significant at p = 0.05; 3/Standard deviation of means.

Towards flowering, there was an increase in the number of tillers per hill. It amounted to 28% for direct-seeded rice, 46% for transplanted rice with at 8-day-old seedlings, and 44.7% for transplanted rice with at 8-day-old seedlings. However, at the harvest, the tiller number was significantly influenced only by the age of rice variety. The MAR variety had a significantly greater tiller number (44 tillers/hill) than SAR rice at harvest as per Table 3.

Plant height at flowering: Among the selected above-ground growth parameters, the plant height was influenced by the two main factors, i.e., age and type of planting material at maximum tillering and flowering. The MAR variety was significantly taller than that of SAR variety, and DSR was taller than TPR as stated in table 3. This is probably attributed to relatively longer vegetative periods of longer-aged rice compared to shorter-aged varieties and direct seeding compared to transplanting.

Leaf area at flowering: Both age of the variety and the type of planting material, and their interaction had significant effects on leaf area per hill at flowering (Table 3 and Fig. 2). The interaction between the two main factors showed that the highest leaf area was in the MAR variety established by DSR and was significantly greater than the rest of the factor-level combinations. No other treatment combinations displayed significant differences.

Fig. 2: Effect of interaction between the age of rice and type of planting material on leaf area at the flowering of rice in SRI

Shoot biomass: The highest shoot biomass was in DSR (144.9 g/hill) and it was significantly greater than TPR₁₂ (114.9 g/hill) (Table 3). There was no significant difference in shoot biomass between DSR and TPR₈ (140.7 g/hill). TPR₁₂ had the lowest shoot biomass.

Root Growth Parameters
Root depth (RD): There were significant effects on root depth by the age of rice and type of planting material and their interaction as per Table 4. The interaction effect on root depth is shown in Fig. 3. The only significant increase in the root depth was found in the DSR of MAR variety, which had 85 cm, whereas the other root depths were in the range of 61 to 67 cm.

Table 4: Effect of age of the rice variety and type of planting material and their interaction on different root growth parameters
Analysis of variance
Source of variation1/	DF	Mean squares
		Root depth (cm)	Total root length (m)	Root biomass (g/hill)	Shoot to root ratio
A	1	364.50***/2	12306.19**	18.34**	188.63*
P	2	253.50***	1248.94	5.72*	134.28*
A*P	2	145.50***	817.15	2.93	88.79
Error	12	4.50	635.76	2.00	31.21
Results
Age of the rice variety (A)
SAR		63.0±3.13/	164.0±20.9	4.8±0.9	27.4±9.0
MAR		72.0±9.8	216.3±31.2	6.9±1.9	20.9±4.7
LSD (p=0.05)		5.0	26.0	1.3	6.4
Type of planting material (P)
DSR		75.1±11.1	191.4±56.1	6.9±2.6	22.9±9.9
TPR8		63.5±1.9	203.9±23.1	5.0±0.9	29.3±6.8
TPR12		64.0±3.7	175.1±25.1	5.7±0.6	20.1±2.1
LSD (p=0.05)		6.1	ns	1.6	7.9
CV%		3.14	13.26	20.19	23.51
1/A - Age of the rice variety; P-Type of planting material; DSR – Direct- seeded rice; TPR8–Transplanted rice with 8-day-old seedlings; TPR12– Transplanted rice with 12-day-old seedlings. 2/ * - F value is significant at p=0.05; ** - p=0.01; and *** - p=0.001; ns – not significant at p = 0.05; 3/Standard deviation of means

Fig. 3: Effect of the interaction between the age of rice and type of planting material on the root depth of rice in SRI

Total root length (TRL): The age of rice variety had a significant influence on the total root length of rice, but neither the type of planting material nor the interaction between age and type of planting material showed significant effects (Table 4). The MAR variety had the highest and significantly greater TRL than that of the SAR variety.

