Effect of planting date on tepary bean yield and yield components sown in Southern Botswana

Tepary bean is a crop that is slowly gaining momentum into the cropping system of Botswana. A study was conducted to determine the effect of planting date on yield and yield components of tepary bean and identify the optimum sowing of the crop. A rain-fed experiment was conducted for two seasons (2012 to 2014), using a randomized complete block design replicated three times in a split plot with three sowing dates in main plots and the nine genotypes in sub plots. Genotype × planting date interaction was not significant among the characters (100 seed weight, shoot dry weight, seed yield, crop loss, plant population, and plant height) which implied stability of the utilized genotypes. However, planting dates and year, and also the interaction between planting date and year were highly significant (P>0.001) for all the characters. Plant height had greater association with shoot dry weight (0.783), 100 seed weight (0.70) and seed yield (0.65), suggesting that it is a character which can be useful in selection for improving the tepary bean productivity. Sowing tepary bean in December and January were found to be good options for farmers in the southern part of Botswana.


INTRODUCTION
The world today depends on few number of crops species for food mainly cereals such as wheat, rice and maize; this leaves a significant number of crops with potential benefit neglected (Collins and Hawtin, 1999;Azam-ali, 2010).Climate change and global populations are key issues forcing researchers to be innovative enough to bring about changes in the crop production systems in order to achieve the world food demand (McClean et al., 2011).Climate change put some pressure on food production particularly on major crops, it is crucial to have some options in the likes of underutilized or orphan crops (Mayes et al., 2014).Therefore, further research and development of minor crops such as tepary bean could be useful in the forthcoming new environments (Porch et al., 2012).
Tepary bean (Phaseolus acutifolius A. Gray) is a drought tolerant but neglected crop that has the potential to provide greater resilience to cope with the climate *Corresponding author.E-mail: omolosiwa@gov.bw.
Author(s) agree that this article remain permanently open access under the terms of the Creative Commons Attribution License 4.0 International License change challenge (Blair et al., 2012;Jiri et al., 2017).The cultivated tepary is an annual legume that originates from the Mesoamerican region in the Sonora desert; it is well adapted to hot arid climates (Thomas et al., 1983).It has received tremendous interest as an arable field crop in a number of regions around the world (Miklas et al., 1998).The crop is preferred for its high biotic and abiotic stress tolerance (Porch et al., 2013), high protein content in edible grain and its fodder value (Bhardwaj, 2013).As a short duration and high yielding crop, this makes it a quick and easy crop to grow (Hamama and Bhardwaj, 2002) and this is a good character for low and erratic rainfall in Botswana.Tepary bean is one pulse crop that is highly promoted for consumption, especially since there is little or no production of common bean in the country.However, there is inadequate documented information on the per capita tepary bean consumption in the country.Planting date is a key factor in the production of crops in semi-arid environment of Botswana (Moatshe et al., 2015).However, the length of growing crops has significantly decreased; the onset and cessation of growing crops have been October (onset) and March (cessation), respectively but this trend has shifted (Adelabu et al., 2010).The decrease in the growing season is attributed mostly to limited rainfall, occurrence of frost and extreme evapotranspiration (Weare, 1971).The crop yields are however, highly dependent on seasonal rainfall (Tsheko et al., 2015), which is affected by the period of planting (Ezeaku et al., 2015).In addition, changes in the rainfall patterns of Botswana are clearly noticed in high inter-annual variation in the rainfall onset, more number of dry days and decreasing amounts of rainfall at the onset and cessation (Simelton et al., 2012).Climate change has caused some changes in the growing seasons in a number of other regions and these lead to a reduction in crop yield (Ezeaku et al., 2015).Negative impacts of climate change on food production can have a serious consequence on food security in Botswana, since to a great extent it is relying on imports for her food requirements (Dube, 2003).
Earlier work on tepary bean revealed that planting dates significantly affected seed yield seed weight and harvest index when tested over three planting dates in Virginia, United States of America (Bhardwaj et al., 2002).The time of planting is important in determining the final seed yield and is a useful agronomic practice as observed in other leguminous crops such as cowpeas (Shiringani and Shimelis, 2011), bambara groundnut (Ngwako et al., 2013) and faba bean (Thalji and Shalaldeh, 2006).Planting dates can also be used effectively to control several pests in cowpea (Ezeaku et al., 2014) and corn rootworn damage in maize crop (Obopile et al., 2012).Therefore, the technique can be a good cultural control method of pest and diseases among subsistence farmers with limited access to resources (Akande et al., 2012).
In Botswana, the sowing date is dependent on rainfall and planting season can start from September to early February and most planting occurs between September and December (Moatshe et al., 2015).However, the cessation of growing season is based on the occurrence of the minimum temperature of ≤3°C to avoid the growth of crops to coincide with the time when black frost occurs resulting in plant senescence (Adelabu et al., 2010).It was discovered that the period during which frost may occur is much longer or comes much earlier in the Southern Botswana than further north (Andringa, 1984).Hence, the planting dates in Botswana are structured in such a way that the last planting dates in the north and south of the country are usually mid February and end of January season, respectively.Therefore, it is usually critical to know the duration of crops to reach physiological maturity to assist establish the best time to plant when conditions are conducive.However, based on some circumstances such as lack of resources, extension dates are provided for farmers to do their planting after the proposed planting dates (Adelabu et al., 2010).Date of sowing is one aspect of crop management that is well explored for a number of crops at the Department of Agricultural Research, Gaborone, Botswana.Studies were conducted to compare early planting (December), intermediate (January) and late planting (February) for several crops.The results suggest that early planting was most suitable for most crops such as groundnut (DAR, 1991), cowpeas and sunflower (DAR, 1988).While for mungbean early planted crops produced better yields, but sowing in January did not significantly affect number of pods and start of flowering, therefore, the crop could be planted between 20th December and 20th January (DAR, 1989).The research also revealed that the optimum planting date is not always suitable for all the cultivars used and this makes determining the right time of planting a challenge.No sowing dates studies have been conducted in Botswana for tepary bean crop, the current changes of the growing season also warrant a study to determine the best planting date for this crop in the country.Therefore, the objective of this study is to determine the effect of planting date on yield and yield components of tepary bean and identify the optimum sowing date for tepary bean in Southern Botswana.

