Impact of Harvesting Time on Grain Yield, Physicochemical Attributes, and 2-Acetyl-1-pyrroline Biosynthesis in Aromatic Rice

Abstract

Achieving a consistent grain yield while preserving persistent aroma remains a substantial challenge in aromatic rice production in Bangladesh. To address this challenge, a field experiment was conducted at the agronomy research area of Sher-e-Bangla Agricultural University, Dhaka-1207, between 15 June 2022, and 25 November 2022 (Aman season). This study aimed to evaluate the influence of harvesting time on aromatic rice performance. The experiment, following a randomized complete block design with three replications, involved two factors: factor 1 comprised various rice varieties [Bangladesh Rice Research Institute (BRRI) dhan34, BRRI dhan70, BRRI dhan80, and Tulshimala], and factor 2 comprised three harvesting times [3, 4, and 5 weeks after flowering (WAF)]. Results revealed significant impacts of variety and/or harvesting time on grain yield, physicochemical characteristics, and aroma of aromatic rice. Notably, Tulshimala and BRRI dhan80 exhibited superior milling quality, biochemical properties, and aroma characteristics among the aromatic rice types. BRRI dhan70 and BRRI dhan80 displayed higher grain yield when harvested at 5 WAF. However, Tulshimala and BRRI dhan80 showed superiority in grain 2-acetyl-1-pyrroline (2-AP) concentration when harvested 3 or 4 WAF. Earlier harvesting at 3 and 4 WAF resulted in higher percentages of grain 2-AP (60.22% and 53.96%, respectively) compared with later harvesting at 5 WAF (used as check). In conclusion, varying harvesting times markedly impact the yield, physiochemical characteristics, and aroma of aromatic rice varieties, with earlier harvesting beneficial for aroma retention in Tulshimala and BRRI Dhan80 and later harvesting for increased economic yield in BRRI dhan70 and BRRI dhan80.

1. Introduction

Rice (Oryza sativa L.) is a crucial staple crop that feeds billions of people worldwide [1], and aromatic rice is highly regarded for its excellent grain quality, potent aroma, and delightful flavor [2]. The allure of aromatic rice, particularly in Asian cuisines, stems from its pleasing scent; thus, aromatic rice is both steamed and used in various culinary preparations, such as polau, biriyani, kacchi, firni, and payesh, as well as being favored by Bangladeshis during special social gatherings. In some cases, aromatic rice is preferred over plain rice for regular consumption [3]. Although the globally recognized varieties Jasmine and Basmati are popular for their long-grain and aromatic qualities, the Bangladesh Rice Research Institute (BRRI) has released several aromatic rice varieties over recent decades, and Bangladesh boasts several indigenous cultivars with competitive potential internationally. However, local demand often exceeds production capacity due to quality constraints. Furthermore, preserving rice to retain its aroma postharvest is challenging, as Bangladeshi rice growers typically aim to complete maturity before a potentially lengthy storage period [4], resulting in prolonged exposure to sunlight and internal field respiration, ultimately diminishing the grains’ natural scent over time [5].

Among the hundreds of volatile chemicals found in aromatic rice, 2-acetyl-1-pyrroline (2-AP) is a key aroma compound [6]. Although nongenetic factors, including nutritional management, environmental stresses, and planting and harvesting strategy modification, influence 2-AP content in aromatic rice, genetic factors predominately affect its regulation [7,8]. Arai and Itani [9] showed that delaying or advancing rice harvesting diminished the taste of cooked rice, emphasizing the substantial influence of grain harvesting timing on cooked rice quality. Indeed, understanding optimal harvesting time is crucial [10], as rice undergoes multiple growth stages from blossoming until maturity. Considering days after flowering (DAF), Jiamyangyuen et al. [11] divided these stages into five periods from blooming to fully ripening (0–35 DAF). Early or late harvesting may result in undesirable traits, such as immature or cracked kernels, leading to an increase in broken milled rice [12]. Harvesting rice before it reaches maturity may offer advantages over mature harvest, including shorter production time and improved grain quality. Early-harvested grains exhibit enhanced sweetness and delectability as cooked rice [13]. However, Hossain et al. [4] noted that early harvesting may lead to an increase in unfilled and immature grains, resulting in thickened bran and aleuronic acid layers and potentially generating partially chalky and milk-white kernels. Conversely, Zhu et al. [14] reported that delaying harvesting until after maturity could detrimentally affect grain processing, appearance, microstructure, and cooking quality. In addition, temperature markedly influences aromatic rice quality during the blooming, grain filling, and maturity stages [15]. Baktiar et al. [16] suggested that rice should be harvested 30–35 days postblooming to achieve superior grain quality in terms of head rice, elongation percentage, and amylose content. Jewel et al. [17] observed that harvesting 30–35 days postflowering positively influenced grain quality characteristics, effectively reducing shattering loss and immature stages. Bao et al. [18] demonstrated that delaying harvesting increases chalkiness and the brown rice rate while lowering milled rice and head rice rates. According to Dong et al. [19], postponing harvesting increased rice yields in various varieties, including Liangyoupeijiu, Wuyujing-3, Nanjing-44, and Nanjing-46. However, Verma and Srivastav [15] indicated that delaying harvesting until maturity may diminish aroma.

The aforementioned studies show that the harvest period substantially impacts both the aroma and grain yield of aromatic rice. Nevertheless, research on the global impact of rice harvesting time on grain quality remains limited, particularly regarding Bangladesh’s native, inbred aromatic rice. Furthermore, investigations into postharvest grain 2-AP concentration variation in aromatic rice are lacking, especially regarding Bangladeshi varieties. Therefore, this study aimed to determine the optimal harvest interval for high-quality aromatic rice, assessing grain yield, physicochemical qualities, and grain 2-AP concentration across several Bangladeshi cultivars.

2. Materials and Methods

2.1. Experimental Site

The experiment was conducted at the agronomy research field of Sher-e-Bangla Agricultural University, Dhaka-1207 (23°7′ N latitude and 93° E longitude; 8.6 m above sea level). The research area is situated within the agro-ecological zone known as the “Madhupur Tract”, AEZ-28. The experimental site experiences a fluctuating sub-tropical climate during the Aman season, from 15 June 2022 to 25 November 2022. Monthly meteorological data and soil component analysis are provided in

Table 1. Monthly meteorological data obtained during the experimental period.