Root biomass (RB): Root biomass per hill of rice was significantly influenced by both the age of the rice variety and the type of planting material (Table 4). There was no significant interaction between the two factors on root biomass. Of the age of the variety, MAR variety had significantly greater RB than the SAR variety, while DSR had significantly greater RB than both TPR. A longer growing period available for root development could have been the contributory factor for higher RB of DSR, and TPR appeared to have had some disturbance to root development due to uprooting and transplanting.

Shoot to root ratio: Both the age of rice variety and the type of planting materials had significant effects on the shoot to root ratio, but there was no significant interaction between the two factors (Table 4). Between the two varieties used, the shoot to root ratio was highest in the SAR variety (27.4) compared to the MAR variety, while TPR₈ (29.3) among the types of planting material used had the highest shoot to root ratio. Both MAR variety and TPR₁₂ had the lowest shoot to root ratio. The lower root biomass of the TPR₈ compared to TPR₁₂ has made a high S:R ratio of TPR₈ than TPR₁₂ which is a bit unrealistic.

Yield and Yield Parameters
Panicle number per hill (PNH): The panicle number per hill was significantly influenced only by the age of rice variety as stated in Table 5. There was neither the type of planting material nor their interaction had any significant effect on the number of panicles per hill. The PNH was significantly greater only in the MAR variety compared to the SAR variety.

Table 5: Effect of age of the variety and type of planting material and their interaction on yield and yield parameters of rice
Analysis of variance
Source of variation1/	DF	Mean squares
		Panicles (No./hill)	Filled grains (No./panicle)	100-grain weight (g)	Panicle length (cm)	Panicle weight (g/panicle)	Grain weight (g/hill)	Spikelet sterility (%)
A	1	361.54**2/	5417.71*	0.16**	3.10*	0.65*	2417.44**	19.18
P	2	106.11	3393.18	0.01	4.87**	0.46	552.16*	114.25*
A*P	2	43.09	725.58	0.02	4.47**	0.01	74.34	333.03**
Error	12	20.48	619.60	0.01	0.62	0.09	131.95	18.81
Results
Age of rice variety (A)
SAR		35.1±6.3/3	165.6±35.9	2.2±0.1	29.9±1.5	3.3±0.2	62.8±16.6	9.9±6.9
MAR		44.1±3.9	130.9±15.6	2.4±0.1	29.1±1.0	3.7±0.4	85.9± 8.9	11.9±9.5
LSD (p=0.05)		4.9	25.5	0.1	1.1	0.3	11.2	ns
Type of planting material (P)
DSR		42.9±5.1	166.9±30.6	2.3±0.1	29.4±0.8	3.4±0.3	81.4±11.2	14.1± 6.0
TPR8		38.9±9.0	134.1±43.1	2.3±0.1	28.7±1.1	3.2±0.4	63.5±20.9	12.7±11.2
TPR12		37.1±5.6	143.7± 8.1	2.3±0.2	30.5±1.4	3.8±0.3	78.3±16.4	5.9± 4.5
LSD (p=0.05)		ns	ns	ns	1.3	ns	13.8	9.9
CV%		11.43	16.69	4.24	2.66	8.68	15.44	9.67
1/A - Age of the rice variety; P- Type of planting material; DSR – Direct- seeded rice; TPR8–Transplanted rice with 8-day-old seedlings; TPR12– Transplanted rice with 12-day-old seedlings. 2/*- F value is significant at p=0.05; ** - p=0.01; and *** - p=0.001; ns – not significant at p = 0.05; 3/Standard deviation of means;

Number of filled grains per panicle (FGNP): Similar to panicle number per plant, the number of filled grains was significantly influenced by the age of rice only, but the grain number was higher in SAR (165.6) than the MAR (130.9) as clearly mentioned in Table 5.

100-grain weight (HGW): The 100-grain weight was also significantly influenced by the age of rice variety and MAR variety had a higher grain weight than SAR. Neither type of planting material nor the interaction between the two factors had any significant effect.