Experimental site
Two field trials were conducted at the Department of Agricultural Research Station in Sebele, Botswana (24°35'S; 25°56'E, 991 m); the climate is semi-arid with a 30 year average annual rainfall of 500 mm.The rains generally starts in October and end in April, and over 90% of the rainfall occurs during November to March (Weare, 1971), however, the rains are not evenly distributed both spatially and temporally with long dry spells within the season.The field site soils are shallow, ferruginous soils, consisting of medium to coarse sandy and sandy loam with low water holding capacity.The site was ploughed and disc harrowed to level the soil and prepare the seed bed for planting.The 2012 to 2013 season was declared a drought year due to prolonged moisture deficiency across the whole country, while 2013 to 2014 season was not considered a drought year (Statistics Botswana, 2015).

Genotypes used
The experiment was performed with a set of nine tepary bean genotypes (GK010, E19, E89, GK012, E105, GK011, E70, GK013 and Motsumi), at three sowing dates (December 19th January 21st and February 19th in the 2012 to 2013 season).The same genotypes were planted in the second season of 2013 to 2014 at three sowing dates (December 18, January 21 and February 19th).Sowing in each season was conducted after every four weeks.December sowing dates are considered early planting while January sowing dates are intermediate planting and the February sowing dates are the late planting dates.These days are within the planting times recommended for Botswana.

Experimental layout
A randomized complete block design was replicated three times in a split plot with three sowing dates in main plots and the genotypes in sub plots.A total plot area of 35 m×54 m was set up after soil preparation by ploughing, and disc harrowing the soil.A unit plot area of 3 m × 5 m, with a row spacing of 0.75 m, and between plant space of 0.2 m, with an expected plant population of 67000 per ha was established.

