Impact of different planting date on shoot growth and tuber development of white Guinea yam grown in Guinea Savanna

ABSTRACT

White Guinea yam is a tuber crop characterized by a long growth period over the entire rainy season, which is affected by changes in rainfall patterns owing to climate change. In this study, we investigated the effects of shifting the planting date of yam seed tubers on shoot growth and tuber development, which could be a plausible measure for adapting to changes in rainfall patterns. Our results revealed that tuber enlargement in plants with different planting dates started depending on the days after planting rather than the day of the year. This indicates that day length had little effect on tuberization. The final tuber yield was rarely affected by low rainfall during the early growth stage, whereas severe yield reduction occurred when the rainfall stopped during the tuberization period. We conclude that shifting the planting date backward to adjust for delayed rainy seasons increases the risk of severe yield reduction.

GRAPHICAL ABSTRACT

KEYWORDS:

• Yam
• Dioscorea rotundata
• shoot growth
• tuber development
• planting date

1. Introduction

White Guinea yam (Dioscorea rotundata) is a staple tuber crop extensively cultivated across the Guinea savanna in West Africa. Yam constitutes 10–20% of the region’s food supply in terms of energy (FAOSTAT, 2023), yet its cultivation is increasingly affected by climate change. Because the cultivation of yams generally spends eight – ten months from planting to harvest, the start and end times of the rainy season are critical factors for shoot growth and tuber development. A previous study concluded that the yam yield in the Upper Ouémé Basin in the Republic of Bénin will decrease by 33% by 2050, owing to an extreme decline in rainfall (Srivastava et al., 2012). Another estimate indicates that the duration of the rainy season in the Guinea savanna will be shortened by 37 days by 2050 (Kwawuvi et al., 2022). Although the effects of climate change on rainfall patterns remain unclear, changes in rainfall patterns are a serious problem for yam production in West Africa.

Shifting the planting date of yam seed tubers is a plausible countermeasure to changes in rainfall patterns (Carr et al., 2022). Because a high tuber yield is ensured by vigorous early vegetative growth (Iseki & Matsumoto, 2020), setting the planting date immediately after the start of the rainy season is important for yam cultivation. Therefore, soil water scarcity owing to changes in rainfall can be avoided by shifting the planting date. However, its effect on tuber yield remains unclear due to the photoperiod sensitivity of tuber enlargement induced by short-day conditions (Hamaoka et al., 2022). If the planting date is delayed due to a delay in the rainy season, tuber enlargement begins before the establishment of shoot biomass, possibly causing a decrease in tuber yield. Therefore, the effect of a shift in planting date on tuber yield must be evaluated from the perspectives of drought avoidance and photoperiod sensitivity of tuber enlargement.

The effect of planting date on tuber enlargement has been extensively studied in water yam (Dioscorea alata). In a field study under non-limiting soil water conditions, early planting caused an early onset of tuber enlargement but a slow tuber growth rate, whereas late planting caused a late onset of tuber enlargement but a fast tuber growth rate (Marcos et al., 2011). This may be explained by the different effects of short day lengths at different growth stages (Shiwachi et al., 2000). Although the effect of photoperiod differs between yam species and varieties (Shiwachi et al., 2002), there have been few studies on white Guinea yams. In addition, previous water yam studies were conducted in high-latitude regions with sufficient rainfall. The effect of planting date on tuber growth and yield of white Guinea yam will be different in equatorial regions, such as the Guinea savanna, where the variation in day length is relatively small and rainfall is more unstable.

The objective of the present study was to clarify the effects of planting date on shoot growth, tuber enlargement, and tuber yield in white Guinea yams grown in the Guinea savanna region of southern Nigeria. The yams were grown on three planting dates for two years with different rainfall patterns. The effectiveness of the shift in planting date on future changes in rainfall patterns is discussed.

2. Materials and methods

2.1. Plant materials

Two accessions of white guinea yam, TDr 95/01932 and TDr2948, were selected from genetic resources, including a mini-core collection (Pachakkil et al., 2020) depending on their maturity date. Both accessions were male and had middle maturity; these characteristics constituted the majority of the genetic resources in our previous evaluation (Darkwa et al., 2020).

