Response of tef (Eragrostis tef [Zucc.] Trotter.) variety to varying seed rates and inter-row spacings for growth, yield and lodging severity in Western Ethiopia

ABSTRACT

Tef is a small-seeded staple crop endemic to Ethiopia. However, its yield is low, constrained partly due to improper agronomic practices and lodging. Thus, during the 2019–2020 cropping season, a field experiment was carried out at Reef Tokko Taanne Hamilee in Hababo Guduru district, Western Ethiopia to examine the effects of seed rate and inter-row spacing on growth, phenology, yield, and lodging severity of tef crop. Three levels of seed rates (10, 15, and 20 kg ha⁻¹) and three styles of row spacing (10, 15, and 20 cm) were combined in a factorial combination in a randomized block design with three replications.Tef variety called ‘Guduru’ was used as planting material. The result showed that seed rate and inter-row spacing had significant (p < 0.01) effects on all phenological and yield-related characteristics studied. Moreover, days to heading, days to maturity, grain yield and harvest index were significantly affected by the interaction effects of both seed rate and inter-row spacing. Higher days to heading (68 days) and days to maturity (127 days) were recorded at a seeding rate of 20 kg ha^–1  and10 cm inter-row spacing. Tef variety (Guduru) was taller (127.2 cm) under 10 kg ha^–1 seed rate as compared to a higher seed rate of 20 kg ha^–1 (120.8 cm). The highest dry biomass yield (51.3 kg plot⁻¹) was obtained from a seeding rate of 20 kg ha⁻¹ in combination with 20 cm inter-row spacing. Besides this, the maximum mean grain yield (19.8 kg plot⁻¹) was recorded from 20 kg ha⁻¹seed rate and 15 cm inter-row spacing to achieve a higher yield in the study area. Furthermore, the highest straw yield (35.99 kg plot⁻¹) was produced from the highest seed rate of 20 kg ha⁻¹, whereas the smallest inter-row spacing (10 cm) produced the highest straw yield (33.5 kg plot⁻¹). The maximum harvest index (56.1%) was obtained from 20 kg ha⁻¹ seed rate and 15 cm inter-row spacing, a similar treatment combination as in the case of grain yield. 10 kg ha⁻¹ seed rate and 20 cm inter-row spacing showed lowest lodging severity (24.1%) and (32.3%), respectively. The result suggests that sowing of tef at 20 kg ha⁻¹ seed rate combined with 10 cm inter-row spacing which resulted in higher grain yield, straw yield and harvest index, could be recommended in the study area to achieve higher tef production.

KEYWORDS:

• Lodging severity
• growth
• phenology
• plant population
• straw yield

Introduction

Tef (Eragrostis tef (Zucc.) Trotter.) is a widely grown and cultivated cereal grain in Ethiopia. It belongs to the Poaceae family and is an allotetraploid (2n = 4x = 40) crop (Paff and Asseng 2018). Tef has the lowest yield among Ethiopia's grains and ranks first in area coverage and second in total annual output after maize (Tesfahun 2018; CSA 2019). In terms of both production and consumption, it is Ethiopia's most valuable commodity (Tesfay and Gebresamuel 2016). Furthermore, its ability to tolerate and grow in a variety of vertisol soils with drainage concerns makes it a more acceptable and preferred cereal crop for farmers to plant inside the country (Workneh 2009). Tef accounts for around 24.17% (3,076,595.02 metric tons) of the total cereal grain output in the country followed by maize, sorghum, and wheat with 18.60%, 14.38%, and 13.73% of total grain production, respectively (CSA 2020).

When compared to other crops, tef has special agronomic and nutritional features, making it the most important crop. This is because of its outstanding ‘Injera’ (flattened bread) quality, which the people of the country often utilize as their main diet, long seed storability, drought tolerance, and high market value. Despite its several benefits, it has low productivity when compared to other major cereals, with an average yield of 1.8 tons per hectare across the country (CSA 2019). Tef straw is the most preferred for cattle feed, is used to fortify mud in plastering the wall of the house, and the grain can be stored for a longer period without storage pests (Tesfahun 2018).