Panicle length (PL): Both the age group of rice and the type of planting material as well as the interaction between the two factors had significant effects on the panicle length (Table 5). The highest panicle length was in SAR variety with TPR₁₂, which was significantly greater than the other factor combinations (Fig. 4). The next highest was in MAR variety established with direct seeding. The lowest panicle length was in MAR with TPR₈.

Fig. 4: Effect of the interaction between the age of rice and type of planting material on panicle length

Panicle weight (PW): The panicle weight was significantly influenced by the age of rice, in which MAR variety had a significantly greater panicle weight (3.7 g) compared to SAR variety (3.3 g).

Grain weight per hill (GWPH): There was a significant influence on grain weight per hill by both age and type of planting material of rice (Table 5). There was no interaction between the factors that significantly influenced the GWPH. Of the two factors, MAR variety had a significantly higher grain weight per hill (85.9 g) compared to SAR variety (62.8 g). DSR had a significantly greater per hill grain weight (81.4g) compared to TPR₈. The per hill grain weight of DSR was not significantly different from TPR₁₂ (78.3 g).

Spikelet Sterility % (SS): The spikelet sterility percentage was significantly influenced by the two factors and their interaction (Table 5), and the interaction effect is shown in Fig. 5. The spikelet sterility was highest in MAR variety with TPR₈ (14.1%). The lowest spikelet sterility was in the MAR rice variety with TPR₁₂. However, the spikelet sterility of SAR variety was non-significantly lower when combined with TPR₁₂ compared to the other two planting materials.

DISCUSSION

The phenology of rice usually is sensitive to water stress during the vegetative period^16,17 and dry weather defers flowering fromone to several weeks. However, the reproductive and grain filling periods from flowering to physiological maturity are often least modified in conventionally grown rice regardless of the age group; and this period is usually consistent around 30-35 days under any change in rainfall and resource supply in conventional rice production systems¹⁸. The SAR rice variety completed vegetative growth in nearly 60 days, while it took 90 days for the MAR variety under normal growing conditions.

The results of the current study showed a deviation from both vegetative and reproductive growth durations under SRI water management practices. Commencement of flowering of TPR₈ had extended from 45 to 72 days (by 27 days), TPR₁₂ from 45 to 77 days (by 32 days), and by 30 days with DSR (Table 2). Fifty percent flowering also extended from 55 days to 83 days (by 28 days)in DSR to 86 and 87 days (by 31 and 32 days) in TPR₁₂ and TPR₈, respectively. As a result, harvesting time (maturity) extended from 90 days to 104 and 109 days in TPR₁₂ and TPR₈, respectively, and 114 days in DSR. Overall, SAR variety delayed its maturity period by 14 days with TPR₁₂ taking 114 days, by 24 days with DSR taking 104 days, and TPR₈ was moderate taking additional 19 days (109 days from establishment).

The MAR variety also delayed its commencement of flowering by 13 to 15 days, 50% flowering by 10-14 days in SRI. However, rice plants in TPR₁₂ reached harvesting time 4 days earlier (106 days against 110 days of expected maturity). The harvesting time of TPR₈ and DSR reached 111 and 113 days, respectively, which is an ignorable variation in the maturity period. The differences in reaching phenological stages were attributed to soil drying resulted from withholding irrigation for two weeks leading to water stress and thus its recovery time during this period. Water stress appeared to have slowed down growth and development related processes thus extending the time taken to recover from the stress with follow up wetting.