2.2. Planting dates and growth conditions

Field experiments were conducted for two cultivate seasons of 2019‒2020 and 2020‒2021 at the experimental fields of the International Institute of Tropical Agriculture (IITA), Ibadan, Nigeria (7° 29 N, 3° 54 E). The soil was a sandy loam with moderate acidity (pH 5.8–6.1). The soil organic carbon, total nitrogen, and Bray-1 phosphate contents were 4.3, 0.39, and 3.1 mg kg⁻¹, respectively. Prior to transplantation, the soil was plowed to ensure uniform field conditions.

Three planting dates were considered: early, middle, and late. Early planting was performed on 19 April 2019, corresponding to 109 days of the year (DOY), and 11 April 2020 (DOY102). Middle planting was performed on 18 May 2019 (DOY138) and 20 May 2020 (DOY141). Late planting was performed on 29 June 2019 (DOY180) and 18 June 2020 (DOY170). The middle planting was the conventional practice of yam cultivation at the experimental site.

To avoid the effects of different seed tuber sizes on early shoot growth (Iseki & Matsumoto, 2020), 100 g of tuber blocks (setts) with a skin surface where a shoot bud could emerge were cut from the center of a normal-sized tuber weighing approximately 1–2 kg. The setts were treated with a fungicide and planted in plastic pots (12 cm in diameter and 10 cm in height) filled with sterilized topsoil (sandy loam soil of pH 7.6 containing 2.0 g kg⁻¹ organic carbon, 0.40 g kg⁻¹ nitrogen, and 3.8 mg kg⁻¹ Bray-1 phosphate) for pre-sprouting.

After one month, plants with adequate sprouts were selected for each accession and transplanted with stakes at the top of 40-cm-high ridges prepared in the field. The distance between the ridges and between the plants on the ridges was 1 m, resulting in a plant density of 1 plant m⁻². Each plot measured 3 m × 3 m and consisted of 16 plants. For each accession, 27 plots were constructed, consisting of 12 plots for early planting, nine plots for middle planting, and six plots for late planting. The total number of 54 plots were arranged in a randomized block design. Weeding was performed manually when required and fertilizers were not used in this study.

2.3. Meteorological conditions

Meteorological data were obtained from a weather station at the study site during the experimental season. The average rainfall pattern was bimodal with a short dry spell from early to mid-August. Annual rainfall at the experimental site was higher in 2019 (1904 mm) and lower in 2020 (1074 mm) than the historical average of 1290 mm (Figure 1a). In 2020, rainfall before September (DOY250) was less than half of that in 2019 and completely ceased after November (DOY304). The average maximum and minimum temperatures remained relatively consistent over the years, at 31.7°C and 22.8°C in 2019, and 32.0°C and 22.5°C in 2020, respectively. Day length at the experimental site was calculated using R version 4.2.1 with the ‘chillR’ package. The longest and shortest day lengths were recorded as 12.6 hours and 11.7 hours on June 22 and December 22, respectively (Figure 1b).

Figure 1. Changes in rainfall, air temperatures, and day length during the experimental periods of day of year from 50 to 400. (a) air temperatures and rainfall. The solid and dashed lines represent maximum and minimum daily air temperatures. Bars represents daily rainfall. The triangles and vertical lines of blue, red, and gray colors indicate planting dates of the early, middle, and late planting, respectively. The days after planting was also indicated by the lines. (b) day length.

2.4. Measurements

Time-course changes in shoot dry weight were evaluated every two weeks, from one month after planting to the end of full shoot senescence. A nondestructive method using a handheld sensor was applied for the normalized difference vegetation index (NDVI) (GreenSeeker, Nikon Trimble, Tokyo, Japan). The NDVI was measured with simultaneous measurements of plant height also being recorded. Shoot dry weight (g plant⁻¹) was estimated using an equation that included NDVI and plant height as explanatory variables, according to Iseki and Matsumoto (2019). The same four plants in the plots used to evaluate final tuber yield were continuously targeted. Four plants were selected from inside the plot to eliminate border plants. Twelve plants, consisting of four plants and three replicate plots, were evaluated for each combination with different planting dates and accessions.