Year	Month	Air Temperature (°C)		Relative Humidity (%)	Total Rainfall (mm)
		Maximum	Minimum
2022 (Aman season)	June	26.5	10.71	69.59	279.9
	July	27.4	11.8	90.1	367.1
	August	24.9	14.7	79.5	333.8
	September	25.8	12.1	74.3	334.5
	October	24.3	10.5	81.9	80.5
	November	21.9	7.9	67.5	40.7

Source: Metrological Centre (Climate Division), Agargaon, Dhaka, Bangladesh.

and

Table 2. Analytical results for initial and postharvest soil from the pot experiment conducted in 2022.

Soil Constituents	Preplanting	Postharvesting
pH	5.59	5.66
Organic matter (%)	1.227	1.616
Total nitrogen (%)	0.125	0.311
K (meq/100 g soil)	0.138	0.169
P (mg/g soil)	5.75	6.98
S (mg/g soil)	21.71	23.38
B (mg/g soil)	0.41	0.47
Zn (mg/g soil)	3.15	3.64

Source: Soil Resource Development Institute, Farmgate, Dhaka, Bangladesh.

.

2.2. Experimental Treatments and Design

The experiment involved two factors: factor 1 comprised four rice varieties (V₁ = BRRI dhan34, V₂ = BRRI dhan70, V₃ = BRRI dhan80, and V₄ = Tulshimala), and factor 2 comprised three harvesting times [D₁ = 3 weeks after flowering (WAF), D₂ = 4 WAF, and D₃ = 5 WAF]. A randomized complete block design was employed, replicated three times, resulting in 36 plots, each of which measured 2.5 × 2.0 m.

2.3. Rice Varieties

Seeds of BRRI dhan34, BRRI dhan70, and BRRI dhan80 were collected from the Plant Genetic Resource division of BRRI in Joydebpur, Gazipur, whereas Tulshimala, a local cultivar, was collected from a farmer in Sherpur district, Bangladesh. BRRI dhan34 rice is small-grained, akin to chinigura or kalojira rice, suitable for cooking polao, with a lifecycle of 135 days. BRRI dhan70 rice features long-slender-aromatic grains, which are potentially exportable and exhibit a lifecycle of 130 days. BRRI dhan80 rice has a pleasant aroma similar to Thai Jasmine rice and is also exportable, with a lifecycle of 130–135 days. Tulshimala rice, similar in size to chinigura rice, possesses a pleasant aroma akin to Thai Jasmine rice, with a lifecycle of 140 days.

2.4. Agronomic Crop Husbandry

Healthy seeds were selected using a specific gravity method and soaked in water for a day before storage in bags. Germination occurred 48 h after sowing, with nutrient application tailored to soil requirements across a meter-wide seeding area. On 15 June 2022, seeds germinating at a rate of 70 g m⁻² were sown in the bed. At 25 days old, seedlings were transplanted into flask planters. Three seedlings were planted per hill. Following BRRI guidelines [20], all available nutrients were used in crop management. Plots were regularly treated with insecticides and fungicides as well as being watered and weeded. Flowering time was determined when 90% of panicles had flowered, with harvesting conducted based on treatment for each variety. Harvested grains were transported to the threshing floor in bundles, separated and clearly marked. For each pot, the grains were washed, dried, and weighed. Moisture content was increased by 12%, and weights were adjusted accordingly.

2.5. Data Collection

2.5.1. Grain Yield (t ha⁻¹)

Harvested grains from each plot were sun-dried and weighed using an electrical balance to determine total grain weight in kilograms. The yield per plot was then converted to the yield per hectare.

2.5.2. Brown Rice Yield (BRY)

After dehusking 100 g of paddy rice seeds using a conventional method, the whole grain BRY was determined as a percentage [21,22].

2.5.3. Head Rice Recovery (HRR)

Using 100 g of hulled rice grains with no obvious breakage, ¾ (i.e., 75% of whole grain) part of grain was employed to calculate the head rice yield. HRR percentage and broken rice were determined using a traditional formula.

$$\mathrm{Head}\mathrm{rice}\mathrm{recovery}(\%)={\displaystyle \frac{\mathrm{W}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t}\mathrm{o}\mathrm{f}\mathrm{t}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l}\mathrm{h}\mathrm{e}\mathrm{a}\mathrm{d}\mathrm{r}\mathrm{i}\mathrm{c}\mathrm{e}}{\mathrm{W}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t}\mathrm{o}\mathrm{f}\mathrm{t}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l}\mathrm{d}\mathrm{e}\mathrm{h}\mathrm{u}\mathrm{l}\mathrm{l}\mathrm{e}\mathrm{d}\mathrm{r}\mathrm{i}\mathrm{c}\mathrm{e}}}\times 100$$

2.5.4. Chalk Index Determination (CID)

A light box was used to detect chalk content in 10 hulled rice grains above 50% of detecting chalk throughout the whole kernel, with the chalk percentage calculated based on grain weight [21,22].

2.5.5. Water Uptake Ratio (WUR)

To determine the WUR, 2 g of each type of whole grain rice was cooked in 20 mL of distilled water in a boiling water bath for the shortest possible duration, following the establish procedure of Oko et al. [23]. Subsequently, surface water was drained from the cooked rice, and cooked samples were weighed. The WUR was calculated as follows:

$$\mathrm{WUR}={\displaystyle \frac{\mathrm{C}\mathrm{o}\mathrm{o}\mathrm{k}\mathrm{e}\mathrm{d}\mathrm{r}\mathrm{i}\mathrm{c}\mathrm{e}\mathrm{w}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t}}{\mathrm{U}\mathrm{n}\mathrm{c}\mathrm{o}\mathrm{o}\mathrm{k}\mathrm{e}\mathrm{d}\mathrm{r}\mathrm{i}\mathrm{c}\mathrm{e}\mathrm{w}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t}}}$$

2.5.6. Imbibition Ratio (IR)

Initially, 4 g of head rice was cooked in 40 mL of distilled water for at least 1 h in a boiling water bath. After draining the water, cooked rice samples were pressed between filter paper to remove any remaining surface water. Finally, the samples were weighed to determine the IR [24].

