Soil Moisture and Growth Rates During Peak Yield Accumulation of Cassava Genotypes for Drought and Full Irrigation Conditions

Abstract

Climate change causes unpredictable weather patterns, leading to more frequent and severe droughts. Investigating the effects of drought and irrigation on soil water status and the performance of various cassava genotypes can provide valuable insights for mitigating drought through designing appropriate genotypes and water management strategies. The objective of this research was to evaluate soil moisture, growth rates, and final yields (total dry weight, storage root dry weight, harvest index and starch yield) of six cassava genotypes cultivated under drought conditions during the late growth phase, as well as under full irrigation. The study utilized a split-plot randomized complete block design with four replications, conducted over two growing seasons (2022/2023 and 2023/2024). The main plots were assigned as two water regimes to prevent water movement between plots: full irrigation and drought treatments. The subplot consisted of six cassava genotypes. Measurements included soil properties before planting, weather data, soil moisture content, relative water content (RWC) in cassava leaves, and several growth rates: leaf growth rate (LGR), stem growth rate (SGR), storage root growth rate (SRGR), crop growth rate (CGR), relative growth rate (RGR), as well as final yields. The results revealed that low soil moisture contents for drought treatment led to variation in RWC, growth, and yield among cassava genotypes. Variations in soil and weather conditions between the 2022/2023 and 2023/2024 growing seasons resulted in differences in the performance of the genotypes. Kasetsart 50 (2022/2023) and CMR38–125–77 (2023/2024) were top performers under late drought stress regarding storage root dry weight and starch yield, showing vigorous recovery upon re-watering, evidenced by their significant increase in LGR (between 240 and 270 DAP) and their high RGR (240–360 DAP). Rayong 9 (2023/2024) demonstrated strong performance in both during the drought period (180–240 DAP), efficiently allocating resources under water scarcity, with SRGR and starch yield reduced by 26.4% and 9.5%, respectively, compared to full irrigation. These cassava genotypes are valuable genetic resources for cassava cultivation and can be used as parental material in breeding programs aimed at improving drought tolerance.

1. Introduction

Drought is one of the main phenomena that occurs in many cropping areas around the world [1]. It can significantly reduce crop yields, causing declines ranging from 20% to 83%, depending on the growth stage of cassava and the duration of the water shortage. It also negatively affects plant growth, leaf size, and leaf area. Drought during cassava’s late growth phase, which is a critical time for peak yield and starch accumulation, it not only inhibits aboveground growth but also critically compromises the translocation of stored sugars to the storage roots, ultimately reducing harvestable yield [2,3,4,5,6].

Thailand is a significant producer and exporter of cassava. Approximately 67% of cassava production is grown in the northeast region [7]. Farmers normally grow cassava under rainfed conditions. Although cassava can grow year-round, water is one of the most limiting factors in some planting seasons and areas. About 65% of cassava in Thailand is planted during the rainy season, which usually starts in May [7]. If planted in the rainy season, cassava is generally only affected by drought during the mid or late growth phases and can reduce yield up to 80% in Thailand, due to a reduction in the number and length of tubers as a result of the imposed slow crop development. In comparison, a yield reduction of approximately 40% has been reported in Ghana [8]. Harvesting can also be postponed from 12 months to 15 or 18 months, depending on the severity of the drought [6,8,9]. Therefore, the selection of appropriate cassava genotypes for this season is an alternative solution to help mitigate drought.

A comparison between cassava growth under full irrigation and drought conditions in different genotypes is crucial in the design of cassava genotypes for different water management. Furthermore, more research is needed that explores the impact of re-watering following exposure to natural drought conditions during the late growth phase, which corresponds to the onset of the rainfall pattern in northeastern Thailand, to obtain valuable insights into the adaptation and recovery mechanisms of cassava genotypes following drought stress. This information can help plant breeders provide suitable genotypes for limited water availability during the late growth phase to improve the yield. There are several parameters, such as leaf photosynthesis, starch content, and tuber yield [6], which have been reported as good indicators of how well cassava genotypes perform under drought conditions. Although measuring leaf photosynthesis, starch content, and tuber yield provides valuable information on mechanisms of crop adaptation, ultimately growth rate and final yield are paramount in order to select genotypes for specific regions or watering regimes. Cassava growth analysis includes evaluating leaf growth rate (LGR), stem growth rate (SGR), storage root growth rate (SRGR), crop growth rate (CGR), and relative growth rate (RGR). These growth analyses can also help develop selective tools for final yield and explain the impact of water supply on the growth of cassava [10]. Growth analysis has been studied, with research covering the effect of planting date on growth and yield of irrigated cassava in Thailand [11], the impact of off-season production after paddy rice [12], and the influence of different water management practices during the early growth phase [13]. While previous research provided valuable insights, a significant gap remains regarding growth analysis for cassava under drought conditions in the late growth phase (180–300 days after planting (DAP)), a period directly impacting yield due to high carbohydrate accumulation [2]. This study aimed to determine soil moisture, growth rates, and the final yield of six different cassava genotypes grown under drought during the late growth phase and under full irrigation.

2. Materials and Methods

2.1. Experimental Design and Plant Materials

The experiment was conducted under field conditions at Field Crop Research Station, Faculty of Agriculture, Khon Kaen University, Khon Kaen, Thailand (latitude 16°28′ N, longitude 102°48′ E, 200 m above sea level). The soil type in the experimental site is Yasothorn soil series (Yt: fine-loamy; siliceous, isohypothermic, Oxic Paleustults) [14]. The experiment was conducted from May 2022 to May 2023 and was repeated from May 2023 to May 2024.

Since the experimental field has a slope of approximately 2%, with the gradient decreasing from west to east, a split-plot design was applied within a randomized complete block design (RCBD) with four replications. The plots within the same replication were arranged at similar elevation levels. The main plot consisted of two water management practices that require the use of large plots to minimize water movement between adjacent plots, while six cassava genotypes were assigned to the subplots with 70 plants per subplot. The two water regimes were: full irrigation, and drought conditions during the dry period. The six cassava genotypes had different morphological and physiological characteristics (

Table 1. Characteristics of six cassava genotypes used in the study.

No.	Genotype	Parent	Characteristic	Source
1	Kasetsart 50	Rayong 1 × Rayong 90	Highly adapted to the environment, high yield, and high starch content	[14,15]
2	Rayong 9	CMR31–19–23 × OMR29–20–118	High yield and high starch content	[10,15]
3	Rayong 72	Rayong 1 × Rayong 5	High-yielding and drought-tolerant	[10,15]
4	CMR38–125–77	RY5 × KU50	High yield and high starch content	[15]
5	CMR35–91–63	OMR29–20–118 × CMR25–105–112	High yield	[15]
6	CM523–7	COL655A × COL1515	Low-yielding and drought-tolerant	[15,16]

Note. RY = Rayong; CMR = Crossed Manihot Rayong; CM = Crossed Manihot; OMR = Open-pollinated Manihot Rayong.