Tuber sampling was conducted four, three, and two times for early, middle, and late plantings, respectively. Sampling was first conducted three months after planting and was subsequently performed every 40 days until the onset of shoot senescence. At full shoot senescence around DOY380, tubers were harvested from each plant. Sampling and harvesting were conducted on four plants from each of the three replicate plots, which were the same plants used for the NDVI measurement. After tuber sampling, the total fresh tuber weight of each plant was measured using an electrical scale (SH-6000WP; A&D Systems, Tokyo, Japan). A subset of tubers from each plant was used to measure dry weight and calculate tuber moisture content. The dry weight was determined by oven-drying the tuber subsets at 80°C for 72 h. As the indicator of tuber maturity, tuber moisture content (%) was determined as follows: (tuber fresh weight – tuber dry weight)/tuber fresh weight × 100.

2.5. Statistical analysis

Shoot dry weight, tuber dry weight, and tuber moisture content were averaged for four plants in each replicate. The effects of cultivation year, accession, planting date, and days after planting (DAP) on the traits were evaluated using a multiway analysis of variance (ANOVA). The DAP factor was defined by categorizing the sampling/evaluating dates into three growth phases: before 120 DAP, from 120 DAP to 200 DAP, and after 200 DAP, corresponding to the periods of early vegetative growth, maximum vegetation, and senescence for the middle planting plants, respectively. The number of levels in each factor was two years, two accessions, three planting dates, and four DAP phases. The total numbers of data points for shoot dry weight using the NDVI evaluation, tuber dry weight, and tuber moisture content were 910 and 403, respectively. The F-values and percentage of contribution (ρ) were calculated from type-II sum of squares and mean squares, and then compared among the factors. ANOVA was conducted using the statistical software R version 4.2.1.

3. Results

The ρ value in ANOVA table shows contribution of the factors on the total variance of each trait. The contribution of DAP was the largest for shoot dry weight (Table 1), tuber dry weight (Table 2), and tuber moisture content (Table 3). These results indicate that shoot biomass, tuber yield, and tuber maturity were significantly different according to DAP. Small ρ values in the factor of accession and its interaction with other factors indicated that the effects of accession on shoot growth and tuber development were small in this study. Based on these results, all data for the two accessions were averaged for subsequent analyses. The ρ values in the factor of year were also small but the analyses were separately performed because the difference in rainfall was distinguished between the years. On the one hand, the contribution of planting date on tuber dry weight was large. This indicates that the tuber weight differed significantly depending on the planting date. However, on the other hand, the contribution of planting date was small for the variation of shoot dry weight and tuber moisture content. This indicates that these traits were little affected by planting date.

Table 1. F-values and contribution (ρ) of the factors obtained by the ANOVA for shoot dry weight.

	Degree of freedom	Sum of squares	Mean square	F-value	ρ (%)
Year (Y)	1	10758	10758	19.9 **	1.4
Accession (A)	1	9387	9387	17.3 **	1.2
Planting date (P)	2	22094	11047	20.4 **	2.8
Days after plating (D)	2	146938	73469	135.7 **	19.4
Y × A	1	7	7	0.0 ns	0.0
Y × P	2	5240	2620	4.8 **	0.6
Y × D	2	49964	24982	46.1 **	6.5
A × P	2	2181	1091	2.0 ns	0.1
A × D	2	2117	1059	2.0 ns	0.1
P × D	4	13609	3402	6.3 **	1.5
Y × A × P	2	2476	1238	2.3 ns	0.2
Y × A × D	2	2653	1327	2.4 ns	0.2
Y × P × D	4	9367	2342	5.8 **	1.0
A × P × D	4	1745	436	0.8 ns	0.0
Y × A × P × D	4	586	147	0.4 ns	0.0
Residuals	874	474438
Total	909	741101

**p < 0.01, *p < 0.05, and ns: not significant.

Table 2. F-values and contribution (ρ) of the factors obtained by the ANOVA for tuber dry weight.