2.5.7. Optimal Cooking Time (OCT)

Samples of milled rice (5 g) of each variety were placed in a graduated cylinder filled with 5 mL of water, which was placed in a water bath. Cooking times were based on gelatinization of 90% of rice kernels, which was determined by pushing selected kernels between two glass plates at various points during cooking [24].

2.5.8. Kernel Elongation Ratio (KER)

The KER was calculated as follows [24,25]:

$${\displaystyle \frac{\mathrm{C}\mathrm{u}\mathrm{m}\mathrm{u}\mathrm{l}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{v}\mathrm{e}\mathrm{l}\mathrm{e}\mathrm{n}\mathrm{g}\mathrm{t}\mathrm{h}\mathrm{o}\mathrm{f}10\mathrm{r}\mathrm{a}\mathrm{w}\mathrm{u}\mathrm{n}\mathrm{c}\mathrm{o}\mathrm{o}\mathrm{k}\mathrm{e}\mathrm{d}\mathrm{r}\mathrm{i}\mathrm{c}\mathrm{e}\mathrm{g}\mathrm{r}\mathrm{a}\mathrm{i}\mathrm{n}\mathrm{s}}{\mathrm{C}\mathrm{u}\mathrm{m}\mathrm{u}\mathrm{l}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{v}\mathrm{e}\mathrm{l}\mathrm{e}\mathrm{n}\mathrm{g}\mathrm{t}\mathrm{h}\mathrm{o}\mathrm{f}10\mathrm{c}\mathrm{o}\mathrm{o}\mathrm{k}\mathrm{e}\mathrm{d}\mathrm{r}\mathrm{i}\mathrm{c}\mathrm{e}\mathrm{g}\mathrm{r}\mathrm{a}\mathrm{i}\mathrm{n}\mathrm{s}}}$$

2.5.9. Gel Consistency (GC)

GC was determined using the method outlined by Cagampang et al. [26]. Briefly, the length of the chilled gel in horizontally held culture tubes for 0.5–1.0 h was used to determine GC, which was <61 mm in all cases. Ten whole-milled rice grains were processed to obtained fine powder (100 mesh) using a Wig-L-Bug amalgamator for 40 s. In each tube, 100 mg of powder was added twice. Using a pipette, 2.0 mL of 0.2 M potassium hydroxide (KOH), 0.2 mL of 95% ethyl alcohol, and 0.025% thymol blue were added, with mixing achieved using a Vortex Genie mixer on the sixth speed. Samples were then heated for 8 min or until the water inside the tubes reached two-thirds height in a rapidly boiling water bath. Afterward, the test tubes were allowed to stand for 5 min at room temperature before cooling in an ice-water bath for 20 min. After being placed horizontally on millimeter graphing paper, the entire length from the tube’s bottom to the gel’s top was measured in millimeters. GC classification followed the criteria established by Tang et al. [27]: length > 61 mm, soft consistency; 60 > length > 41 mm: medium consistency; length < 40 mm: hard consistency.

2.5.10. Apparent Amylose Content (AAC)

Amylose content in rice samples was determined following the method of Juliano [28]. For starch gelatinization, boiling water was added to a volumetric flask containing 100 mg of milled rice, 1 mL of 1 M NaOH, and 1 mL of 95% ethanol. Subsequently, 5 mL of starch extract was added to a 100 mL volumetric flask, with 1 mL of 1 N acetic acid and 2 mL of iodide solution added to the starch extract to yield a total volume of 100 mL. Following shaking, the mixture was allowed to stand for 20 min, after which a potato amylose standard curve was used to calculate amylose content, which was expressed as a percentage. Absorbance was measured at 620 nm using an Agilent Technologies Cary 60 UV–VIS spectrophotometer (Santa Clara, CA, USA).

2.5.11. Amylopectin Content (APC)

Based on AAC, amylopectin content was calculated using the following equation [29]:

Amylopectin (%) = 100 − AAC (%).

2.5.12. Alkali Spreading Value (ASV) and Gelatinization Temperature (GT)

ASV was calculated following the method of Chemutai et al. [30]. Six fully ground uncracked cores were placed in a plastic box (5 × 5 × 2 cm) with 10 mL of 1.7% KOH solution. Sample placement allowed sufficient space for nuclei dispersal. The boxes were covered and baked for 23 h at 30 °C, after which the starchy endosperm was visually rated based on a seven-point numerical spreading scale. Spreading ratings of 1–3 indicated high (75–79 °C) gelatinization temperature, 4–5 indicated intermediate (70–74 °C) temperature, and 6–7 indicated low (55–69 °C) temperature [31].

2.5.13. Percent Protein Content

Protein content was determined using a near-infrared grain tester (AN-820, Kett Co., Ltd., Tokyo, Japan), as employed by Nakamura et al. [32].

2.5.14. Sensory Aroma Test

For 30 min at room temperature, 40 grains of each treatment were immersed in 10 mL of 1.7% KOH solution in a covered glass Petri dish. After removing the Petri dish lid, the participants immediately evaluated the grains’ aroma using a 4-point category scale for aroma intensity: 1 = no aroma, 2 = slight aroma, 3 = moderate aroma, and 4 = strong aroma. Mean aroma scores for each sample were classified as nonaromatic (<1.5), slightly aromatic (1.5–2.5), moderately aromatic (2.5–3.5), and highly aromatic (>3.5) [33,34].

2.5.15. Grain 2-AP Content (μg g⁻¹)

Before analysis, grains were pulverized using a mortar and pestle, following the method of Huang et al. [35] to determine 2-AP concentration. An extraction head was connected to a 500 mL round-bottom flask for continuous steam distillation, which was conducted using 150 mL of filtered water and approximately 10 g of grains, with the mixture heated to 150 °C in an oil pot. The extraction solvent, 30 mL of dichloromethane in a 500 mL round-bottom flask, was connected to the second head of the continuous steam distillation system, and the flask was heated to 53 °C in a water pot. To maintain a temperature of 10 °C, the continuous steam distillation extraction system was connected to a cold water circulation unit. Extraction lasted around 25 min. For water absorption, anhydrous sodium sulfite was added to the extract. The dried extract was filtered using an organic needle filter, and the GCMS-QP2010 Plus system (Shimadzu Corporation, Kyoto, Japan) was employed to determine 2-AP content with a carrier gas flow of 2 mL of premium helium min⁻¹. The gas chromatography oven temperature gradient was programmed as follows: 40 °C (1 min), increased at 2 °C min⁻¹ to 65 °C and maintained at 65 °C for 1 min, then raised to 220 °C at 10 °C min⁻¹, and finally held at 220 °C for 10 min. A validation study confirmed a 2-AP retention time of 7.5 min, with grain 2-AP concentration measured in µg g⁻¹.