Tef production is constrained by a variety of problems that can limit its yield potential, from lodging, inappropriate input usage, high post-harvest losses, and non-optimal seed rate (Tsegay et al. 2015;Wubante and Menzir 2017). Broadcast sowing is one of the most frequently used planting methods by local farmers which mostly encourages high crop density, which in turn increases resource competition, and eventually causes a reduction in tef yield due to plant density exceeding its optimum level for resource competition (Wubante and Menzir 2017). Controlling plant population density and seed distribution in the field is more difficult during planting of the small-seeded crop (Tareke 2010). However, optimum seed rate is essential for efficient resource utilization by the plants for better growth and yield performances, yet this is not commonly practiced by the farmers.

The inter-row spacing is another method in which optimal crop density is preserved for a higher yielding advantage than broadcasting. Mitiku (2008) found a substantial difference in raising tef yield components with decreasing seed rate from highest to lowest. Thus, it was recommended to establish the optimum density of plant population per unit area to produce the highest tef seed yield (ATA 2013; Tsegay et al. 2015). According to Tareke and Nigusse (2008), the majority of farmers use the broadcasting conventional method of sowing, which results in an excess crop density and promotes competition between the plants for nutrients, water, sunlight, and carbon dioxide (CO₂) absorption. Additionally, the disseminating method of sowing requires more seed per area than the row sowing method, leading to increases in production costs. Furthermore, lodging is a major constraint on tef production in Ethiopia and its productivity remains extremely low in part due to its susceptibility to lodging (Blösch et al. 2020; Muluken et al. 2020). To get the highest yield production, it is required to plant the ideal plant population density per unit area (Tareke 2009).

One study showed that the transplanted and drilled tef plots produced 3400–5100 kg ha⁻¹ with a three- to four-fold increase in grain production as compared to the broadcasted plot's 500–1200 kg ha⁻¹ (Tareke 2008). Tef tillers had enough room and caused a highly substantial rise in grain output at a decreased seed rate than broadcasted plots, and the highest plant height was achieved by using 10 kg ha⁻¹ seed rate with a 25 cm row spacing (Getahun et al. 2018). Another research finding by Alemat et al. (2016) showed an optimal inter-row spacing of 20 cm, indicating variation in the recommendation of rates may be due to difference in soil fertility, environmental differences and variety of tef.

Tef is grown as a major crop in Hababo Guduru in Horro-Guduru Wollega Zone, Ethiopia. However, farmers are still sowing tef by broadcasting an estimated 25 and 50 kg seed rate per hectare, causing yield reduction and crop lodging. Nowadays, an optimal seed rate and sowing strategy are getting increasingly important as necessary practices to minimize the effect of lodging in tef and increase its yield. There has been no clear guidance on the recommended inter-row spacing for drilling or transplanting the tef seeds, hence requiring the need to determine ideal tef seed rate and inter-row spacing for the region. Therefore, this study aimed to investigate the effects of seed rates and inter-row spacing on the growth, phenological characteristics, yield, and lodging severity of tef cultivar (Guduru) in the study area.

Materials and methods

Description of the study area

The research was carried out in Horro-Guduru Wollega Zone, Hababo Guduru District, Western Oromia, Ethiopia. The study region had a span of 1500–2400 metres above mean water level and was characterized by mid-altitude agroecology (m.a.s.l.). The realm's rainfall ranges from 500 to 1200 mm throughout the season, with temperatures ranging from 18° to 24°C all year. Farmers practice diverse farming methods (Figure 1).

Figure 1. Map of the study area.