	Degree of freedom	Sum of squares	Mean square	F-value	ρ (%)
Year (Y)	1	23774	23774	5.8 *	0.4
Accession (A)	1	33580	33580	8.2 **	0.7
Planting date (P)	2	735075	367538	90.2 **	16.5
Days after plating (D)	2	1831082	915541	224.7 **	41.4
Y × A	1	217	217	0.1 ns	0.0
Y × P	2	34643	17322	4.3 *	0.6
Y × D	2	8082	4041	1.0 ns	0.0
A × P	2	1622	811	0.2 ns	0.0
A × D	2	42595	21298	5.2 **	0.8
P × D	4	138133	34533	8.5 **	2.8
Y × A × P	2	2144	1072	0.3 ns	0.0
Y × A × D	2	5778	2889	0.7 ns	0.0
Y × P × D	4	512	128	0.1 ns	0.0
A × P × D	4	12413	3103	0.8 ns	0.0
Y × A × P × D	4	14126	3532	3.5 ns	0.0
Residuals	367	1519548
Total	402	4265921

**p < 0.01, *p < 0.05, and ns: not significant.

Table 3. F-values and contribution (ρ) of the factors obtained by the ANOVA for tuber moisture content.

	Degree of freedom	Sum of squares	Mean square	F-value	ρ (%)
Year (Y)	1	9	8.8	0.6 ns	0.0
Accession (A)	1	94	93.7	5.9 *	0.4
Planting date (P)	2	360	180.0	11.3 **	1.7
Days after plating (D)	2	11938	5969.2	376.4 **	59.9
Y × A	1	6	5.6	0.4 ns	0.0
Y × P	2	85	42.6	2.7 ns	0.3
Y × D	2	93	46.4	2.9 ns	0.3
A × P	2	147	73.4	4.6 *	0.6
A × D	2	213	106.6	6.7 **	0.9
P × D	4	641	160.2	10.1 **	2.9
Y × A × P	2	100	50.1	3.2 *	0.3
Y × A × D	2	71	35.5	2.2 ns	0.2
Y × P × D	4	2	0.4	0.1 ns	0.0
A × P × D	4	181	45.2	2.9 *	0.6
Y × A × P × D	4	19	4.8	1.2 ns	0.0
Residuals	367	5915
Total	402	20284

**p < 0.01, *p < 0.05, and ns: not significant.

Figure 2 shows the changes in shoot dry weight, tuber dry weight, and tuber moisture content with increasing DAP. Because the DOY was different for the same DAP on different planting dates (Figure 1b), it should be noted that the meteorological conditions also differed depending on the planting dates. The changes in the traits according to DOY was shown in Supplemental Figure S1.

Figure 2. Changes in shoot dry weight, tuber dry weight, and tuber moisture content according to days after planting. Each point represents mean and standard errors for 12 plants consisting of four plants and three replications. Blue, red, and gray lines represent the early, middle, and late planting, respectively. The dashed vertical lines indicate the date of tuber harvest conducted at the same time for the plants from different planting dates.

Shoot dry weight in 2019 reached its maximum at approximately 150 DAP, regardless of the planting date. The maximum shoot dry weight in 2020 was similar to that in 2019, but growth was retarded compared to that in 2019. The DAP at the maximum shoot dry weight was similar for plants with different planting dates, but the onset of senescence differed. Plants in the later planting showed earlier senescence in terms of DAP bases. Between these two years, senescence occurred earlier in 2020 than in 2019 because of the early rain cessation.

The final tuber dry weight at full shoot senescence was the highest in the early planting and lowest in the late planting. The initial increases in tuber dry weight were similar in the early and middle planting periods until the second sampling was conducted at approximately 140 and 190 DAP in 2019 and 2020, respectively. After the second sampling, the increase in tuber weight slowed or stopped in the mid-planting period, whereas that in the early planting period continued until approximately 200 DAP. Increases in tuber weight during the late planting period were small and consistently observed in both years.

Tuber moisture content on different planting dates decreased similarly with increasing DAP. It decreased to 70–75% at approximately 200 DAP, regardless of the planting date.

4. Discussion

At the experimental site in the Guinea savanna, tuber enlargement did not depend on DOY but on DAP, indicating that day length had little effect on tuber enlargement in the white guinea yam used in this study. Therefore, the phase change from the vegetative growth stage to the tuber growth stage is mainly controlled by genetically determined growth characteristics (Poeting, 1990) rather than the photoperiod. The results obtained for white guinea yam were inconsistent with those for water yam, in which tuber enlargement started with a day length shorter than 12 h, regardless of the growth stage (Vaillant et al., 2005). Marcos et al. (2009) also reported that small changes in photoperiod in the tropics had a significant effect on tuber development in water yams. This inconsistency may have been caused by the different yam species. Shiwachi et al. (2002) concluded that the photosensitivity of tuber enlargement in white Guinea yam and water yam was weaker than that in other yam species, although sensitivity also differed between the early and late genotypes of the same species. Because our results were obtained for middle-maturity genotypes, which make up the majority of the genetic resources of white Guinea yam, this species was presumably less photosensitive than the water yam, even when varietal differences were considered.