The drying period induced by suspending irrigation in alternate wetting and drying (AWD)seemed to modify crucial physiological activities; in this situation, the vegetative period was eventually extended in both age groups (SAR and MAR). This could be attributed to the additional time required to fulfill key physiological and phenological requirements that will eventually ensure the plant to be in a state to support reproductive functions. Although it has been reported that long-aged rice varieties have a better defense than short-aged varieties in responding to water stresses that occur during the vegetative period¹⁹, this study shows that when the water stress disturbs physiological processes supporting key developmental activities and reproductive functions, rice plants seem to make all efforts to support its reproductive function, even by extending the vegetative growth period to fulfill growth and developmental needs before reaching flowering. There is literature indicating that water stress promoting the synthesis of Abscisic Acid (ABA) in the rice plant that ensures increasing partitioning of assimilates to root development, prolonging leaf functions, and improving tolerance to water stress^20,21. Although the reproductive period is known to be more or less consistent, Garrity, and O’Toole²² reported that it is also sensitive to water stress and variable. Being grain filling period critical in determining the size and weight of grains that contribute to final yield, SRI adopts continuous inundation to 5 cm depth during the reproductive and grain filling periods to avoid water shortage or stress and ensure adequate water availability so that the rate of assimilate transfer to grains would be maintained as needed. The issue of water stress occurs in SRI during the vegetative period because of alternate wetting and drying. After two weeks of flooding, suspension of water availability for two weeks may gradually expose the rice plants to water stress which seems to stimulate the plant's recovery and physiological processes in favour of yield components. In this aspect, the synthesis of cytokinins governs the phenological development of the rice plant. It has been reported that cytokinins promotes several growth functions such as deep root development ensuring the maintenance of water uptake following water stress periods as well as during follow up water stress periods²³, maintenance of functioning leaves over a longer period by delaying leaf senescence and maintaining the photosynthetic capacity of existing leaves^24,25, increasing tiller production and tiller development with the support of the root and leaf functioning^23,24,26, etc., and all these positive changes are effectively manipulating vegetative period to produce the optimum leaf area and increasing the number of tillers and spikelets. These two components eventually ensure higher yields of rice in SRI compared to the conventionally grown rice crop. Therefore, more than the stipulated period of 90-days of SAR, prolonging the vegetative period has taken place to ensure its potential yield, which led its overall growth period to take a full-time duration similar to the maturity period of MAR variety. This period seemed to have not been required for MAR variety as the period of water stress resulted due to suspension of water supply and the periods for stress recovery and growth promotion with follow-up flooding were adequate for its usual growth and development periods. This agrees with Stoop¹⁹ who reported that longer duration varieties can recover physiological drawbacks that resulted from water stress during vegetative periods due to their longer vegetative growth period compared to a short-day variety like 90 days having 60 days of the vegetative period under SRI. Latif et al.²⁸ also reported similar results with longer duration varieties indicating that such varieties perform better in SRI giving higher yield than short-aged and medium-aged rice varieties. Similar findings were also reported by Raju and Sreenivas²⁹ by evaluating 90-, 110-, 130-,135- and 145-day-old varieties. In their work, 145-day-old varieties performed better than medium (110-, 130-, and 135-day-old) and short-aged (90-day-old) rice varieties, specifically having a significantly higher tiller number in the longer duration variety.

Katayama²⁹ reported that phyllochrons are responsible for ensuring the number of tillers formed in rice. Longer duration rice varieties have a longer vegetative period and these varieties can complete full phyllochrons¹⁹ since the vegetative period of these varieties is long enough to complete full phyllochrons even when rice is established with either by direct seeding or early transplanting. Youki et al.³⁰ reported that the developmental change of phyllochron is highly dependent upon the genotypes and the growth duration of the crop. The varieties with shorter vegetative periods possess faster growth phases, and any type of stress including water stress could drastically reduce the growth rate thus delaying the reach of specific phenological stages. For short-aged rice varieties, water stress if occurred during the vegetative period will provide only a short period to recover from the stress once the stress is alleviated by irrigating the crop, compared to longer-aged rice varieties³¹. This shows that only those varieties that could prolong the vegetative period under water stress conditions would appear to be suitable for SRI, in case if short-aged rice varieties are used. The SAR variety (Suphan Buri 2) used in this study shows its growth plasticity and hence suitability for SRI when only the growth plasticity is concerned.