). The genotypes Kasetsart 50, Rayong 9, and Rayong 72 are cassava commercial cultivars in Thailand. The genotypes CMR38–125–77, CMR35–91–63, and CM523–7 are breeding lines obtained from the Department of Agriculture, Thailand.

The experimental field was plowed three times, and soil ridges were created with a spacing of 1 m between ridges before planting. The stems for six cassava genotypes were cut into 20 cm-long sticks. To prevent mealybugs, the cassava sticks were then soaked for 20 min in a solution of thiamethoxam (3-(2-chloro-thiazol-5-ylmethyl)-5-methyl-(1,3,5)-oxadiazinan-4-ylidene-N-nitroamine 25% WG) water dispersible granules at a rate of 4 g per 20 L of water (Syngenta crop protection limited, Bangkok, Thailand). The sticks were planted by vertically placing two-thirds of the cutting in the soil at a spacing of 1 × 1 m.

After 30 days of planting, fertilizer was applied based on chemical and physical soil properties following [17]. At 60 DAP, the fertilizers carbonic diamide (CH₄N₂O; Saksiam group, Nonthaburi, Thailand) and potassium chloride (KCl; Saksiam group, Nonthaburi, Thailand) were applied at a rate of 46.9 and 56.3 kg ha⁻¹, respectively. In addition, the compound fertilizer of N–P₂O₅–K₂O formula 15–7–18 was applied at a rate of 312.5 kg ha⁻¹ at 60 DAP [18] (Saksiam group, Nonthaburi, Thailand). Prior to planting, soil samples were collected from two points per replication at depths of 0–30 cm and 30–60 cm to assess their physical and chemical properties (

Table 2. Physical and chemical properties of soils before planting at 0–30 cm and 30–60 cm depth.

Soil Property	2022/2023		2023/2024
	0−30 cm	30−60 cm	0−30 cm	30−60 cm
Physical property
Texture class	Sandy loam	Sandy loam	Sandy loam	Sandy loam
Sand (%)	67.5	60.3	73.7	73.4
Silt (%)	24.7	25.4	21.1	19.3
Clay (%)	7.9	14.3	5.2	7.3
Chemical property
pH (1:1 H2O)	6.2	6.6	5.4	5.5
CEC (cmol kg−1)	5.6	5.8	2.5	3.2
EC (dS m−1)	0.6	1.3	0.4	0.3
OM (mg kg−1)	2900	2700	600	600
Total N (mg kg−1)	300	200	300	300
Available P (mg kg−1)	31.8	7.5	28.1	18.8
Exchangeable K (mg kg−1)	28.9	12.8	27.0	23.9

Note. pH = potential of hydrogen; CEC = cation exchange capacity; EC = electrical conductivity; OM = organic matter; N = nitrogen; P = phosphorus; K = potassium.

). The soil texture at the Field Crop Research Station of Khon Kaen University was a sandy loam, the values for soil pH ranged from 5.42 to 6.61, total nitrogen varied from 200 to 300 mg kg⁻¹, available phosphorus was between 7.50 and 31.75 mg kg⁻¹, and exchangeable potassium varied from 12.83 to 28.87 mg kg⁻¹. The soil chemical analysis indicated low total nitrogen and exchangeable potassium.

Weed control was done manually throughout the crop establishment phase (0–90 DAP). Mini-sprinkler irrigation system was applied to all experimental plots to maintain soil moisture at field capacity (the amount of soil moisture content retained in the soil after excess water has drained away due to gravity, 9.0%) to a depth of 60 cm from 0 to 180 DAP. The drought treatments during 180–240 DAP were discontinued to receive water. After 240 DAP, drought plots were watered again to assess cassava re-growth and plant recovery in comparison with irrigation plots (Figure 1). The rehydration-driven recovery phase of cassava is timed to the onset of the rains and represents a key determinant of both crop growth rates and final yield.

Crop water requirement (ETcrop) was calculated using the formula of Doorenbos and Pruitt [19]:

ET crop = ETo × Kc

(1)

where ETcrop is the crop water requirement (mm day⁻¹), ETo is the evapotranspiration of a reference plant calculated using the pan evaporation method [19], and Kc is the crop water requirement coefficient, which varies depending on the crop growth stage. The Kc values in cassava were followed by Ruangyos et al. [20].

2.2. Data Collection

Soil moisture for each experimental plot was measured at 180, 210, 240, 270, and 360 DAP at depths of 0–30 cm and 30–60 cm. The soil samples were oven-dried at 105 °C for 72 h or until weights were constant, and the moisture percentage was calculated. Soil moisture was determined by the gravimetric method described by Shukla et al. [21] as below. The field capacity and permanent wilting point were determined using a wide-range pF meter and the pressure plate method in the laboratory.

$$\mathrm{S}\mathrm{o}\mathrm{i}\mathrm{l} \mathrm{m}\mathrm{o}\mathrm{i}\mathrm{s}\mathrm{t}\mathrm{u}\mathrm{r}\mathrm{e} \mathrm{c}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{e}\mathrm{n}\mathrm{t}\left(\mathrm{\%}\right)={\displaystyle \frac{\mathrm{S}\mathrm{o}\mathrm{i}\mathrm{l} \mathrm{w}\mathrm{e}\mathrm{t} \mathrm{w}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t}\left(\mathrm{g}\right)-\mathrm{S}\mathrm{o}\mathrm{i}\mathrm{l} \mathrm{d}\mathrm{r}\mathrm{y} \mathrm{w}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t} \left(\mathrm{g}\right)}{\mathrm{S}\mathrm{o}\mathrm{i}\mathrm{l} \mathrm{d}\mathrm{r}\mathrm{y} \mathrm{w}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t} \left(\mathrm{g}\right)}}\times 100$$

(2)

RWC was measured at 180, 210, 240, 270, and 360 DAP using the second or third fully expanded leaves from the top of the main stem of two plants from each plot from 9.00 a.m. to 12.00 p.m. The leaf discs were sampled from 10 leaves per plot, with 1 cm² in size. Leaf fresh weight was determined in the laboratory, and was soaked in distilled water for 24 h, and was weighed to get the turgid weight. Sampled leaves were oven-dried at 80 °C until constant weight, then weighed. RWC was calculated following the formula described by [19,22] as below:

$$\mathrm{R}\mathrm{W}\mathrm{C}\left(\mathrm{\%}\right)={\displaystyle \frac{\mathrm{F}\mathrm{r}\mathrm{e}\mathrm{s}\mathrm{h} \mathrm{w}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t}\left(\mathrm{g}\right)-\mathrm{D}\mathrm{r}\mathrm{y} \mathrm{w}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t} \left(\mathrm{g}\right)}{\mathrm{T}\mathrm{u}\mathrm{r}\mathrm{g}\mathrm{i}\mathrm{d} \mathrm{w}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t}\left(\mathrm{g}\right)-\mathrm{D}\mathrm{r}\mathrm{y} \mathrm{w}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t} \left(\mathrm{g}\right)}}\times 100$$