In our previous studies conducted at the same experimental site, the final tuber yield was strongly correlated with the maximum shoot biomass during the vegetative growth stage, when the tuber yield was compared for yam plants with the same planting date (Iseki & Matsumoto, 2019, 2020). This is because large leaf area confers rapid tuber growth (Diby et al., 2011). However, in the present study, tuber yields were significantly different for different planting dates, although shoot biomass was slightly different. Even in late planting, sufficient shoot biomass was obtained at approximately 100 DAP, when tuber enlargement began. Therefore, maximum shoot biomass was not a major factor in the yield differences among the different planting dates.

The tendency for similar shoot biomass among the planting dates was consistent for the two years with different rainfall levels, although there was an almost three-fold difference in the total rainfall during the growth periods between early planting in 2019 and late planting in 2020. This is consistent with a previous report that concluded that yams are relatively tolerant to drought during the vegetative phase (Daryanto et al., 2016). Therefore, low rainfall during the early growth stage had little effect on the shoot growth and tuber yield of white guinea yam.

In contrast to shoot growth, soil drying during tuberization significantly affects tuber yield (Daryanto et al., 2016). This suggests that differences in water availability from the start of tuber enlargement to the end of the rain were the major cause of yield differences among the different planting dates. For the middle and late plantings, the rate of tuber enlargement slowed from earlier DAP (approximately 150–180 DAP and 100–150 DAP, respectively) than in early planting. This was because of the cessation of rainfall on DOY 300. Because the rain stopped immediately after the start of tuber enlargement for late planting, the rate of tuber enlargement was slow, and the duration for tuberization could not be ensured, resulting in the lowest yield.

In contrast, in early planting, the peak of tuber enlargement fell in the periods immediately after the restart of rain, following a dry spell around DOY230‒250. Higher water availability sustained shoot biomass for a longer period, facilitating tuber enlargement, and resulting in a higher tuber growth rate. This is because the newly assimilated carbon is primarily allocated to the tubers (Iseki et al., 2022). Therefore, the benefits of early planting were much greater than the risks of early drought during the vegetative growth period. The planting date should not be shifted to a later date, even in the year of a delayed rainy season, to avoid severe yield reductions due to early rain stopping, as observed in 2020.

Generally, the tuber moisture content at maturity is approximately 65%, regardless of the growth environment (Matsumoto et al., 2021). Shoot senescence is used as an indicator of tuber maturity (Sartie et al., 2012). In the present study, moisture content decreased to nearly 70% at 200 DAP, but shoot biomass remained at its maximum level in the early and middle plantings. This indicates that tuber maturity is independent of shoot senescence. Therefore, early harvesting before the rain stops is possible for early planting, which would improve farmers’ income by selling tubers at high prices before the general harvest periods (Fu et al., 2011).

5. Conclusion

A shift in planting dates has been employed by local farmers who are aware of changes in rainfall patterns (Baffour-Ata et al., 2023). However, our results revealed that the final tuber yield was rarely affected by low rainfall during the early growth stage, whereas severe yield reduction occurred when rain stopped during the tuberization period. Therefore, we conclude that early drought expected from early planting would have a limited effect on tuber yield, whereas late drought caused by late planting would have a high risk of severe yield reduction. This effect would be more and less pronounced for late maturity and early maturity accessions, respectively, but further studies are needed to reveal how the planting dates affect to the tuber yield of different maturity groups.

Acknowledgments

We thank Mr. Oyedele Sunday O., Mr. Oketokun Abass, Ms. Obaude Oyebola O., and Ms. Farinde Oluwaseun for their assistance with field management and data collection.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/1343943X.2024.2351623.

Additional information

Funding

This work was conducted under the JIRCAS research program “Design of crop breeding and food processing of indigenous resources to create new and diversified demands.” Part of the data used in this study were generated under the IITA-MAFF yam project of the IITA, supported by the Ministry of Agriculture, Forestry, and Fisheries (MAFF) of Japan.