Data collection and analysis
Climatic data were obtained using weather station (Pessl Instruments, Weiz, Austria).weather station monitors air temperature, precipitation, air humidity, solar radiation, wind direction and speed, dew point and leaf wetness; the data was recorded hourly.The weather station was located less than 2 km from the planting sites.Data for 2012 to 2013 and 2014 crop season were used to relate the crop yield to the weather conditions.
The agronomic data recorded were the initial plant stand recorded 21 days after sowing; final plant stand recorded at harvest when the plant has reached maturity at 60 to 70 days from sowing.Shoot dry weight is above ground biomass averaged from five plants per plot, it was oven dried (Oven, series 2000) for 48 h at 70°C.Seed yield is the weight of seed recorded per plot collected within the two middle rows, crop loss percentage was estimated from subtracting the initial plant stand from final plant stand, plant population was estimated from the final plant stand count, and plant height was measured from an average of 5 plants per plot within the two middle rows.Hundred seeds were counted using a seed counter (Numigral) and 100 seed weight was measured using (AND HR-300i), from each plot.Data was analyzed based on statistical package SAS version 9.4 and the means were compared using Duncan Multiple Range Test.

RESULTS
There were some differences on meteorological observations between 2012 to 2013 and 2013 to 2014 crop seasons (Table 1).The 2012 to 2013 season experienced and average rainfall of 334 mm much less than the 30 year average and temperature of 22.8°C.(Table 1).Only December month received more than Molosiwa and Kgokong 139 100 mm of rainfall, therefore most field crops in the country failed to produce yield.The 2013 to 2014 season received an average rainfall of 467 mm and an average temperature of 21.9°C (Table 1).Low genetic variation was observed for 100 seed weight, shoot dry weight, seed yield, crop loss, plant population and plant height among the nine genotypes (Table 2).The average 100 seed weight was relatively low at 7.74 g and a low average seed yield of 136.4 kg/ha, attributed possibly to low rainfall received in both seasons (Table 1).However, E105, GK011 and GK013 produced seed weight of more than 8 g, while E70 produced the highest seed yield of 181.57kg/ha, within the two seasons.
Early planting (December) in the first season (2012 to 2013), increased 100 seed weight, shoot dry weight, seed yield, plant population and plant height compared to the January planting.While the intermediate (January) planting led to higher crop loss at 64% and reduction in all the selected characters (Table 3).Late planted crops in February did not yield any results; therefore, it is strongly discouraged to grow crops late, when the rainfall is below normal as it was received in 2012 to 2013 season.Some contrasting results were observed in the second season (2013 to 2014), since early planting in December led to a reduction in 100 seed weight, shoot dry weight, seed yield, and plant population as compared to the crops planted in January (Table 3).However, crop losses between the early and the intermediate planted crops were not significantly different in the 2013 to 2014 season.Higher crop losses at 77% were observed in the late planted crops, which led to significantly lower 100 seed weight, shoot dry weight, seed yield and also plant height.In the second season of planting significantly taller plants were identified in the intermediate sowing date and this lead to relatively higher yields 679 kg/ha (Table 3).
A combination of the three planting dates averaged over two years indicated that December and January planting are not significantly different in terms of 100 seed weight and shoot dry weight (Table 4).Early planted crops were not significantly different from the late planted crops for seed yield, crops lost and plant population.The yield realized for these two seasons were relatively low, early planting and late plantings were not significantly different because both plantings had significantly lower plant populations.Poor germination occurred in the early planting and late plantings possibly because the crop received low moisture and the heat dried up the seedlings.Even though January plantings overall seed yield are higher at 348 kg/ha, it could be because intermediate planted crops benefited from moisture received in December and some rains in February and March (Table 1).The January planted crops were also significantly taller than those planted early and late (Table 4).
The analysis of variance results for planting date, genotype and year and the interactions for the two seasons are presented in Table 5.The genotype effect was not significant for the six characters selected which is an indication of low genetic diversity among the nine cultivars.The genotype × planting date (G × E) was not significant which also imply the stability of the given cultivars in the three environments in the two seasons.However, planting dates and year, and also the interaction between planting date and year were highly significant (P>0.001) for all the selected characters (Table 5).The time of planting on the six selected characters for tepary beans production is very important.The two seasons were significantly different from each other as shown in Table 5 which reveals the contrasting environment experienced by the crops.The pooled correlation matrix was highly significant (P< 0.001) among most characters (Table 6).The results illustrated that plant height had highly significant and positive association with shoot dry weight (0.783), 100 seed weight (0.70) and seed yield (0.65) demonstrating that these characters can be useful in selection for improving tepary bean productivity (Table 6).Some negative and non-significant associations were observed between crop lost to shoot biomass (-0.108) and seed yield (-0.027), indicating that as more crops are lost, there is a reduction in crop biomass and final seed yield (Tables 3 and 6).