Description of experimental materials

The experiment used a tef variety called DZ-011880 (‘Guduru’) that was released by Bako Agricultural Centre (BARC) and was generally appropriate and adaptive to the agroecology between 1850 and 2500 m.a.s.l, as well as within the rainfall range of 1000–1200 mm throughout the crop growing season. Because the seeds are white in color and mature in 132 days, the yield is estimated to be between 15 and 23 q ha⁻¹ at the research station and 14–20 q ha⁻¹ in the farmers’ field. As a main source of phosphorus nutrients, 100 kg NPS (18.9% N, 37.7% P₂O₅, and 6.9% S) and 100 kg urea ha⁻¹ were recommended as fertilizer doses for this cultivar. At the time of sowing, whole NPS and 50 percent urea fertilizer were applied. To meet the needs of the urea, a base dose and the remaining 50% following urea fertilizer were administered as top dressing after 20 days after sowing.

Experimental design

The method of treatment in three replications, three x three possibilities were spread out in a randomized complete block design with a factorial layout. The three seed rates as Factor A (10, 15 and 20 kg ha⁻¹) and three inter-row spacings as Factor B (10, 15 and 20 cm) were duplicated three times and randomly dispersed, resulting in a total of nine treatment combinations. For proper separation, the unit plot size was 2 m × 2 m = 4 m², with 0.5 m spacing between plots and 1 m between blocks.

Experimental procedure and land management

For this experiment, land was chosen from the sphere of advanced farmers in the study area. Requests were made of a single farmer's farm for a plot of land that was sufficient for performing the experiment without payment, and only an oral agreement was made to provide the gathered produce to the farmer in situ when the data collection was completed. Farmers were not involved in data collection, only members of the research team undertook this task. Seeds from the DZ-011880 (‘Guduru’) variety were used for sowing since they were clean and healthy. Before sowing, the chosen research site was adequately prepared by being correctly laid out, cleaned, and ploughed three to four times with oxen. The initial ploughing was done to a depth of 25–30 cm. With the aid of human labor and the traditional hoe, the land was appropriately leveled. According to the treatment combination, the rows were built. Seeds were manually drilled at the rate of 10, 15, and 20 kg ha⁻¹ after seed beds were thoroughly prepared, as per the experimental protocol. At the time of planting, blended NPS fertilizer was administered at a rate of 100 kg ha⁻¹ to the middle of the rows for each unit plot, and half of the urea was applied at the time of sowing, therefore the remaining half dose was applied as fecundation at the tillering stage to meet the remaining nitrogen needs while also reducing nitrogen loss through leaching and volatilization. On 12 July 2019, the sowing was completed. For all of the experimental plots, the entire agronomic package of procedures necessary for tef, including hoeing, weeding, and control techniques for insect and pest incidence, were followed consistently.

Data collection

Phenological data

Days to 50% heading: This was recorded as the number of days from the date of sowing up to the date when 50% tips of the panicles first emerged from the main shoot of the plant per plot.

Days to 90% maturity: This was recorded as the number of days from the date of sowing up to the date when 50% of the crop stand – stems, leaves, and floral bracts – in a plot changed to light yellow color, based on visual observation. It was indicated by the senescence of the leaves as well as the free threshing of grain from the glumes when pressed between the forefinger and thumb.

Growth data

Plant height (cm): This was calculated as the average of ten plants that were randomly selected from the middle rows of a net plot area and measured using a measuring scale from the base of the main stem to the tip of each chosen plant.

Number of total tillers: This was measured at the point of physiological maturity, also known as the dough stage, by counting all the tillers from ten randomly chosen plants that were located in the middle rows of the net plot area, and the average mean was taken into account.

Panicle length (cm): This was measured using a measuring scale for ten randomly chosen samples of plants in the middle rows of the net plot area, from the node (where the initial panicle branches emerged) up to the tip of the panicle and the average mean was then taken into consideration.

Yield and yield components of tef

Dry biomass yield (kg plot⁻¹): After being sun-dried, the weight (kg) of the entire above-ground plant biomass – which includes leaves, stems, seeds, and chaff – of all the crops collected from the net plot area was recorded at the point of maturity and this weight was then translated to kg plot⁻¹.

Grain yield (kg plot⁻¹): This was measured from the net plot area and converted to grains in kg plot⁻¹ after threshing and sun-drying the grains to a moisture content that was nearly optimal.

Straw yield (kg plot⁻¹): It was obtained by subtracting the entire grain yield after threshing in kg plot⁻¹ from the total above ground biomass yield in kg plot⁻¹.