2.5.16. Percent Grain Yield and Grain 2-AP Content Increases

The percent grain yield increase was calculated based on total yield per hectare, whereas the percent grain 2-AP content increase was determined based on 2-AP content from different harvesting times using the following formula:

$$\mathrm{Grain}\mathrm{yield}/\mathrm{Grain}\mathrm{2-AP}\mathrm{increase}(\%)={\displaystyle \frac{\begin{array}{c}\\ (\mathrm{V}\mathrm{a}\mathrm{l}\mathrm{u}\mathrm{e}\mathrm{o}\mathrm{f}\mathrm{e}\mathrm{x}\mathrm{p}\mathrm{e}\mathrm{c}\mathrm{t}\mathrm{e}\mathrm{d}\mathrm{t}\mathrm{r}\mathrm{e}\mathrm{a}\mathrm{t}\mathrm{m}\mathrm{e}\mathrm{n}\mathrm{t}-\mathrm{v}\mathrm{a}\mathrm{l}\mathrm{u}\mathrm{e}\mathrm{o}\mathrm{f}\mathrm{c}\mathrm{o}\mathrm{m}\mathrm{p}\mathrm{a}\mathrm{r}\mathrm{e}\mathrm{d}\mathrm{t}\mathrm{r}\mathrm{e}\mathrm{a}\mathrm{t}\mathrm{m}\mathrm{e}\mathrm{n}\mathrm{t})\end{array}}{\mathrm{V}\mathrm{a}\mathrm{l}\mathrm{u}\mathrm{e}\mathrm{o}\mathrm{f}\mathrm{c}\mathrm{o}\mathrm{m}\mathrm{p}\mathrm{a}\mathrm{r}\mathrm{e}\mathrm{d}\mathrm{t}\mathrm{r}\mathrm{e}\mathrm{a}\mathrm{t}\mathrm{m}\mathrm{e}\mathrm{n}\mathrm{t}}}\times 100$$

2.5.17. Monetary Advantage

Monetary advantage for aromatic rice was calculated based on total grain yield per hectare given different harvesting times. Grain yield was multiplied by 1000 for conversion into kilograms, and this value was multiplied by the market price for 1 kg of aromatic rice, estimated at BDT 160 (Tk.) kg⁻¹. Finally, the total monetary return was converted into US dollar (USD).

2.6. Statistical Analysis

Two ways ANOVA technique was conducted using Statistix-10 (Analytical Software, Tallahassee, FL, USA). Data for various parameters underwent mean correction via least significant difference (LSD) correction at a 5% significance level. Correlation analysis was performed using Statistix-10 to examine relationships between grain quality attributes based on mean parameter values.

3. Results

3.1. Grain Yield

Harvesting time and rice variety significantly impacted aromatic rice grain yield (

Table 3. Effect of variety and harvesting time on aromatic rice yield and milling traits.

Treatments	Grain Yield (t ha−1)	Brown Rice Yield (%)	Head Rice Recovery (%)	Chalk Index Determination (%)
V1D1	1.97 g	70.2 f	69.1	32.0 d
V1D2	2.21 fg	77.4 bcd	70.0	36.5 c
V1D3	2.53 ef	81.7 ab	70.2	40.7 b
V2D1	4.01 c	70.3 f	66.3	30.6 d
V2D2	4.31 bc	72.2 ef	69.9	37.0 c
V2D3	5.09 a	84.0 a	73.1	43.9 a
V3D1	4.13 c	73.9 def	68.8	30.3 d
V3D2	4.57 b	81.7 ab	70.2	36.0 c
V3D3	5.12 a	80.1 abc	73.6	44.2 a
V4D1	2.79 e	76.4 cde	68.8	30.4 d
V4D2	3.15 d	74.3 def	69.2	44.5 a
V4D3	3.19 d	82.3 a	73.0	44.8 a
CV (%)	5.81	3.37	5.24	3.58
LSD (0.05)	0.353	4.393	---	2.281
F-test	*	**	NS	**
Varieties	**	NS	NS	**
Harvesting time	**	**	*	**

In columns, means with the same letter are not statistically different, whereas those with different letters differ significantly at the 0.05 probability level. *: significant at the 5% probability level; **: significant at the 1% probability level; NS, nonsignificant. V₁ = BRRI dhan34, V₂ = BRRI dhan70, V₃ = BRRI dhan80, V₄ = Tulshimala. D₁ = 3 WAF, D₂ = 4 WAF, D₃ = 5 WAF. Coefficient of variation (CV); Least significant difference (LSD).

). The highest grain yield was observed in V₃D₃ (5.12 t ha⁻¹), which was statistically similar to V₂D₃ (5.09 t ha⁻¹). The lowest yield occurred in V₁D₁ (1.97 t ha⁻¹). Late harvesting increased grain yield compared with early harvesting.

3.2. Brown Rice Yield

Significant regulation occurred in BRY in response to rice variety and harvesting time (Table 3). The highest BRY was observed in V₂D₃ (84.0%), which was statistically similar to V₄D₃ (82.3%), V₃D₃ (80.1%), V₃D₂ (81.7%), and V₁D₃ (81.7%), whereas the lowest yield occurred in V₁D₁ (70.2%). Early harvesting led to decreased BRY.

3.3. Head Rice Recovery

A nonsignificant response in HRR was observed due to varieties and harvesting time (Table 3), but numerically the highest HRR was observed in V₃D₃ (73.6%), whereas the lowest HRR was observed in V₂D₁ (66.3%).

3.4. Chalk Index Determination

The presence of chalk reduces the market value and acceptance of aromatic rice. A significant response in CID occurred due to rice varieties and harvesting times (Table 3). The highest CID was found in V₄D₃ (44.8%), which was statistically similar to V₄D₂ (44.5%), V₃D₃ (44.2%), and V₂D₃ (43.9%). The lowest CID was observed in treatments where rice varieties were harvested at three WAF. Therefore, chalkiness increased with prolonged harvesting time.