(3)

Crop data were collected from two plants of each plot at 180, 210, 240, 270, and 360 DAP. The plants were separated into leaves, stems, storage roots, and fibrous roots. All plant parts were subsampled (about 10% of the total fresh weight of each organ). A subsample of fresh leaves was then used to measure leaf area by using a leaf area meter (LI-3100, LI-Cor Inc., Lincoln, NE, USA). Subsamples were oven-dried at 80 °C to achieve a constant dry weight. The harvest index (HI) was calculated as the ratio of storage root dry weight to total dry weight. Calculations for CGR during 180–240 DAP, 240–270 DAP, and 270–360 DAP were performed based on the function below [12]:

$$\mathrm{C}\mathrm{G}\mathrm{R}\left(\mathrm{g} {\mathrm{m}}^{-2} {\mathrm{d}}^{-1}\right)={\displaystyle \frac{1}{\mathrm{G}}}\times {\displaystyle \frac{{\mathrm{D}\mathrm{W}}_2-{\mathrm{D}\mathrm{W}}_1}{{\mathrm{T}}_2-{\mathrm{T}}_1}}$$

(4)

where G is the sample area (2 m²) and DW₁ and DW₂ are the crop dry weight (g) of two sampled plants at the sampling times T₁ and T₂ (d). The equations for calculating LGR, SGR, and SRGR were based on the calculation concept of CGR.

Relative growth rate (RGR) was calculated using the following equation [12]:

$$\mathrm{R}\mathrm{G}\mathrm{R}\left(\mathrm{g} {\mathrm{g}}^{-1} {\mathrm{d}}^{-1}\right)={\displaystyle \frac{\mathrm{I}\mathrm{n}\left({\mathrm{D}\mathrm{W}}_2\right)-\mathrm{I}\mathrm{n} \left({\mathrm{D}\mathrm{W}}_1\right)}{{\mathrm{T}}_2-{\mathrm{T}}_1}}$$

(5)

where DW₁ and DW₂ are crop dry weight (g) at the times T₁ and T₂ (d).

Cassava storage roots weighing approximately 5 kg were sampled at final harvest for starch content determination. The starch content of the storage root was determined with the specific gravity method [12], and starch yield was calculated using the following equation:

Starch yield (t ha⁻¹) = starch content × storage root dry weight

(6)

2.3. Statistical Analysis

The homogeneity of variance was assessed, and analysis of variance (ANOVA) was performed to determine the significance of the main effects and interaction between water regimes and cassava genotypes, using a model for split-plot in RCBD [23,24]. The model was described the following equation:

Y_hij = μ + ρ_h + α_i + δ_hi + β_j + αβ_ij + ε_hij

(7)

where Y_hij is the observed response, μ is a general mean effect, ρ_h represents the replicate effect, α_i and β_j are the main effects of water management practices and cassava genotypes, respectively, δ_hi indicates the whole plot random error term, αβ_ij is the interaction between water management practices and cassava genotypes, and ε_hij is a split plot random error effect.

Mean comparisons were conducted based on the least significant difference test (LSD) at p ≤ 0.05. All statistical analyses were carried out by using the statistix10 program (Analytical Software, Tallahassee, FL, USA) [25].

3. Results

3.1. Climatic Conditions

Climatic conditions at the Field Crop Research Station, Khon Kaen University (Figure 2), were recorded daily over the 2022/2023 and 2023/2024 growing seasons using an on-site automatic weather station (WatchDog 2000, Spectrum Technologies Inc., Lincoln, NE, USA). Average daily temperatures ranged from 22.0 to 31.4 °C and 23.2 to 32.5 °C, respectively. Solar radiation varied from 5.3 to 26.0 MJ m⁻² day⁻¹ in 2022/2023 and from 3.6 to 26.0 MJ m⁻² day⁻¹ in 2023/2024. Total rainfall in the experiment was 1771.8 mm in 2022/2023 and 1760.0 mm in 2023/2024. Despite the cumulative rainfall, a key limiting factor for cassava storage root accumulation was the low rainfall during the dry season (180–240 DAP). During this critical period, total rainfall amounted to 54.1 mm for 14 rainy days in the 2022/2023 growing season and 4.4 mm for 15 rainy days in the 2023/2024 growing season. The historical data from the Thai Meteorological Department also support that the dry period normally occurs during November to January (the drought treatment was imposed) [26]. Insufficient rainfall during the dry season limits water resources for crop production. This limitation can lead to undesirable yields unless suitable genotypes and appropriate water management practices are implemented [13,14].

3.2. Soil Moisture Content and Leaf Relative Water Content (RWC)

The results based on soil moisture content during the experimental period are shown in Figure 3, revealing significantly higher values of soil moisture content at 210 and 240 DAP under the irrigation treatment compared to the drought treatment for both the 2022/2023 and 2023/2024 growing seasons (p ≤ 0.05). The reduction in soil moisture contents under the drought treatment correlated with a smaller value of RWC during 210 and 240 DAP compared to the irrigation treatment (

Table 3. Relative water content (RWC, %) for cassava leaf at 180, 210, 240, 270, and 360 days after planting (DAP) of six cassava genotypes under two water regimes in the 2022/2023 and 2023/2024 growing seasons.