Transplanting young seedlings of 8- to 10-day-old and not after 15days preserves the optimum tillering and rooting potential. This may be reduced if transplanting is done after the 4^th phyllochron, i.e., 15 days after emergence¹. However, three types of planting materials used in the present experiment were within 8-12-days range, and hence plants were able to make use of the maximum tillering potential of the rice plant. Direct seeded rice (DSR) irrespective of the age of the variety had greater plant height than transplanted rice (TPR₈ and TPR₁₂). The DSR experienced minimum disturbance after establishment and hence produced taller plants compared to transplanted rice plants (TPR). Medium-aged variety (MAR) had taller plants compared to SAR in this study. Leaf area per plant was significantly greater with DSR of MAR variety than that of SAR variety. Longer vegetative period and higher tiller production of the long age variety were the main reasons for contributing to greater leaf area of MAR variety than that of SAR variety. This was previously reported by Schnier et al.³², Dingkuhn et al.³³, and Ginigaddara and Ranamukhaarachchi⁹.

Root biomass was significantly greater in MAR variety compared to SAR variety. This was due partly to the ability of longer aged varieties to produce greater root mass compared to shorter aged varieties. However, mediation of deep root development by cytokinin under water stress conditions seemed to have also contributed to increased root biomass as reported by Ookawa et al.²³. The lower shoot to root ratio of MAR withTPR₁₂ indicates relatively higher root weight in the treatments compared to those with higher shoot to root ratio, which is an adjustment of the root systems to maintain plant water uptake during the water stress period and hence the photosynthetic capacity. Since SRI induced water stress during the period of suspension of irrigation, adjustment of the shoot to root ratio indicates the ability of rice plants to grow deeper in search of water to overcome the periods of water stress. TPR₈ of this study has shown lower root depth and root biomass and hence higher S:R ratio when compared to the TPR₁₂ which should have been the other way around. The continuous standing water layer in the pots after PI may have retarded the root growth of TPR₈. Thakur et al.³⁴ have proved that a long-standing continuous 5 cm water layer of the SRI field after PIis not advisable compared to continuing with AWD as it may lead to retardation of the growth of the rice plants in addition to wasting water at the same time.

The significantly higher panicle weight per plant, 100-grain weight, and grain weight per panicle of the medium-aged rice variety (MAR) than that of short-aged rice variety (SAR) explain the mechanisms contributing to better performance of medium duration varieties than shorter duration varieties in SRI. Tolerance to stress and the availability of time in the vegetative phase to adjust to stresses occurring might have been the reasons for this result. The higher per hill grain yield of DSR in shorter duration rice variety might be due to the least disturbance for the plant when established with direct seeds than transplanting.

CONCLUSION

In conclusion, 90-day rice variety irrespective of the type of planting material used in the current study showed weaker performance in the System of Rice Intensification with the extension of their phenological stages, performing poor growth, and lowering yields compared to 110-day rice variety in the Central Plain of Thailand. This study indicates the suitability of the 110-day variety compared to the 90-day variety for SRI. However, further experiments would be recommended under field-level assuring inducement of water stress with the suspension of irrigation to confirm the findings further. In cases where soil moisture depletion induced by the AWD leading to water stress in the sub-surface soil profiles is complemented by water in deeper layers due to deep root growth, the benefit of cytokinin promoting growth and yield advantages in SRI may be hindered. In this respect, the identification of a characteristic site in a suitable environment for further studies would enable us to confirm the results of the current study.

ACKNOWLEDGMENT

Authors thank the Government of Norway for providing a scholarship for graduate studies of the first author and Asian Rice Foundation, the USA for partial financial assistance for in-depth studies at the Asian Institute of Technology, Thailand.