3.5. Water Uptake Ratio

The cooking quality of rice is influenced by its water absorption during boiling, affecting grain expansion and cooking time. A notable effect was observed in WUR compared across different harvesting times (

Table 4. Effects of variety and harvesting time on various aromatic rice cooking traits.

Treatments	Water Uptake Ratio	Imbibition Ratio	Optimal Cooking Time (min)	Kernel Elongation Ratio	Gel Consistency (mm)
V1D1	6.7633 b	4.0100 e	17.670 a	1.4400 d	70.447 g
V1D2	6.2167 c	4.6200 bc	16.240 bc	1.9500 b	75.347 ef
V1D3	5.1533 d	4.7900 b	15.450 c	2.4400 a	78.207 e
V2D1	6.6033 b	4.2200 de	17.910 a	1.6700 c	72.387 fg
V2D2	6.0667 c	4.4500 cd	16.110 bc	1.8800 b	85.187 cd
V2D3	4.5133 e	5.2100 a	12.950 d	2.4100 a	88.707 bc
V3D1	6.7533 b	4.5000 c	18.150 a	1.3200 d	77.963 e
V3D2	4.2500 f	5.1400 a	17.010 ab	2.4600 a	93.963 a
V3D3	4.4533 ef	5.2100 a	12.980 d	2.4800 a	89.840 b
V4D1	7.6667 a	4.6200 bc	18.090 a	1.3900 d	84.200 d
V4D2	4.4933 e	5.1100 a	13.010 d	2.4900 a	89.983 b
V4D3	4.5533 e	5.2300 a	12.920 d	2.5000 a	94.300 a
CV (%)	2.25	3.45	4.46	4.75	2.82
LSD (0.05)	0.2146	0.2780	1.1864	0.1639	3.9766
F-test	**	*	**	**	**
Varieties	**	**	**	**	**
Harvesting time	**	**	**	**	**

In columns, means with the same letter are not statistically different, whereas those with different letters differ significantly at the 0.05 probability level. *: significant at the 5% probability level; **: significant at the 1% probability level. V₁ = BRRI dhan34, V₂ = BRRI dhan70, V₃ = BRRI dhan80, V₄ = Tulshimala. D₁ = 3 WAF, D₂ = 4 WAF, D₃ = 5 WAF. Coefficient of variation (CV); Least significant difference (LSD).

). The highest WUR occurred in V₄D₁ (7.6667), whereas the lowest WURs were found in V₃D₂ (4.2500) and V₃D₃ (4.4533).

3.6. Imbibition Ratio

The quality of aromatic rice can be gauged by measuring the time taken for the cooked rice grain to expand after milling. This measurement also indicates water adsorption by rice kernels based on the IR. Variety and harvesting time significantly influenced the IR (Table 4). The highest IR occurred in V₄D₃ (5.2300), which was statistically similar to V₄D₂ (5.1100), V₃D₃ (5.2100), V₃D₂ (5.1400), and V₂D₃ (5.2100). The lowest IRs were found in V₁D₁ (4.0100) and V₂D₁ (4.2200).

3.7. Optimal Cooking Time

Cultural cooking practices impact rice cooking quality, with cooking time being a crucial element. Variations in harvesting time and rice variety significantly impacted cooking time (Table 4). The shortest cooking time was required for rice from V₄D₃ (12.920 min), which was statistically similar to V₄D₂ (13.010 min), V₃D₃ (12.980 min), and V₂D₃ (12.950 min). The longest cooking was required for rice from V₃D₁ (18.150 min), which was statistically similar to treatments where all varieties were harvested at 3 WAF. Shorter cooking times are considered best practice for preparing aromatic rice.

3.8. Kernel Elongation Ratio

Kernel elongation ratio is key indicator of cooking quality, comparing cooked and uncooked rice lengths. Both harvesting time and variety substantially influenced KER (Table 4). The highest KER occurred in V₄D₃ (2.5000), which was statistically similar to V₄D₂ (2.4900), V₃D₃ (2.4800), V₃D₂ (2.4600), V₂D₃ (2.4100), and V₁D₃ (2.4400). The lowest KER was found in treatments where BRRI dhan34, BRRI dhan80, and Tulshimala were harvested at three WAF. Kernels cooked for longer are typically preferred by consumers.

3.9. Gel Consistency

Gel consistency, which regulates aromatic rice cooking quality, was significantly impacted by variety and harvesting time (Table 4). The highest GC was observed in V₄D₃ (94.300 mm), which was statistically similar to V₃D₂ (93.963 mm). The lowest GC was recorded in V₁D₁ (70.447 mm). In case of Bangladesh, the people prefer dry nature of cooked rice. So, high GC is good for the Bangladeshi nationals.

3.10. Apparent Amylose Content

Apparent amylose content affects the quality of cooked rice kernels, particularly their texture when chewed. Harvesting period and variety notably influenced AAC (

Table 5. Effects of variety and harvesting time on various aromatic rice physicochemical traits.

Treatments	Apparent Amylose Content (%)	Amylopectin Content (%)	Alkali Spreading Value	Gelatinization Temperature	Protein Content (%)
V1D1	24.917 a	75.083 e	3.0100 f	High	6.8797 cd
V1D2	24.727 ab	75.273 de	5.1100 c	Intermediate	7.2197 cd
V1D3	24.617 abc	75.383 cde	6.8700 a	Low	8.5697 b
V2D1	23.647 a–d	76.353 b–e	4.0100 e	Intermediate	6.5597 d
V2D2	23.477 bcd	76.523 bcd	4.4400 d	Intermediate	7.3697 c
V2D3	23.157 d	76.843 b	4.4900 d	Intermediate	8.5597 b
V3D1	23.807 a–d	76.193 b–e	3.3800 f	High	6.6597 d
V3D2	20.017 e	79.983 a	5.5200 b	Intermediate	9.6897 a
V3D3	20.047 e	79.953 a	6.8500 a	Low	9.7097 a
V4D1	23.407 cd	76.593 bc	3.1900 f	High	7.0697 cd
V4D2	20.087 e	79.913 a	4.5600 d	Intermediate	9.7497 a
V4D3	20.010 e	79.990 a	6.8800 a	Low	9.7797 a
CV (%)	3.36	0.98	4.77	-------	4.85
LSD (0.05)	1.2889	1.2889	0.3928	-------	0.6696
F-test	**	**	**	-------	**
Varieties	**	**	**	-------	**
Harvesting time	**	**	**	-------	**

In columns, means with the same letter are not statistically different, whereas those with different letters differ significantly at the 0.05 probability level. **: significant at the 1% probability level. V₁ = BRRI dhan34, V₂ = BRRI dhan70, V₃ = BRRI dhan80, V₄ = Tulshimala. D₁ = 3 WAF, D₂ = 4 WAF, D₃ = 5 WAF. Coefficient of variation (CV); Least significant difference (LSD).