Genotype (G)	RWC at 180 DAP				RWC at 210 DAP				RWC at 240 DAP				RWC at 270 DAP			RWC at 360 DAP
	Irrigation		Drought		Irrigation		Drought		Irrigation		Drought		Irrigation		Drought	Irrigation		Drought
	2022/2023
Kasetsart 50	97.4	A	90.8	B	93.1	BC	92.1	A	93.4	C	89.6	A	96.5	A	97.1	96.3	B	95.6	B
Rayong 9	96.5	A	91.5	AB	91.1	C	85.1	C	83.4	D	80.7	C	96.3	A	94.5	93.9	B	98.4	A
Rayong 72	91.9	B	92.7	AB	97.5	A	92.6	A	94.7	BC	85.5	B	92.4	B	97.4	95.8	B	98.7	A
CMR38–125–77	92.3	B	95.2	A	95.0	AB	93.5	A	96.7	AB	86.4	B	95.8	A	94.1	95.0	B	99.1	A
CMR35–91–63	92.3	B	94.8	A	92.4	BC	88.9	B	96.4	AB	86.8	B	93.3	B	97.1	99.0	A	97.6	A
CM523–7	93.8	B	89.3	B	92.1	BC	87.1	BC	97.1	A	79.8	C	92.7	B	96.2	94.7	B	92.6	C
F-test (G)	**		*		**		**		**		**		**		ns	**		**
Water regime (W)	94.0		92.4		93.5	a	89.9	b	93.6	a	84.8	b	96.1		94.5	95.7		97.0
F-test (W)	ns				*				**				ns			ns
	2023/2024
Kasetsart 50	92.0	B	92.1	B	92.3	B	86.5	BC	90.7	C	89.2		93.7	B	93.4	92.8	B	93.8
Rayong 9	89.6	C	85.1	D	88.7	C	85.0	C	89.6	C	88.4		91.7	C	91.8	87.4	D	92.8
Rayong 72	97.5	A	92.6	AB	95.3	A	88.9	AB	94.1	AB	88.8		95.8	A	91.4	90.3	C	93.7
CMR38–125–77	85.5	D	95.1	A	83.6	D	87.2	A–C	94.7	A	89.4		92.7	BC	93.9	93.2	B	92.9
CMR35–91–63	90.8	BC	88.8	C	88.7	C	86.6	BC	92.2	A–C	87.1		93.7	B	91.6	88.2	D	93.8
CM523–7	92.1	B	90.2	BC	92.4	B	89.3	A	91.2	BC	87.2		92.0	C	92.9	94.7	A	90.5
F-test (G)	**		**		**		*		*		ns		**		ns	**		ns
Water regime (W)	91.3		90.7		90.2	a	87.2	b	92.1	a	88.4	b	93.3		92.5	91.1		92.9
F-test (W)	ns				*				**				ns			ns

Note. Different letters in the same column (uppercase) and the same row (lowercase) represent significant differences (least significant difference test at p ≤ 0.05). ns, *, ** = non-significant, significant at p ≤ 0.05, and significant at p ≤ 0.01, respectively.

). Under drought treatment for the 2022/2023 growing season, Kasetsart 50 showed a significantly higher RWC at both 210 and 240 DAP (92.1% and 89.6%, respectively; p ≤ 0.01), indicating a greater ability to maintain water in its leaves compared to the other genotypes. Whereas Rayong 72 and CMR38–125–77 had a high value of RWC at 210 DAP (92.6% and 93.5%, respectively; p ≤ 0.01). Based on the 2023/2024 growing season, Rayong 72, CMR38–125–77, and CM523–7 performed higher RWC under drought treatment at 210 DAP (88.9%, 87.2% and 89.3%, respectively; p ≤ 0.05). However, non-significant differences were observed among genotypes under drought treatment at 240 DAP. This indicates that all genotypes exhibited similar ability to maintain leaf water under drought conditions.

3.3. Analysis of Variance for Growth Rates

Growth rates were calculated during the treatment period (180–240 DAP) and subsequently during the irrigation period for all experimental plots (240–270 DAP and 270–360 DAP). In the 2022/2023 growing season, analysis of variance indicated that two water regimes significantly (p ≤ 0.01) impacted most growth rate parameters, except for LGR (180–240 DAP), SGR (240–270 DAP), and CGR (270–360 DAP) (

Table 4. The results for analysis of variance for leaf growth rate (LGR), stem growth rate (SGR), storage root growth rate (SRGR), crop growth rate (CGR), and relative growth rate (RGR) for cassava under the 2022/2023 and 2023/2024 growing seasons.

Growth Rate	Duration	Water Regime (W)	Genotype (G)	Water Regime × Genotype (W × G)
	2022/2023
LGR (g m−2 day−1)	180–240 DAP	ns	**	**
	240–270 DAP	**	**	**
	270–360 DAP	**	**	**
SGR (g m−2 day−1)	180–240 DAP	**	**	**
	240–270 DAP	ns	**	**
	270–360 DAP	**	**	**
SRGR (g m−2 day−1)	180–240 DAP	**	**	**
	240–270 DAP	**	**	**
	270–360 DAP	**	**	**
CGR (g m−2 day−1)	180–240 DAP	**	**	**
	240–270 DAP	**	**	**
	270–360 DAP	ns	**	*
RGR (g g−1 day−1)	180–240 DAP	**	**	**
	240–360 DAP	**	**	**
	2023/2024
LGR (g m−2 day−1)	180–240 DAP	**	**	**
	240–270 DAP	*	**	**
	270–360 DAP	ns	**	**
SGR (g m−2 day−1)	180–240 DAP	**	**	**
	240–270 DAP	**	**	**
	270–360 DAP	ns	ns	**
SRGR (g m−2 day−1)	180–240 DAP	**	**	**
	240–270 DAP	**	**	**
	270–360 DAP	**	**	**
CGR (g m−2 day−1)	180–240 DAP	**	**	**
	240–270 DAP	**	**	**
	270–360 DAP	**	**	**
RGR (g g−1 day−1)	180–240 DAP	ns	**	**
	240–360 DAP	ns	**	*

Note. DAP = days after planting; ns, *, ** = non-significant, significant at p ≤ 0.05, and significant at p ≤ 0.01, respectively.

). The significant differences for genotypes (p ≤ 0.01) and the interaction between water regimes and genotypes (p ≤ 0.01 and p ≤ 0.05) were also observed. For the 2023/2024 growing season, non-significant differences between the two water regimes were observed for LGR and SGR (270–360 DAP) and RGR (180–240 DAP and 240–360 DAP). Genotypic differences (p ≤ 0.01) were observed for nearly all growth rate parameters, with the exception of SGR (270–360 DAP). Furthermore, the water regime and genotype interaction significantly (p ≤ 0.01 and p ≤ 0.05) affected all growth rate parameters.

3.4. Cassava Performances

The total dry weight, storage root dry weight, HI, and starch yield for six cassava genotypes at final harvest (360 DAP) for the two watering regimes are presented in

Table 5. Total dry weight, storage root dry weight, starch yield, and harvest index at 360 days after planting (DAP) of six cassava genotypes under two water regimes in the 2022/2023 and 2023/2024 growing seasons.