REFERENCES

1. Stoop, W.A., N. Uphoff and A. Kassam, 2002. A review of agricultural research issues raised by the System of Rice Intensification (SRI) from Madagascar: Opportunities for improving farming systems for resource-poor farmers. Agric. Syst., 71: 249-274.
2. Uphoff, N. and R. Randriamiharisoa, 2002. Reducing Water use in Irrigatedrice Production with the Madagascar System of Rice Intensification (SRI). In: Water-Wise Rice Production, Bouman, B.A., H. Hengsdijk, B. Hardy, P.S. Bindraban, T.P. Thuong and J.K. Ladha (Eds.). International Rice Research Institute, Los Banos, Philippines 89-102.
3. Krishna, A., N.K. Biradarpail, K. Manjappa and B.B. Channappagoudar, 2008. Evaluation of system of rice intensification cultivation, seedling age and spacing on seed yield and quality in Samba Masuhri (BPT-5204) rice. Kartaka J. Agric. Sci., 21: 20-25.
4. Krishna, A. and N.K. Biradirpatil, 2009. Influence of seedling age and spacing on seed yield and quality of short duration rice under system of rice intensification cultivation. Karnataka J. Agric. Sci., 22: 53-55.
5. Makarim, A.K., V. Balasubramanian, Z. Zaini, I. Syamsiah and I.G.P.A. Diratmadja et al., 2002. Systems of rice intensification (SRI): evaluation of seedling age and selected 20 components in Indonesia. In: Water-wise Rice Production, Bouman, B.A.M., A. Hengsdijk, B. Hardy, P.S. Bindraban, T.P. Tuong, and J.K. Ladha, IRRI, Los Banos, Philippines, pp: 129-139.
6. Thakur, A.K., R.K. Mohanty, D.U. Patil and A. Kumar, 2013. Impact of water management on yield and water productivity with system of rice intensification (SRI) and conventional transplanting system in rice. Paddy Water Environ., 12: 413-424.
7. Reuben, P., F.C. Kahimba, Z. Katambara, H.F. Mahoo and W. Mbungu et al., 2016. Optimizing plant spacing under the systems of rice intensification (SRI). Agric. Sci., 07: 270-278.
8. Thakur, A.K., S. Rath, S. Roychowdhury and N. Uphoff, 2010. Comparative Performance of rice with system of rice intensification (SRI) and conventional management using different plant spacings. J. Agron. Crop Sci., 196: 146-159.
9. Ginigaddara, G.S. and S.L. Ranamukhaarachchi, 2009. Effect of conventional, SRI and modified water management on growth, yield and water productivity of direct-seeded and transplanted rice in central Thailand. Aust. J. Crop Sci., 3: 278-286.
10. Nissanka, S.P. and T. Bandara, 2004. Comparison of productivity of system of rice intensification and conventional rice farming systems in the dry-zone region of Sri Lanka. Proceedings of the 4th international crop science congress. September, 26-1 October, 2004, Australian Society of Agronomy Inc.
11. Gupta, R., 2002. Tour Report to the International Conference on System of Rice Intensification (SRI), Sanya, Hainan Province, China. Proceedings of an International Conference on Assessments of the System of Rice Intensification (SRI) Sanya, China, April 1-4, 2002 SRI International Network and Resources Center.
12. Vijayakumar, M., S.D. Sundar Singh, N.K. Prabhakaran and T.M. Thiyagarajan, 2005. Effect of SRI (System of Rice Intensification) practices on the yield attributes, yield and water productivity of rice (Oryza sativa L.). Acta Agronomica Hungarica, 52: 399-408.
13. Yoshida, S., 1981. Fundamental of Rice Crop Science. International Rice Research Institute, Los Banos, Laguna, Philippines, Pages: 269.
14. Bohm, W., 1979. Methods of studying root systems. Ecol. Stud.
15. Steel, R.G.D. and J.H. Torrie, 1980. Principles and Procedures of Statistics: A Biometrical Approach. 2nd Edn., JH McGraw-Hill, New York.,