). The highest AAC was found in rice from V₁D₁ (24.917%), which was statistically similar to V₁D₂ (24.727%), V₁D₃ (24.617%), V₂D₁ (23.647%), and V₃D₁ (23.807%). The lowest AAC was recorded in rice from V₄D₃ (20.010%), which was statistically similar to V₄D₂ (20.087%), V₃D₃ (20.047%), and V₃D₂ (20.017%). Lower amylose content increases kernel stickiness, a trait preferred by aromatic rice consumers.

3.11. Amylopectin Content

Cooked rice quality, especially its texture, is influenced by amylopectin levels, with rice quality inversely correlated with amylose and amylopectin content. Harvesting time and varieties notably affected perceived APC (Table 5). The highest APC was found in rice from V₄D₃ (79.990%), which was statistically similar to V₄D₂, V₃D₃, and V₃D₂, whereas the lowest APC was recorded in rice from V₁D₁ (75.083%).

3.12. Alkali Spreading Value

The ASV is used to calculate GT, a key indicator of cooking quality. According to Oko et al. [23], a low ASV indicates a high GT, and vice versa. Harvesting times substantially impacted ASV (Table 5). The highest ASV occurred in V₄D₃ (6.8800), which was statistically similar to V₃D₃ (6.8500), and V₁D₃ (6.8700). The lowest ASV was recorded in treatments where BRRI dhan34, BRRI dhan80, and Tulshimala were harvested at three WAF.

3.13. Gelatinization Temperature

Based on ASV data, GT was evaluated, reflecting aromatic rice’s cooking quality. Harvesting time and variety significantly impacted GT (Table 5). A low GT was found in V₄D₃, V₃D₃, and V₁D₃ treatments, whereas a high GT was observed in V₄D₁, V₃D₁, and V₁D₁. All other treatments exhibited an intermediate GT.

3.14. Protein Content

The uptake of aromatic rice is severely constrained by its grain protein content, which is important when considering potential health implications. Both variety and harvesting time significantly affected protein content (Table 5). The highest protein content was found in rice from V₄D₃ (9.7797%), which was statistically similar to V₄D₂ (9.7497%), V₃D₃ (9.7497%), and V₃D₂ (9.6897%). The lowest protein content was recorded in V₂D₁ (6.5597%) and V₃D₁ (6.6597%).

3.15. Sensory Aroma Test

Given that 2-AP serves as the precursor of grain scent, the level of 2-AP in aromatic rice grains can be inferred from a random panel test evaluating grain aroma. A stronger grain aroma is typically associated with a higher grain 2-AP content. Harvesting time and variety substantially impacted grain aroma (

Table 6. Effects of variety and harvesting time on sensory aroma scores and grain 2-acetyl-1-pyrroline (2-AP) content in aromatic rice.

Treatments	Sensory Aroma Score	Grain 2-AP Content (µg g−1)
V1D1	3.8897 a	0.0707 ef
V1D2	3.8697 a	0.0727 de
V1D3	2.8597 c	0.0587 g
V2D1	3.8997 a	0.0948 c
V2D2	3.8897 a	0.0777 d
V2D3	2.4097 d	0.0657 f
V3D1	3.9097 a	0.1447 ab
V3D2	3.8997 a	0.1437 ab
V3D3	2.3197 d	0.0887 c
V4D1	3.9497 a	0.1507 a
V4D2	3.9397 a	0.1487 ab
V4D3	3.4097 b	0.0747 de
CV (%)	3.46	4.10
LSD (0.05)	0.2060	0.0231
F-test	**	**
Varieties	**	**
Harvesting time	**	**

In columns, means with the same letter are not statistically different, whereas those with different letters differ significantly at the 0.05 probability level. **: significant at the 1% probability level. V₁ = BRRI dhan34, V₂ = BRRI dhan70, V₃ = BRRI dhan80, V₄ = Tulshimala. D₁ = 3 WAF, D₂ = 4 WAF, D₃ = 5 WAF. Coefficient of variation (CV); Least significant difference (LSD).

). The highest sensory aroma score was found in treatments where all varieties were harvested at 3 and 4 WAF. Thereafter, the sensory aroma score decreased for all rice varieties. The lowest sensory aroma scores were recorded for rice in V₃D₃ (2.3197) and V₂D₃ (2.4097).

3.16. Grain 2-AP Content

When aromatic rice is boiled, the grain 2-AP concentration contributes to its pleasant aroma, making rice variants with stronger scents more appealing. Notable variations in grain 2-AP content were observed due to variety and harvesting time (Table 6). High grain 2-AP content was detected in V₄D₁ (0.1507 µg g⁻¹), which was statistically similar to V₄D₂ (0.1487 µg g⁻¹), V₃D₁ (0.1447 µg g⁻¹), and V₃D₂ (0.1437 µg g⁻¹), whereas the lowest 2-AP content was recorded in V₁D₃ (0.0587 µg g⁻¹).

3.17. Increases in Grain Yield and Grain 2-AP Content

Harvesting time markedly affected the percentage increases in grain yield and grain 2-AP content in aromatic rice (

Table 7. Effect of harvesting time on aromatic rice’s percent increase in grain yield, percent increase in grain 2-AP content, and monetary advantage.

Harvesting Time	Grain Yield Increase (%)	Grain 2-AP Increase (%)	Monetary Advantage (USD ha−1)
D1	---	60.22	4701.97
D2	10.38	53.96	5190.44
D3	23.49	---	5806.49

D₁ = 3 WAF, D₂ = 4 WAF, and D₃ = 5 WAF.; --- sign is used to denote the check treatment.