Harvest index (HI) (%): This was the proportion of grain yield per plot to above-ground biomass and expressed in percent.

Lodging severity (%): Caldicott and Nuttal's (1979) recommendations were followed while recording on a scale of 0–5.

Data analysis

All the data collected was subjected to the analysis of variance (ANOVA) using SAS software version 9.3 (SAS 2014) programs. When ANOVA showed significant differences, the mean separation test was carried out using the Least Significant Difference test at a 5% level of significance (Steel et al. 1997).

Results and discussion

Analysis of variance

The analysis of variance showed that seed rate and inter-row spacing had significant (p < 0.001) effects on days to heading, days to maturity, plant height, dry biomass yield, grain yield, straw yield, harvest index and lodging severity of tef. Besides this, plant height and lodging severity were significantly affected by seed rate and inter-row spacing, whereas panicle length was not significantly influenced by these parameters (Table 1). The interaction effects of both seed rate and inter-row spacing caused significant effects on days to heading, days to maturity, grain yield and harvest index (Table 1). This revealed that seed rate and inter-row spacing had notable influence, especially on phenological, yield, and yield components.

Table 1. Analysis of variance on growth, phenological, yield components and lodging severity under different seed rates and row spacing of tef cultivar in Hababo Guduru district, Ethiopia in 2019–2020.

		Growth, phenology, yield components and lodging of tef
Source of variation	DF	DH	DM	PH	NTt	PL	DBMY	GY	SY	HI	LS
Replication	3	0.04	0.11	0.01	0.12	8.03	9.92	0.67	10.31	11.80	6.62
Seed rate (SR)	2	49.00**	52.30**	120.30**	4.12**	2.59^ns	224.79**	119.26**	668.88**	1492.63**	1720.75**
Inter-row spacing (RS)	2	0.25*	14.30**	10.08**	0.07^ns	1.00^ns	133.47**	9.33**	209.29**	278.50**	54.20**
SR * RS	4	4.25**	1.67**	0.41^ns	0.26^ns	0.96^ns	12.24^ns	3.13**	16.40^ns	62.54**	3.50^ns
Error	24	0.06	0.11	0.13	0.21	0.99	10.00	0.29	11.07	9.69	2.96
CV (%)		0.37	0.27	0.28	13.6	1.88	7.0	3.77	10.8	9.4	5.0

Notes: Where DF = degree of freedom, ** and * showed significance at 1% and 5% probability levels, respectively, ns = non-significant at 5% significance level; DH = day to 50% heading, DM = day to 90% maturity, PH = plant height (cm), NTt = number of total tillers, PL = panicle length (cm), DBMY = dry biomass yield (kg plot⁻¹), GY = grain yield (kg plot⁻¹), SY = straw yield (kg plot⁻¹), HI = harvest index (%), and LS = lodging severity (%).

Phenological characteristics

Days to 50% heading

The mean separation test showed that the greatest value (68) days to heading was noted at a high seeding rate of 20 kg ha⁻¹ at a 10 cm inter-row spacing, whereas the lowest value (62) days to heading was noted at a high seed rate of 20 kg ha⁻¹ and a wider inter-row spacing (20 cm) (Table 2). At narrow spacing, the days to heading time increased with increased seed rate, and this may be due to the influence of resource competition. Days to heading time were higher at 10 cm inter-row spacing in combination with the three seed rate levels than in other combinations, indicating that the narrower spacing had a longer tef heading time (Table 2). Adugna (2017) discovered that the phenological parameters of tef were considerably influenced by seeding rates. Similar to this, Hoshikawa (1975) discovered that increased planting density accelerated early heading and flowering in rice by changing the sequence in which plants, hills, and populations headed and flowered. The author also stated that the heading time was shortened by 4–5 days when the number of productive tillers per plant was low, but required more days when the number of productive tillers was high. Additionally, Abraham (2014) observed that a higher seeding rate resulted in earlier tef heading. However, according to Adugna (2017), sowing at a rate of 25 kg ha⁻¹ significantly delayed heading by 3.6 days when compared to seeding at 2.5, 5 and 7.5 kg ha⁻¹.