16. Pantuwan, G., S. Fukai, M. Cooper, S. Rajatasereekul and J.C. O"Toole, 2002. Yield response of rice (Oryza sativa L.) genotypes to drought under rainfed lowlands. 2. Selection of drought resistant genotypes. Field Crops Res., 73: 169-180.
17. Centritto, M., M. Lauteri, M.C. Monteverdi and R. Serraj, 2009. Leaf gas exchange, carbon isotope discrimination and grain yield in contrasting rice genotypes subjected to water deficits during the reproductive stage. J. Exp. Bot., 60: 2325-2339.
18. Chang, T.T. and E.A. Bardenas, 1965. The morphology and varietal characteristics of rice plant. Int. Rice. Res. Inst. Tech. Bull., 4: 40-40.
19. Stoop, W.A., 2005. The System of Rice Intensification (SRI): Results from Exploratory Field Research in Ivory Coast -- Research Needs and Prospects for Adaptation to Diverse Production Systems of Resource-Poor Farmers. West African Rice Development Association, WARDA, Ivory Coast.
20. Wilkinson, S. and W.J. Davies, 2010. Drought, ozone, ABA and ethylene: new insights from cell to plant to community. Plant, Cell Environ., 33: 510-525.
21. Ye, N., G. Zhu, Y. Liu, Y. Li and J. Zhang, 2011. ABA controls H₂O₂ accumulation through the induction of OsCATB in rice leaves under water stress. Plant Cell Physiol., 52: 689-698.
22. Garrity, D.P. and J.C. O`Toole, 1995. Selection for reproductive stage drought avoidance in rice, using infrared thermometry. Agron. J., 87: 773-779.
23. Ookawa, T., Y. Naruoka, T. Yamazaki, J. Suga and T. Hirasawa, 2003. A comparison of the accumulation and partitioning of nitrogen in plants between two rice cultivars, akenohoshi and nipponbare, at the ripening stage. Plant Prod. Sci., 6: 172-178.
24. Soejima, H., T. Sugiyama and K. Ishihara, 1995. Changes in the chlorophyll contents of leaves and in levels of cytokinins in root exudates during ripening of rice cultivars nipponbare and akenohoshi. Plant Cell Physiol., 36: 1105-1114.
25. Peleg, Z., M. Reguera, E. Tumimbang, H. Walia and E. Blumwald, 2011. Cytokinin-mediated source/sink modifications improve drought tolerance and increase grain yield in rice under water-stress. Plant Biotechnol. J., 9: 747-758.
26. Jiang, C.Z., T. Hirasawa and K. Ishihara, 1988. Physiological and ecological characteristics of high yielding varieties in rice plants. II. leaf photosynthetic rates. Jap. J. Crop Sci., 57: 139-145.
27. Latif, M.A., M.R. Islam, M.Y. Ali and M.A. Saleque, 2005. Validation of the system of rice intensification (SRI) in Bangladesh. Field Crops Res., 93: 281-292.
28. Raju, R.A. and C. Sreenivas, 2008. Agronomic evaluation of system of rice intensification methods in godavari delta. ORYZA, Int. J. Rice, 45: 280-283.
29. Katayama, T., 1951. Studies on Tillering in Rice, Wheat and Barley. Yokendo, Tokyo, Japan.
30. Itoh, Y., S. Sato and Y. Sano, 2001. Developmental changes of phyllochron in near-isogenic lines of rice (Oryza sativa L.) with different growth durations. Euphytica, 119: 271-278.
31. Farooq M., S.M.A. Basra and B.A. Saleem, 2006. Integrated rice growing system. Daily Dawn, Lahore, Pakistan.
32. Schnier, H.F., M. Dingkuhn, S.K. de Datta, K. Mengel and E. Faronilo, 1990. Nitrogen fertilization of direct-seeded flooded vs. transplanted rice: I. Nitrogen uptake, photosynthesis, growth and yield. Crop Sci., 30: 1276-1284.
33. Dingkuhn, M., H.F. Schnier, S.K.D. Datta, E. Wijangco and K. Dorffling, 1990. Diurnal and developmental changes in canopy gas exchange in relation to growth in transplanted and direct-seeded flooded rice. Funct. Plant Biol., 17: 119-134.
34. Thakur, A.K., K.G. Mandal, R.K. Mohanty and S.K. Ambast, 2018. Rice root growth, photosynthesis, yield and water productivity improvements through modifying cultivation practices and water management. Agric. Water Manage., 206: 67-77.