). The highest percent yield increase was found in D₃ (23.49%) followed by D₂ (10.38%), with D₁ showing a lower yield increase. Thus, as harvesting time postflowering was extended, the grain yield increased. Conversely, the highest percent grain 2-AP content increase was observed in D₁ (60.22%) followed by D₂ (53.96%), with D₃ exhibiting the lowest increase. This suggests that early harvesting led to higher 2-AP content, whereas delayed harvesting enhanced grain yield. The present study also detected a visual grain color variation due to different harvesting time in aromatic rice varieties (Figure 1).

3.18. Monetary Advantage

Harvesting time markedly influenced the monetary advantages of aromatic rice production (Table 7). The highest economic return was obtained from D₃ (USD 5806.49) followed D₂ (USD 5190.44), with D₁ leading to the lowest return. Thus, later harvesting resulted in higher grain yield and improved the subsequent economic return compared with earlier harvesting.

3.19. Pearson Correlation Analysis

This study revealed that harvesting time substantially altered the aroma and physicochemical traits of aromatic rice. Various physicochemical traits and fragrance parameters exhibited either positive or negative associations (

Table 8. Correlation matrix for aromatic rice grain yield, physicochemical traits, and grain 2-AP content.

	GY	BRY	HRR	CID	WUR	IR	OCT	KER	GC	AAC	APC	ASV	PR	SA	G2-AP
GY	1.0000
BRY	0.2827 NS	1.0000
HRR	0.3282 NS	0.5567 *	1.0000
CID	0.2732 NS	0.6177 *	0.5937 *	1.0000
WUR	−0.4252 NS	−0.6063 *	−0.3197 NS	0.8358 **	1.0000
IR	0.4707 NS	0.7853 *	0.6605 *	0.8225 **	−0.7784 *	1.0000
OCT	−0.2444 NS	−0.4741 NS	−0.2315 NS	0.8884 **	0.8041 **	−0.6585 *	1.0000
KER	0.3248 NS	0.6878 *	0.5077 *	0.9094 **	0.9277 **	0.8492 **	−0.7590 *	1.0000
GC	0.5158 *	0.6521 *	0.4721 NS	0.6634 *	−0.6942 *	0.8624 **	−0.5791 *	0.6942 *	1.0000
AAC	−0.3943 NS	−0.2993 NS	−0.0144 NS	−0.5000 *	0.7127 *	−0.6171 *	0.6239 *	−0.5754 *	−0.7654 *	1.0000
APC	0.3943 NS	0.2993 NS	0.0144 NS	0.5000 *	−0.7127 *	0.6171 *	−0.6239 *	0.5754 *	0.7654 *	−1.0000 **	1.0000
ASV	0.1832 NS	0.6606 *	0.4756 NS	0.7395 *	−0.7197 *	0.6785 *	−0.5877 *	0.8330 *	0.4993 NS	−0.4331 NS	0.4331 NS	1.0000
PR	0.3282 NS	0.6518 *	0.5710 *	0.8531 **	−0.8733 **	0.9048 **	−0.6891 *	0.9186	0.8253 **	−0.7254 *	0.7254 *	0.7490 *	1.0000
SA	−0.4122 NS	−0.5809 *	−0.3099 NS	−0.6120 *	0.5621 *	−0.4814 NS	0.6761 *	−0.5393	−0.2747 NS	0.2161 NS	−0.2161 NS	−0.5592 *	−0.4114 NS	1.0000
G2-AP	0.1406 NS	−0.1602 NS	−0.1526 NS	−0.3082 NS	0.1579 NS	0.1201 NS	0.3178 NS	−0.2167	0.2646 NS	−0.3280 NS	0.3280 NS	−0.3703 NS	0.0597 NS	0.5120 *	1.0000

Abbreviations: GY, grain yield; BRY, brown rice yield; HRR, head rice recovery; WUR, water uptake ratio; IR, imbibition ratio; OCT, optimal cooking time; KER, kernel elongation ratio; GC, gel consistency; AAC, apparent amylose content; APC, amylopectin content; ASV, alkali spreading value; PR, protein content; SA, sensory aroma score; G2-AP, grain 2-AP content. *: Significant at p < 0.05 (n = 12); **: significant at p < 0.01 (n = 12); NS: nonsignificant (n = 12).

). A significant positive association was found between grain yield and GC. Negative associations were observed between BRY and WUR, and sensory aroma score, whereas positive correlations were found between BRY and HRR, CID, IR, KER, GC, ASV, protein content, and 2-AP content. Positive associations were found between HRR and CID, IR, KER, protein content, and 2-AP content. Negative associations were detected between CID and AAC, sensory aroma score, and 2-AP content, whereas positive correlations were observed between CID and rest other traits. WUR exhibited negative associations with IR, GC, APC, ASV, and protein content; however, it displayed positive correlations with OCT, KER, AAC, and sensory aroma score. IR exhibited positive associations with most of the characteristics but negative correlations with OCT, AAC, and sensory aroma score. OCT was negatively associated with most of the traits but positively correlated with AAC, sensory aroma score, and 2-AP content. KER exhibited negative associations with AAC, and sensory aroma score but positive correlations with the other characteristics. GC was positively correlated with APC, and protein content; however, it was negatively correlated with AAC. AAC was negatively correlated with protein content. APC was positively correlated with protein content. ASV was found to be positively correlated with protein content but negatively correlated with sensory aroma score. Protein content was positively correlated with 2-AP content but negatively correlated with sensory aroma score. Sensory aroma score and 2-AP content were positively correlated. Higher BRY and HRR percentages may be attributed to higher protein content in the kernel, potentially impacting grain intactness and reducing the amount of broken rice. Additionally, higher amylopectin content may reduce starch degradation into amylose, resulting in higher ASV and GC. The higher content may lead to increased levels of proline, a precursor of 2-AP, in the kernel, thereby enhancing sensory aroma.