Table 2. Mean values of days to heading, days to maturity, biomass yield, grain yield, and harvest index as affected by the interaction effect of seed rate and inter-row spacing on Guduru tef variety at Hababo Guduru district, Ethiopia in 2019–2020.

	Days to 50% heading			Days to 90% maturity			Biomass yield (kg plot^−1)			Grain yield (kg plot^−1)			Harvest index (%)
	Inter-row spacing (cm)
Seed rate (kg ha^−1)	10	15	20	10	15	20	10	15	20	10	15	20	10	15	20
10	66.0	64.75	64.0	124.0	122	120	48.9	42.7	49.3	12.6	16.4	11.69	25.6	38.5	23.6
15	67.0	64.75	63.0	125.0	123	122	47.2	41.9	49.3	13.3	17.3	11.69	28.2	42.4	23.1
20	68.0	64.00	62.0	127.0	124.	122	45.2	35.4	51.3	14.2	19.8	11.80	31.3	56.1	27.7
LSD (5%)		0.20			0.28			2.66			0.45			2.60

where LSD = least significant difference.

Days to 90% maturity

The highest days to maturity was 127 days at a seed rate of 20 kg ha⁻¹ with a 10 cm inter-row spacing, while the lowest value (120 days) was obtained at a seed rate of 10 kg ha⁻¹ with a 20 cm inter-row spacing, demonstrating that low plant population and wider inter-row spacing speed up the maturation process (Table 2). Similarly, Sate and Tafese (2016) reported that days to 90% physiological maturity of tef was significantly affected by the main parameter of seed rates, and a reduction in the number of days to reach maturity with decreasing seed rate from 25 to 10 kg ha^–1 might be due to the availability and use of growth resources. Additionally, according to Adugna (2017), 25 kg ha⁻¹ of planting considerably delayed maturity by 2 days when compared to 2.5, 5 and 7.5 kg ha⁻¹. In contrast, Abraham (2014) found that higher seeding rate accelerated tef maturity, while Bekalu and Arega (2016) found no significant affect of seed rate on maturity.

Growth data

Plant height (cm)

The main effect of seeding rate showed a highly significant (p < 0.01) effect on plant height of tef, as did inter-row spacing (p < 0.01), although the interaction effects of seed rate and spacing was not significant (Table 1). Tef variety (Guduru) was taller (127.2 cm) at 10 kg ha^–1 seed rate as compared to a higher seed rate of 20 kg ha^–1 (120.8 cm) (Table 3). This indicated that plants grow taller at low plant population, while decreases in height when the population has increased may be associated with the lack of available resources. In contrast, Abraham (2014) noted that the tallest plants were produced using a seeding rate of 25 kg ha⁻¹, whereas the smallest plants were produced using a seeding rate of 2.5 kg ha⁻¹. Consequently, tef plants grown from seed at a rate of 25 kg ha⁻¹ were on average 14% taller than those grown from seed at a rate of 2.5 kg ha⁻¹. On the other hand, in our study the variation in inter-row spacing from 10 cm to 20 cm had a smaller effect on plant height in absolute terms (123.1–124.9 cm) but this difference still reached statistical significance (Table 3). In tef, plant height is among the characteristics that show a moderate-to-strong correlation with the lodging index (Dejene 2019).

Table 3. Main effects of seed rates and row spacings on plant height (cm), number of total tillers per plant, panicle length (cm), straw yield (kg ha⁻¹), and lodging severity percent of Guduru tef variety at Hababo Guduru district, Ethiopia in 2019-2020.