Genotype (G)	Total Dry Weight (t ha−1)						Storage Root Dry Weight (t ha−1)						Starch Yield (t ha−1)						Harvest Index
	Irrigation		Drought		Reduction (%)		Irrigation		Drought		Reduction (%)		Irrigation		Drought		Reduction (%)		Irrigation		Drought
	2022/2023
Kasetsart 50	29.0	B	35.1	A	−25.3	D	16.1	BC	26.7	A	−55.3	F	5.45	AB	6.00	A	−67.5	E	0.56	B	0.76	A
Rayong 9	31.6	A	37.4	A	−15.4	C	17.5	B	22.8	B	−35.0	D	6.00	A	5.66	B	−26.1	C	0.55	B	0.62	B
Rayong 72	23.3	C	30.0	B	−36.4	E	15.2	BC	22.1	B	−47.2	E	4.92	B	4.39	C	−25.9	C	0.65	A	0.74	A
CMR38–125–77	30.9	AB	24.5	C	23.7	A	21.5	A	16.4	C	15.5	B	5.28	AB	3.44	D	38.2	B	0.69	A	0.67	AB
CMR35–91–63	30.9	AB	25.6	BC	14.9	B	14.9	C	4.2	D	66.3	A	5.39	AB	1.23	F	66.1	A	0.48	C	0.16	C
CM523–7	19.1	D	26.2	BC	−43.3	F	11.7	D	15.1	C	−25.8	C	2.94	C	2.49	E	−42.1	D	0.58	B	0.58	B
F-test (G)	**		**		**		**		**		**		**		**		**		**		**
Water regime (W)	27.5	b	29.8	a	−13.6		16.2	b	17.9	a	−13.6		3.74	b	3.87	a	−9.5		0.59		0.59
F-test (W)	**						**						*						ns
	2023/2024
Kasetsart 50	34.9	B	25.1	E	27.9	B	19.2	C	11.7	C	50.2	A	4.08	C	2.09	C	57.7	A	0.55	BC	0.36	E
Rayong 9	47.1	A	40.8	A	15.8	C	30.1	A	24.9	A	26.4	C	6.85	A	5.95	A	9.5	C	0.64	A	0.58	B
Rayong 72	34.8	B	29.8	CD	9.2	D	6.5	E	19.2	B	−73.6	F	1.36	E	4.00	B	−73.6	F	0.60	AB	0.53	C
CMR38–125–77	32.4	B	35.6	B	−12.9	E	21.2	C	25.6	A	−44.0	E	5.38	B	6.66	A	−50.7	E	0.66	A	0.66	A
CMR35–91–63	24.3	C	27.4	DE	−28.4	F	11.2	D	16.7	B	−39.6	D	2.16	DE	3.31	B	−41.5	D	0.50	C	0.48	D
CM523–7	49.7	A	30.8	C	36.3	A	25.8	B	16.7	B	35.3	B	2.58	D	1.67	C	36.8	B	0.52	BC	0.54	BC
F-test (G)	**		**		**		**		**		**		**		**		**		**		**
Water regime (W)	37.2	a	31.6	b	8.0		19.1		19.0		−7.6		3.95		3.73		−10.3		0.58	a	0.53	b
F-test (W)	**						ns						ns						*

Note. Reduction (%) = (Dry weight for irrigation − Dry weight for drought) × 100/Dry weight for irrigation; Different letters in the same column (uppercase) and the same row (lowercase) represent significant differences (least significant difference test at p ≤ 0.05). ns, *, ** = non-significant, significant at p ≤ 0.05, and significant at p ≤ 0.01, respectively.

. The data clearly show significant differences (p ≤ 0.01) in these four crop traits among the various genotypes within each water regime for both growing seasons. In the 2022/2023 growing season, CMR38–125–77 stood out under full irrigation, performing strongly across all four metrics: total dry weight (30.9 t ha⁻¹), storage root dry weight (21.5 t ha⁻¹), HI (0.69), and starch yield (5.28 t ha⁻¹; p ≤ 0.01). This genotype also showed a higher LGR between 180 and 240 DAP and RGR between 240 and 360 DAP (

Table A1. Means for leaf growth rate (LGR) and stem growth rate (SGR) during 180–240, 240–270, and 270–360 days after planting (DAP) of six cassava genotypes under two water regimes in the 2022/2023 and 2023/2024 growing seasons.

Genotypes (G)	LGR (g m–2 day–1)												SGR (g m–2 day–1)
	180–240				240–270				270–360				180–240				240–270				270–360
	Irrigation		Drought		Irrigation		Drought		Irrigation		Drought		Irrigation		Drought		Irrigation		Drought		Irrigation		Drought
	2022/2023
Kasetsart 50	−1.6	B	−2.0	C	0.9	B	1.5	A	3.2	B	0.6	B	3.2	D	9.5	C	26.8	D	32.7	C	0.8	D	6.7	C
Rayong 9	−0.8	A	0.7	A	0.2	C	1.4	A	0.8	D	1.9	A	4.6	C	3.1	E	45.6	B	28.7	D	0.8	D	14.5	A
Rayong 72	−0.8	A	−0.9	B	−1.5	D	−1.9	C	2.0	C	0.6	B	1.5	E	5.4	D	6.4	F	27.6	E	1.7	D	2.6	D
CMR38–125–77	−0.9	A	−1.1	B	1.3	B	−1.1	B	1.7	C	1.5	A	8.6	B	10.7	B	33.9	C	47.1	A	6.2	B	2.3	D
CMR35–91–63	−1.8	BC	−1.3	B	2.9	A	−5.6	D	0.8	D	2.0	A	2.3	E	14.7	A	54.7	A	14.0	F	3.1	C	3.0	D
CM523–7	−2.2	C	−3.7	D	−2.5	E	−1.3	B	6.3	A	1.9	A	9.6	A	9.3	C	18.7	E	40.1	B	11.1	A	8.5	B
F-test (G)	**		**		**		**		**		**		**		**		**		**		**		**
Water regime (W)	−1.3		−1.4		0.2	a	−1.2	b	2.5	a	1.4	b	5.0	b	8.8	a	31.0		31.7		3.9	b	6.3	a
F-test (W)	ns				**				**				**				ns				**
	2023/2024
Kasetsart 50	8.3	B	3.8	B	0.9	E	4.2	C	1.4	C	2.5	B	14.3	B	12.0	B	43.8	B	26.6	D	10.7	B	4.9	C
Rayong 9	0.5	F	5.2	A	4.1	C	5.3	B	2.2	B	1.8	BC	10.4	C	10.7	C	22.7	C	32.1	BC	13.0	A	5.0	C
Rayong 72	1.4	E	1.2	C	11.5	A	5.5	B	1.9	BC	0.9	D	10.1	C	4.7	D	7.7	F	12.9	E	3.6	C	3.7	D
CMR38–125–77	2.6	D	1.5	C	2.3	D	6.3	A	1.6	BC	1.7	C	40.1	A	11.3	BC	20.1	D	29.0	CD	4.1	C	7.9	B
CMR35–91–63	4.3	C	5.8	A	2.3	D	0.9	D	1.6	BC	2.1	BC	15.4	B	2.6	E	11.2	E	34.4	B	12.9	A	3.1	D
CM523–7	9.1	A	2.8	B	8.6	B	4.1	C	5.2	A	4.6	A	15.3	B	18.6	A	64.1	A	76.1	A	1.1	D	12.5	A
F-test (G)	**		**		**		**		**		**		**		**		**		**		**		**
Water regime (W)	4.4	a	3.4	b	5.0	a	4.4	b	2.3		2.3		17.6	a	10.0	b	28.3	b	35.2	a	7.6		6.2
F-test (W)	**				*				ns				**				**				ns

Note. Different letters in the same column (uppercase) and the same row (lowercase) represent significant differences (least significant difference test at p ≤ 0.05). ns, *, ** = non-significant, significant at p ≤ 0.05 and significant at p ≤ 0.01, respectively.

and

Table A3. Means for relative growth rate (RGR, g g^–1 day^–1) during 180–240 and 240–360 days after planting (DAP) of six cassava genotypes under two water regimes in the 2022/2023 and 2023/2024 growing seasons.