4. Discussion

The findings of this study highlight the substantial impact of harvesting time on both the physicochemical characteristics and grain yield of aromatic rice cultivars. Despite genetic predispositions, harvesting time exerted control over grain productivity, aroma, and 2-AP content. Present research revealed an increase in grain production with extended harvesting periods, aligning with the findings of Dong et al. [36], who attributed this rise to prolonged starch and storage protein accumulation. Zhu et al. [14] similarly observed increased rice output with extended harvest periods. Liu et al. [37] noted a significant increase in 1000-grain weight, seed setting rate, and harvest duration during various phases of japonica rice harvest, supporting present findings. Early harvesting may affect grain filling by shortening the active filling period, potentially leading to undesirable panicle characteristics, such as low panicle fertility, abundant immature grains, and light 1000-grain weights, similar to the findings of Afifah et al. [38], Atapattu et al. [39], and Yang et al. [40]. Present observations regarding harvesting time also mirrored trends in BRY and HRR, consistent with findings from Atapattu et al. [39] regarding low head rice output at early harvest due to high proportions of immature, green, chalky grains that were easily broken during milling. They also reported that low-density, immature grains increased the likelihood of fracture during milling. Additionally, Hossain et al. [4] identified the optimal harvest time for fragrant rice to be 30–35 DAF to maximize head rice output, a finding reinforced by Atapattu et al. [39], who recorded the highest head rice production at 35 DAF. We found that the abundance of chalky kernels increases with longer harvested periods. This may be attributed to the drop in temperature in Bangladesh between November and December, leading to diminished aggregation of starch granules in rice kernels and a chalkier texture. Consistent with present findings, Zhang et al. [41] observed that chalkiness increased with an extended harvest period. However, Kabir et al. [42] noted temperature-dependent variations in chalkiness, with Zhu et al. [14] also reporting effects of temperature on kernel chalkiness in aromatic rice in China.

In the present study, later-harvested grains required less water, could be cooked faster, exhibited higher GC, and were more palatable compared with earlier-harvested grains. In contrast, Hossain et al. [4] found that early-harvested crops exhibited increased protein content, appearing to limit starch swelling at 25 DAF, as well as prolonged cooking times. Rice quality is impacted by harvest time, variety genetics, and growing environment [43]. Present findings indicated lower amylose levels in later-harvested grains, consistent with the findings of Zhang et al. [41]. Additionally, Atapattu et al. [39] observed a considerable decrease in amylose levels when the harvest period was extended from 25 to 40 DAF. Conversely, Hossain et al. [4] revealed that amylose content could be enhanced by harvesting fragrant rice between 30 and 35 DAF. Some Asians appreciate the stickiness that decreased amylose levels produce in cooked kernels.

Regarding protein content, the present study showed that rice varieties exhibited higher content with extended harvesting times. Arai and Itani [13] noted a comparable trend, observing a progressive rise in protein content in the rice varieties Koshihikari and Nakateshinsenbon as they matured. In contrast, Dong et al. [36] demonstrated a reduction in protein content during grain development, suggesting that early harvesting yielded higher protein content compared with later harvesting. According to Atapattu et al. [39], the protein content of rice kernels is initially high during early maturation but decreases around 30–35 days after heading. However, the present study postulates that a later harvest date may facilitate extended grain development, greater 1000-grain weight, and enhanced translocation of photosynthates toward grains, all contributing to higher protein content.

Previous research indicates that 2-AP is the primary component of fragrant rice aroma, with proline serving as a precursor to 2-AP biosynthesis. Aromatic rice can arise from genetic mutation or occur naturally [41]. The glutamate and ornithine pathways are used to synthesize proline, with P5CS in the glutamate pathway acting as a rate limiting enzyme [44]. In the current study, it is noted that both the indirect (sensory) and direct (2-AP) scent content in several rice varieties decreased as harvesting times increased. Reduced scent at later harvesting times may be attributed to the declining activity of P5CS in grains as the harvest period extends, consistent with the findings of Zhang et al. [41]. Additionally, Verma and Srivastav [15] reported that scent may be diminished if harvesting is delayed until maturity. Conversely, Itani et al. [45] found that levels of 2-AP, the primary fragrance component in brown rice, peaked after 4 or 5 weeks of heading.

Numerous microorganisms have also been shown to create the aromatic chemical 2-acetyl-1-pyrroline (2AP), including plant-growth-promoting rhizobacteria (PGPR) [46]. By fixing nitrogen from the atmosphere, rhizobacteria have the potential to increase plant uptake of nitrogen (N) and eventually make L-proline, one of the precursors needed for 2-AP synthesis, more readily available [47]. Several bacterial taxa, including Bacillus, Acinetobacter, Pseudomonas, Enterobacter, Micrococcus, Sinomonas, and Burkholderia, are able to boost the 2-AP compound in aromatic rice, according to Deshmukh et al. [46] and Chinachanta et al. [48]. Therefore, compared to nonaromatic rice, it can be firmly inferred that the rhizosphere of aromatic rice is a favorable environment for a bigger population of rhizobacteria that produce 2-AP. When harvesting late as opposed to early, there may be a decrease in the availability of rhizospheric microbial activities.

The enzymatic reactions and precursor hydrolysis activities that result in the synthesis of 2-acetyl-1-pyrroline in aromatic rice also depend on moisture [49]. Because of the predominant winter season, Bangladesh’s atmospheric moisture content decreases daily from October to December. Due to the lack of appropriate moisture, late harvesting of aromatic rice may result in a lower 2-AP concentration. The present study shows limitations and draws future attention to evaluate the performance of rice growth stages on grain quality and aroma along with harvesting time.

5. Conclusions

From the current study, it can be concluded that the yield, physicochemical characteristics, and aroma of several aromatic rice varieties are markedly influenced by varying harvesting times. The varieties BRRI dhan70 and BRRI dhan80 are recommended for higher grain production, with the crop harvested at 5 WAF. Conversely, Tulshimala and BRRI dhan80 exhibited superior milling, biochemical characteristics, and fragrance. To achieve higher grain 2-AP concentrations, harvesting the crop 3–4 WAF is recommended. It is suggested that the varieties Tulshimala and BRRI dhan80 have the potential to influence consumer preferences and may be suitable for future biotechnological studies. For rice growers seeking higher yields, harvesting the crop later could lead to enhanced economic returns. However, for those prioritizing aroma, harvesting the grain as soon as possible without prolonged field time is advisable.