Treatments	Parameters
Seed rate (kg ha^−1)	PH	NTt	PL	SY	LS
10	127.2^a	3.51^a	53.3^a	33.8^a	24.1^c
15	124.0^b	3.80^a	52.4^b	22.1^b	31.8^b
20	120.8^c	2.69^b	53.0^ab	35.9^a	47.5^a
LSD (0.05)	0.30	0.38	0.83	2.8	1.45
Row spacing (cm)
10	123.1^c	3.4^a	52.6^a	33.5^a	36.6^a
15	124.0^b	3.3^a	53.1^a	32.6^a	34.6^b
20	124.9^a	3.4^a	53.1^a	25.8^b	32.3^c
LSD (0.05)	0.30	0.38	0.84	2.8	1.5
CV (%)	0.28	13.6	1.88	10.8	5.8

Notes: Means in the same column within the same treatment category followed by different letters are significantly different at p < 0.05, probability level. CV: coefficient of variation; NTt: number of total tillers per plant: PH: plant height(cm); PL: panicle length (cm); SY: straw yield (kg plot−¹); LS: lodging severity (%).

Number of total tillers per plant

The main effect of seeding rate showed significant effect (p < 0.01) on the number of total tillers per plant while the main effect of inter-row spacing and the interaction effect of both factors had non-significant influence on number of total tillers (Table 1). Thus, relatively highest total tillers per plant (3.8) was recorded at a seed rate of 15 kg ha⁻¹, whereas low tillering (2.69) was obtained at 20 kg ha⁻¹ seed rate showing a tendency of reduction of tillering capacity of tef under an increasing seed rate. On the other hand, under varying row spacing, there was no significant difference in total tillers per plant revealing that intra-row plant density is more important than inter-row spacing for tillering capacity of tef (Table 3). This showed that the overall growth and size of the growing plant was reduced under dense planting and the number of total tillers was smaller. According to Fekremariam et al. (2014), broadcast tef had the fewest tillers in relation to row spacings. As the seeding rate level dropped, more total tillers per square metre of tef were produced (Shiferaw 2012). According to a study by Abraham et al. (2018), row-sown tef plants generated 32% more productive tillers per plant than broadcast tef plants.

Panicle length (cm)

The variation in seed rate from 10 to 20 kg ha^–1 made no significant difference on panicle length (53.0–53.3 cm) and similarly the variation in inter-row spacing from 10 cm to 20 cm caused a negligible difference of 52.6 cm to 53.1 cm (Table 3). Therefore, in the present study, the effect of seed rate and inter-row spacing similarly caused nonsignificant variation on panicle length of tef. Some reports indicate that the angle of the panicle contributes to the likelihood of lodging in tef. In comparison to the other kinds, cultivars with compact panicles and shorter heights have higher lodging resistance. It was discovered through comparing several types that tef landraces generally lodged less than improved cultivars (Blösch et al. 2020).

Yield and yield components of tef

Dry biomass yield (kg plot⁻¹)

The highest dry biomass yield (51.3 kg plot⁻¹) was achieved with a seeding rate of 20 kg ha⁻¹ in combination with 20 cm inter-row spacing, while the lowest dry biomass yield (35.4 kg plot⁻¹) was obtained from a treatment combination of 20 kg ha⁻¹ and 15 cm inter-row spacing (Table 2). It appears that at the widest inter-row spacing (20 cm), the biomass yield was higher and the biomass yield tends to increase at increasing seed rate under wider spacing, which may be triggered due to an increased plant population and the higher tiller formation under wider spacing. As a result, seed rates and inter-row planting methods have demonstrated considerable influence on dry biomass output of the crop, all of which have an effect on yield. Similarly, Sate and Tafese (2016), indicated sowing method and seed rate had a significant effect on tef biomass with greater biomass yield obtained from inter-row sowing than broadcast sowing and a considerable increase in dry biomass yield was gained by decreasing the seed rate from 25 to 10 kg ha⁻¹. In contrast to our study, a report showed that seeding rates of 5 kg ha⁻¹ generated higher biomass yields than seeding rates of 10, 20, and 25 kg ha⁻¹ (Bekalu and Arega 2016).