Genotypes (G)	RGR for 180–240 DAP				RGR for 240–360 DAP
	Irrigation		Drought		Irrigation		Drought
	2022/2023
Kasetsart 50	0.0062	A	0.0083	AB	0.0026	B	0.0104	A
Rayong 9	0.0046	B	0.0068	C	0.0017	B	0.0036	C
Rayong 72	0.0064	A	0.0075	BC	0.0070	A	0.0047	B
CMR38–125–77	0.0046	B	0.0091	A	0.0062	A	0.0016	E
CMR35–91–63	0.0062	A	0.0073	C	0.0071	A	0.0001	F
CM523–7	0.0041	B	0.0043	D	0.0069	A	0.0027	D
F-test (G)	**		**		**		**
Water regime (W)	0.0053	b	0.0072	a	0.0053	a	0.0039	b
F-test (W)	**				**
	2023/2024
Kasetsart 50	0.0050	C	0.0042	C	0.0034	C	0.0017	C
Rayong 9	0.0080	B	0.0071	AB	0.0022	DE	0.0030	BC
Rayong 72	0.0019	D	0.0042	C	0.0015	E	0.0018	C
CMR38–125–77	0.0054	C	0.0057	BC	0.0028	CD	0.0047	AB
CMR35–91–63	0.0055	C	0.0055	C	0.0044	B	0.0030	BC
CM523–7	0.0125	A	0.0082	A	0.0068	A	0.0061	A
F-test (G)	**		**		**		**
Water regime (W)	0.0064		0.0058		0.0035		0.0034
F-test (W)	ns				ns

Note. Different letters in the same column (uppercase) and the same row (lowercase) represent significant differences (least significant difference test at p ≤ 0.05). ns, ** = non-significant, and significant at p ≤ 0.01, respectively.

). When faced with late drought, Kasetsart 50 proved to be the top performer for all four crop characteristics: total dry weight (35.1 t ha⁻¹), storage root dry weight (26.7 t ha⁻¹), HI (0.76), and starch yield (6.00 t ha⁻¹; p ≤ 0.01) (Table 5), and this was linked to high values of LGR and CGR from 240 to 270 DAP and RGR from 180 to 240 DAP and 240 to 360 DAP (Table A1,

Table A2. Means for storage root growth rate (SRGR) and crop growth rate (CGR) during 180–240, 240–270, and 270–360 days after planting (DAP) of six cassava genotypes under two water regimes in the 2022/2023 and 2023/2024 growing seasons.

Genotypes (G)	SRGR (g m–2 day–1)												CGR (g m–2 day–1)
	180–240				240–270				270–360				180–240				240–270				270–360
	Irrigation		Drought		Irrigation		Drought		Irrigation		Drought		Irrigation		Drought		Irrigation		Drought		Irrigation		Drought
	2022/2023
Kasetsart 50	26.5	D	27.4	B	39.0	C	52.9	B	24.1	B	6.7	F	42.4	B	37.4	B	41.7	C	76.7	A	21.5	C	19.0	C
Rayong 9	37.6	A	14.4	C	42.5	B	25.9	F	23.1	C	13.4	C	38.4	C	39.9	A	59.2	A	34.5	F	18.0	CD	10.8	E
Rayong 72	33.9	B	28.4	A	44.7	A	47.4	C	16.4	E	12.5	D	48.6	A	33.9	C	27.5	D	56.7	D	35.2	A	30.9	A
CMR38–125–77	22.6	E	7.7	D	36.3	D	43.0	D	18.9	D	23.6	B	29.6	D	17.1	E	44.3	C	62.8	C	13.7	D	14.8	D
CMR35–91–63	28.5	C	6.8	E	5.4	F	28.3	E	32.9	A	11.2	E	35.2	C	30.4	D	54.7	B	43.7	E	29.3	B	29.5	AB
CM523–7	16.5	F	4.7	F	21.8	E	65.6	A	11.0	F	25.9	A	12.3	E	15.1	E	19.8	E	69.2	B	22.1	C	25.7	B
F-test (G)	**		**		**		**		**		**		**		**		**		**		**		**
Water regime (W)	27.6	a	14.9	b	31.6	b	43.9	a	21.2	a	15.5	b	34.4	a	29.0	b	41.2	b	57.3	a	23.3		21.8
F-test (W)	**				**				**				**				**				ns
	2023/2024
Kasetsart 50	14.0	C	24.0	A	37.7	D	33.6	B	51.8	A	7.3	D	27.0	C	36.8	A	31.6	C	29.0	C	44.5	A	25.7	C
Rayong 9	18.1	B	15.4	C	49.5	C	30.8	C	14.3	D	14.9	BC	24.2	D	29.1	B	49.8	B	25.5	D	15.5	D	24.6	C
Rayong 72	4.7	E	14.8	C	70.5	A	10.8	DE	17.5	C	8.5	CD	5.4	F	3.4	E	32.7	C	46.4	A	11.1	E	30.0	BC
CMR38–125–77	3.9	E	16.7	BC	54.4	B	12.1	D	18.9	C	19.8	B	38.7	B	13.3	D	21.5	D	35.5	B	24.1	C	28.6	C
CMR35–91–63	10.4	D	11.7	D	55.0	B	9.4	E	29.4	B	33.5	A	15.4	E	18.8	C	55.8	A	43.2	A	37.0	B	36.0	AB
CM523–7	32.8	A	18.4	B	23.3	E	40.0	A	12.7	D	29.6	A	49.9	A	12.1	D	49.9	B	30.4	C	26.7	C	40.4	A
F-test (G)	**		**		**		**		**		**		**		**		**		**		**		**
Water regime (W)	14.0	b	16.8	a	48.4	a	22.8	b	24.1	a	18.9	b	26.8	a	18.9	b	40.2	a	35.0	b	26.5	b	30.9	a
F-test (W)	**				**				**				**				**				**

Note. Different letters in the same column (uppercase) and the same row (lowercase) represent significant differences (least significant difference test at p ≤ 0.05). ns, ** = non-significant, and significant at p ≤ 0.01, respectively.

and Table A3). Significant differences between the two water regimes were also recorded for total dry weight (p ≤ 0.01), storage root dry weight (p ≤ 0.01), and starch yield (p ≤ 0.05) for the 2022/2023 growing season. Under drought treatment, total dry weight (29.8 t ha⁻¹), storage root dry weight (17.9 t ha⁻¹), and starch yield (3.87 t ha⁻¹) were higher than those under full irrigation.