Grain yield (kg plot⁻¹)

Highest mean yield (19.8 kg plot⁻¹) was recorded from 20 kg ha⁻¹ seed rate and 15 cm inter-row spacing followed by 16.4 kg plot⁻¹ from 10 kg ha⁻¹ seed rate and 15 cm inter-row spacing (Table 2), while the lowest grain yield (approximately 12 kg plot⁻¹) was produced at the widest inter-row spacing (20 cm) at all seed rates tested. Thus, the result showed that sowing at 20 kg ha⁻¹ seed rate with 15 cm inter-row spacing resulted in higher yield, implicating this as optimum conditions; at 15 cm inter-row spacing an increasing seed rate revealed a tendency of yield increment. Getahun et al. (2018) found that at 10 kg ha⁻¹ seeding rate, a 25 cm spacing gave greatest grain yield, and at 15 kg ha⁻¹, a 20 cm spacing gave greatest grain yield; but broadcasting sowing yielded the lowest grain yields (Paff and Asseng 2018). Another finding by Shifera et al. (2020), indicated that highest grain yields of tef were recorded from at 10 kg ha⁻¹ seed rate and 25 cm row spacing.

Straw yield (kg plot⁻¹)

The statistical analysis of variance showed the presence of a significant difference in straw yield from the main effect of seed rate and inter-row spacing whereas the interaction effects of seed rates and row spacing were not significant (Tables 1 and 3). The highest straw production (35.99 kg plot⁻¹) was obtained from the highest seed rate of 20 kg ha⁻¹, while the lowest straw yield (22.1 kg plot⁻¹) was from a seed rate of 15 kg ha⁻¹. On the other hand, regarding the effect of inter-row spacing, the highest straw yield (33.5 kg plot⁻¹) was recorded with the smallest inter-row spacing (10 cm), followed by 15 cm inter-row spacing (32.6 kg plot⁻¹). This shows that seed rate and planting method have influence on tef straw yield. Tef straw is specifically valuable as animal feed following its food use, which is probably one of the major reasons that forces farmers to grow tef. Therefore, the importance of tef straw is not significantly different from its use for consumption, which makes tef special from other crops. Large and small seed rates may have affected vegetative growth in terms of plant height and the number of tillers, which could explain why straw output rose or decreased in some combinations (Shifera et al. 2020). According to Tareke and Nigusse (2008), reducing the seeding rate boosted tef straw yields. However, Wubante and Menzir (2017) found that increasing row spacing reduced tef straw yields. Other research (Shifera et al. 2020), showed that raising row spacing from 15 cm to 20 cm and reducing seed rate from 25 to 15 kg ha⁻¹ both considerably increased the straw yields in tef.

Harvest index (%)

The highest harvest index (56.1%) was obtained from a treatment combination of 20 kg ha⁻¹ seed rate and 15 cm inter-row spacing, while the lowest harvest index value (23.1%) was from 15 kg ha⁻¹ seed rate and 20 cm inter-row spacing. The finding showed that at increasing seed rates, the harvest index value showed an increasing trend for 10 and 15 cm inter-row spacing but the amount of increase was larger for 15 cm inter-row spacing (Table 2). This might be because the crop is able to employ its growth resources more effectively to maximize larger sizes of plant height, leaf length, and leaf number, which in turn results in declines in the grains. Row spacing may have this effect by reducing interspecific plant competition and leading to fewer plants in the plot. Contrary to our findings, some researchers have found that decreasing seed rate from 25 to 10 kg ha⁻¹ and increasing row spacing from 15 to 25 cm resulted in a noticeably rising trend in the harvest index. Lower seed rates significantly increased the harvest index of tef, and vice versa, with the use of seed rate at 10 kg ha⁻¹ producing a higher harvest index of tef (45%), followed by 15 kg ha⁻¹ producing a harvest index of 40% (Shifera et al. 2020). The higher harvest index reached at the lowest seed rate was attributed to increased light penetration through the plant canopy and increased nutrient availability. Similar to this, Tareke (2010) claimed that decreasing the seed rate in tef resulted in a higher harvest index. Harvest index is useful to assess the flexibility of converting dry matter into economic yield, and can be used to express the link between total biological yields of the crop (Marschener 1995).