According to the 2023/2024 growing season (Table 5), however, the picture changed slightly. Rayong 9 emerged as a strong performer, showing excellent total dry weight, storage root dry weight, and starch yield under both full irrigation and drought (p ≤ 0.01). This genotype showed high LGR from 180 to 240 and SGR from 270 to 360 for drought and irrigation, respectively (Table A1 and Table A2). Under drought, Rayong 9 also exhibited a great value for RGR from 180 to 240 DAP (Table A3). Interestingly, CMR38–125–77 also maintained high storage root dry weight (25.6 t ha⁻¹), HI (0.66), and starch yield (6.66 t ha⁻¹) even under drought treatment in this later season (p ≤ 0.01) (Table 5) with high LGR from 240 to 270 and RGR from 240 to 360 DAP (Table A1 and Table A3). As expected, the results confirm higher total crop dry weight (37.2 t ha⁻¹; p ≤ 0.01) and HI (0.58; p ≤ 0.05) under full irrigation compared to the drought treatment during the late growth phase. This suggests that biomass accumulation was negatively impacted by drought in this season, and the proportion of total dry matter allocated to storage roots was reduced with drought stress.

The percentage reduction values (Table 5), which represent the decrease in dry weight under drought conditions relative to the dry weight under full irrigation, revealed that drought during the late growth period reduced both dry weight and starch yield for certain cassava genotypes during the 2022/2023 growing season. However, the same results were not observed in the 2023/2024 growing season, indicating year-to-year variability in drought impact. However, the storage root dry weight and starch yield of Rayong 72 under drought were higher than those under full irrigation.

3.5. Relationship Between Crop Characteristics

The experimental results revealed correlations between final yield data and plant growth rates (n = 48). Considering the significant correlation with moderate to strong levels, it was found that under full irrigation (Figure 4), total dry weight exhibited a significant positive correlation with LGR during 180–240 DAP (r = 0.52, p ≤ 0.05), LGR during 240–270 DAP (r = 0.65, p ≤ 0.01). In terms of storage root dry weight, a positive correlation was observed with RGR during 180–240 DAP (r = 0.62, p ≤ 0.01). Under drought conditions imposed during the late growth phase (Figure 5), storage root dry weight was positively correlated with LGR during 240–270 DAP (r = 0.53, p ≤ 0.01) and RGR during 240–360 DAP (r = 0.62, p ≤ 0.01). However, a negative correlation was observed between starch yield and LGR during 270–360 DAP (r = −0.50, p ≤ 0.01). Notably, HI under drought conditions exhibited a positive correlation with RGR during 240–360 DAP (r = 0.57, p ≤ 0.01).

4. Discussion

This study demonstrated soil moisture, growth rates, and final yield of six different cassava genotypes grown under drought during the late growth phase and under full irrigation. A direct consequence of low soil moisture contents under drought treatment was observed in low RWC in cassava leaves. As RWC serves as a robust indicator of cellular hydration level, low values signify a water deficit, resulting in outcomes such as reduced photosynthesis, stunted growth, and decreased yield [22]. The different water regimes resulted in differences in the growth of six cassava genotypes across two experimental years (2022/2023 and 2023/2024). During the treatment period (180–240 DAP) and subsequent irrigation periods (240–270 DAP and 270–360 DAP), the influence of water availability was evident in shaping growth dynamics. In the 2022/2023 growing season, most growth rate parameters were significantly affected by water availability (Table 4), indicating high sensitivity of cassava to water during both the treatment and post-treatment stages. This finding is consistent with prior studies reporting genotype-dependent resilience in shoot and canopy traits under different water regimes in cassava [8,27]. However, exceptions were observed for some growth rate parameters, such as LGR (180–240 DAP), SGR (240–270 DAP), and CGR (270–360 DAP). Moreover, significant genotype effects and genotype × water regime interactions indicate genetic variation and the different responses of genotypes to water availability, respectively. This supports breeding programs to exploit such interactions in developing drought-resilient genotypes [28,29]. In contrast, the 2023/2024 season revealed a slightly different pattern. The water regime did not significantly affect SGR during the final phase (270–360 DAP) and RGR during both the treatment and overall periods (180–240 DAP and 240–360 DAP). The difference between 2022/2023 and 2023/2024 growing seasons might be due to a seasonal variation in environmental conditions (e.g., rainfall distribution, temperature, solar radiation, and soil properties) and an adaptive adjustment by plants subjected to prior stress exposure [27,30]. However, genotypic variation remained a dominant factor influencing growth rate parameters across nearly all measurement phases, except for SGR in the final growth stage. Additionally, the water regime × genotype interaction was significant for all parameters, highlighting the complexity of genotype-environment interplay in cassava. The 2022/2023 growing season demonstrates that the drought treatment during the late growth phase provided slightly higher total dry weight, storage root dry weight, and starch yield than the full irrigation treatment (Table 5). This was due to a positive performance of certain genotypes, even growing under the drought treatment, as supported by the percentage of reduction. A small amount of rainfall during the drought treatment (180–240 DAP) (Figure 2) for the 2022/2023 growing season could explain the observed result. During the 2023/2024 growing season, however, full irrigation treatment continued to show a positive effect (Table 5), resulting in significantly higher total dry weight and HI compared to the drought treatment. The inconsistencies observed between the two growing seasons were primarily due to a combination of varying environmental factors (as mentioned previously) and the distinct seasonal responses of the genotypes. The results for the 2023/2024 growing season align with previous studies that emphasize the critical role of adequate water supply in promoting biomass accumulation in cassava [27,31]. The improved performance under irrigation likely reflects optimal physiological processes, including sustained photosynthesis and carbohydrate translocation to storage organs, which are often impaired under water deficit conditions [27,30]. The increase in HI under full irrigation in the 2023/2024 growing season indicates more efficient partitioning of assimilates to storage roots rather than to vegetative tissues. Together, these results highlight the strong dependency of cassava performance on water availability, with inconsistent observations across both years under full irrigation. The observed variation in HI across years may also reflect genotype-by-environment interactions and seasonal differences in rainfall distribution or temperature [11,18]. This study collectively demonstrates the correlation between final yield and growth rate parameters (Figure 4 and Figure 5). Under full irrigation, the positive correlations of total dry weight with LGR during 180–240 DAP and 240–270 DAP suggest that sustained canopy development during the late growth phase enhances assimilate production and partitioning to storage organs. Similar findings were reported by [6,32], who emphasized that vigorous canopy growth contributes to higher photosynthetic capacity and dry matter accumulation in cassava under non-stress conditions. A higher RGR reflects the plant’s ability to efficiently increase dry matter per unit of existing biomass over time, signifying strong physiological vigor and effective assimilation [33,34]. The positive correlation between storage root dry weight and RGR during 180–240 DAP under full irrigation underscores the importance of rapid biomass accumulation during the early growth phase in determining final yield potential. In cassava, early increases in RGR indicate vigorous photosynthetic activity and early establishment of source–sink relationships, which promote subsequent storage root expansion [2,27].

Under drought treatment, the positive correlations of storage root dry weight with LGR during 240–270 DAP and RGR during 240–360 DAP indicate that genotypes capable of maintaining or recovering leaf area and crop growth after stress onset tend to produce higher yields. The positive association between HI and RGR during 240–360 DAP under drought also suggests that sustaining or recovering growth in the late developmental phase can more effectively allocate assimilates to storage roots [35]. Conversely, the negative correlation between starch yield and LGR during 270–360 DAP under drought indicates that excessive vegetative regrowth after stress may compete with storage root filling for assimilates. This trade-off between vegetative recovery and storage allocation under post-stress conditions has been documented in cassava and other root crops [36].

Based on the evaluation of the suitable cassava genotypes under full irrigation and drought conditions in the 2022/2023 growing season (Table 5), CMR38–125–77 exhibited superior performance at final harvest under full irrigation, indicating its suitability for well-watered environments aimed at maximizing yield. In contrast, under late drought conditions, Kasetsart 50 outperformed all other genotypes at final harvest, demonstrating its drought tolerance and ability to maintain productivity under water-limited conditions. The 2023/2024 growing season, however, presented a slightly different picture. CMR38–125–77 maintained high storage root dry weight, HI, and starch yield under drought treatment. This suggests that CMR38–125–77 may possess some level of drought resilience, although its performance might be more pronounced under optimal water conditions. Kasetsart 50 (2022/2023 growing season) and CMR38–125–77 (2023/2024 growing season) proved to be the top performers when faced with late drought stress in terms of storage root dry weight and starch yield. These genotypes performed well in LGR (240–270 DAP) and RGR (240–360 DAP) (Table A1 and Table A3), indicating vigorous leaf growth and a good recovery after the drought period [11,37]. Therefore, the genotypes Kasetsart 50 and CMR38–125–77 are being proposed as drought-tolerant checks. A study by Wongnoi et al. [38] identified CMR38–125–77 and Kasetsart 50 as top-performing cassava genotypes for growing under drought during the late growth phase. These genotypes demonstrated superior water use efficiency and crop biomass production when subjected to prolonged dry periods during the critical storage root accumulation phase. The consistent high performance of CMR38–125–77 across diverse growing conditions has also been documented in previous research [11,12]. This suggests its potential as a valuable genotype for improving cassava cultivation in drought-prone environments. In the 2023/2024 growing season, Rayong 9 showed excellent performance in total dry weight, storage root dry weight, and starch yield under both full irrigation and drought. This makes Rayong 9 a promising candidate for cultivation in regions with unpredictable rainfall patterns, offering consistent yields regardless of water availability. Good performances based on LGR and RGR (180–240 DAP) for Rayong 9 during the drought stress (2023/2024 growing season) suggest that this genotype was able to maintain vigorous leaf growth and overall crop growth, and efficiently allocate resources even under water scarcity, which is crucial for drought tolerance [31]. Moreover, an understanding of Rayong 9’s performance under full irrigation could guide cultural practices and support varietal recommendations under full irrigation.

In addition, crop yield reduction under drought conditions is one of the important criteria for identifying drought-tolerant genotypes. Genotypes that show a smaller decrease in yield when subjected to water scarcity are considered more resilient and valuable for breeding programs aimed at improving crop performance in drought-prone environments. This study demonstrated that the natural drought experienced during the late growth phase did not consistently pose a severe problem for certain cassava genotypes across specific years, as evidenced by the varying percentage reductions in dry weight and starch yield. This suggests that the performance of individual cassava under different water regimes exhibited inter-annual variability, raising the need for multi-location or controlled-environment trials to validate genotype responses under standardized stress conditions [20,39,40]. Remarkably, drought had no adverse effect on the storage root dry weight and starch yield of Rayong 72 for both the 2022/2023 and 2023/2024 growing seasons (Table 5). Despite yield reductions of –47.2% (storage root dry weight) and –25.9% (starch yield) in 2022/2023, and –73.6% for both traits in 2023/2024, Rayong 72 consistently exhibited a superior capacity to maintain yield under drought conditions. Although Rayong 72 did not outperform other genotypes in terms of total dry weight, storage root dry weight, or starch yield, its drought tolerance indicates that it is a valuable genetic resource for breeding programs. Nevertheless, further studies across multiple locations and years are needed to validate the genotype’s performance.

5. Conclusions

The study assessed soil moisture, growth rates, and final yields of six cassava genotypes under different water management. Across the 2022/2023 and 2023/2024 growing seasons, this study revealed that low soil moisture contents for drought treatment affect genotypic variation in cassava CMR38–125–77 performed well under full irrigation in 2022/2023 and also demonstrated drought resilience in 2023/2024 by maintaining high storage root dry weight (25.6 t ha⁻¹), HI (0.66), and starch yield (6.66 t ha⁻¹). Kasetsart 50 proved to be a top performer under late drought conditions in 2022/2023, with 26.7 t ha⁻¹ for storage root dry weight, 0.76 for HI, and 6.00 t ha⁻¹ for starch yield, indicating its potential as a drought-tolerant genotype. Under drought stress, the genotypes Kasetsart 50 (2022/2023) and CMR38–125–77 (2023/2024) demonstrated that better LGR and RGR during the rewatering phase were associated with higher storage root dry weight and starch yield. Rayong 9 showed excellent adaptability in 2023/2024, performing well under both full irrigation and drought in total dry weight (47.1 and 40.8 t ha⁻¹, respectively), storage root dry weight (30.1 and 24.9 t ha^–1, respectively), and starch yield (6.85 and 5.95 t ha⁻¹, respectively), making it suitable for erratic rainfall regions. High LGR and RGR during drought stress were linked to good performance at final harvest for Rayong 9 (2023/2024). In addition, Rayong 72 displayed remarkable drought tolerance across both seasons, with no adverse effects on its storage root dry weight and starch yield, indicating its value as genetic material for breeding programs. The differences in soil and weather conditions between the 2022/2023 and 2023/2024 growing seasons led to variations in the genotypes’ performance; therefore, validation through experiments across multiple locations and years remains necessary. Overall, this study indicates the necessity of selecting genotypes with robust performance under both stress and recovery conditions to support climate-resilient cassava production. Furthermore, integrating physiological traits such as photosynthetic rate, leaf area index, and specific leaf area into selection criteria could enhance understanding of drought tolerance mechanisms and improve the efficiency of breeding programs.