Lodging severity (%)

The maximum lodging severity (47.58%) occurred at 20 kg ha⁻¹ seed rate while the lowest lodging severity (24.1%) was recorded at 10 kg ha⁻¹. On the other hand, at the wider inter-row spacing (20 cm), the lowest lodging severity (32.3%) was recorded, whereas the highest severity (36.6%) was obtained from 10 cm inter-row spacing (Table 3). Therefore, the results showed that an increased seed rate favors lodging, and an increased inter-row spacing reduces lodging severity. This may be because at higher seed rates and low inter-row spacing, there is an enhanced plant density and biomass causing quick vegetative development and succulent stem elongation in tef, resulting in lodging. Lodging is a major problem in small cereal crops, reducing photosynthesis within the canopy, damaging vascular bundles by bending or breaking stems, and causing mechanical harvesting issues, all of which have an impact on yield and quality. The grains of stuck plants may germinate on the panicle in cultivars with limited seed dormancy. As a result, grain quantity and quality suffer severe losses due to lodging. Furthermore, it complicates harvesting operations, raises the demand for grain drying, and, as a result, raises costs. Hence, in Hababo Guduru district, usage of 10 kg ha⁻¹ seed rates with 20 cm inter-row spacing would be important in reducing the logging severity of Guduru tef cultivars. In another study, Muluken et al. (2020) indicated the lodging index based on visual score ranged from 22% to 100% at Adet, and from 12.5% to 100% at Bichena, Ethiopia. Dejene (2019) reported variability in the lodging index of tef among three locations, Alem Tena (75.0%), Debrezeit (52.5%) and Holeta (62%) in Ethiopia.

Conclusion

The result showed that the main parameters of seed rate and inter-row spacing had significant (p < 0.01) effect on all phenological and yield related characteristics studied. Moreover, days to heading, days to maturity, grain yield and harvest index were significantly affected by the interaction effects of both seed rate and inter-row spacing. Thus, higher days to heading (68 days) and days to maturity (127 days) was recorded at seeding rate of 20 kg ha^–1 and10 cm inter-row spacing. Tef variety (Guduru) was taller (127.2 cm) under 10 kg ha^–1 seed rate as compared to a higher seed rate of 20 kg ha^–1 (120.8 cm). The highest dry biomass yield (51.3 kg plot⁻¹) was obtained from a seeding rate of 20 kg ha⁻¹ in combination with 20 cm inter-row spacing. Besides this, the maximum mean grain yield (19.8 kg plot⁻¹) was recorded from 20 kg ha⁻¹ seed rate and 15 cm inter-row spacing to achieve higher yield in the study area.

Furthermore, the highest straw yield (35.99 kg plot⁻¹) was produced from the highest seed rate of 20 kg ha⁻¹ whereas the smallest inter-row spacing (10 cm) produced the highest straw yield (33.5 kg plot⁻¹). Maximum harvest index (56.1%) was obtained from 20 kg ha⁻¹ seed rate and 15 cm inter-row spacing, a similar treatment combination as in the case of grain yield. 10 kg ha⁻¹ seed rate and 20 cm inter-row spacing showed lowest lodging severity (24.1%) and (32.3%), respectively. The results suggest that sowing of tef at 20 kg ha⁻¹ seed rate combined with 10 cm inter-row spacing which resulted in higher grain yield, straw yield and harvest index could be recommended in the study area to achieve higher tef production.

Authors' contributions

To complete this project, all of the authors collaborated. AT and BCN were responsible for the study's design and data collection. BCN interpreted the information and drafted the first and original draft of the manuscript. The manuscript was modified, read, and approved by authors BCN and ZJ. The statistical analysis was done by authors ZJ and KL.

Acknowledgements

We appreciate the administrative assistance provided by Ethiopia's Oromia Agricultural and Natural Resources Office during the study period.

Disclosure statement

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

Data availability statement

The data that backs up this study's findings is publicly available in figshare.com at https://doi.org/10.6084/m9.figshare.22794620

Additional information

Funding

The author(s) reported there is no funding associated with the work featured in this article.