DISCUSSION
Early planting (December), intermediate (January) and late (February) planting dates were utilized to evaluate nine tepary bean genotypes over two seasons.The seasons experienced below average rainfall, with 2012 to 2013 recorded as the driest period in Botswana in 13 years since 2001 (Agromet Update, 2012), while 2013 to 2014 was not considered a drought year (Statistics Botswana, 2015).Even though low rainfall were received in Sebele, tepary bean was able to produce some yields (136.45 kg/ha) (Table 2).Our results confirm previous observations by Jiri et al. (2017), that tepary bean can produce moderate yield under drought stress conditions, they recorded 245.9 kg/ha in the semi-arid environment of Zimbabwe.Under good soil moisture tepary bean has a potential to produce more than 2000 kg/ha (Thomas et al., 1983;Bhardwaj et al., 2002).Future work on the best sowing dates for tepary bean especially in wet years is encouraged since this study was conducted for two years in which they were drought and normal year.For subsistence farmers for the crop to be able to produce some yield in the presence of harsh conditions could be more important than the yield potential under favorable conditions (Porch et al., 2012).
Our results also indicated that when planting early in December more yields were realized, especially under normal rainfall in the 2012 to 2013 crop season.However, in relatively good year such as 2013 to 2014 crop season taller plant and higher yields were expressed in the intermediate planting in January (Table 3).Our findings are in agreement with those of other researchers who discovered that planting dates have a significant effect on yield and yield components of the crops.Similar findings were in observed in peanut in semi-arid environment of Turkey by Canavar and Kaynak (2008) and in cowpeas in the South Eastern Nigeria by Ezeaku et al. (2015), who found early planting to be higher yielding than late planted crops.
The genotype effect was not significant for the six characters selected which reveals low genetic diversity among the nine cultivars (Table 5).Limited diversity in tepary bean was also noted by Blair et al. (2012) and Gujaria-Verma et al. (2016) when using simple sequence repeats (SSR) and single-nucleotide polymorphism (SNP) markers, respectively.Their results suggested it could indicate that tepary bean may have arisen from a single domestication event that led to genetic bottle neck which limits diversity within domesticated cultivars.Our tepary bean were relatively small seeded at 7.74 g (Table 4), compared with those of mean 14.5 g observed by Bhardwaj et al. (2002), possibly because the genotypes they used consisted of more variation such as tan coloured seeds, which are bigger than cream coloured seed found in Botswana.However, small seed size were found to have an added advantage of grain filling and producing higher number of seeds, with less abortion which occurs during drought stress as discovered in a close relative in common bean (Rao et al., 2013).
Genotype × environment interaction is varying responses of cultivars for particular characters in different environments.In this study, genotype × planting date interaction was not significant among the six characters which imply that the genotypes are relatively stable (Table 5), and stable genotypes are useful in the risky varied environments.It is therefore important to have stable performing genotypes across environments to realize higher seed yields (Nath et al., 2013).Lack of cultivar × planting dates interaction in yield was reported in bambara groundnut by Makanda et al. (2009), which they associated with the crop adaptability to the four planting dates.In tepary bean (Bhardwaj et al., 2002) found lack of genotype × environment in seed yield and harvest index which indicates that these characters were stable among the planting dates observed.Even though there is vast difference between the dates, year, and date × year interaction, lack of genotype × environment in all the characters indicates that selection of these traits could therefore be conducted either during the early or intermediate planting dates with similar effects.However, Shiringani and Shemilis (2011), noted that multi-location evaluation of genotypes could reveal genotype × environment better compared to when conducting the study in one location like in our case.
Correlation analyses give a measure of relationship between traits and assist in identifying useful traits in increasing yield.Plant height had significant and positive correlation with most characters such as shoot dry weight, seed weight, and seed yield (Table 6).Selection of relatively taller plant would lead to a significant increase in yield and yield related traits of tepary bean.Bashir et al. (2001), when studying correlation of economically important traits in forage crops cowpea, lab and rice bean discovered that plant height have a significant correlation with pod length, 100 seed weight, straw yield and total dry weight, and argued that it is an important character that maybe exploited to improve production of these forage legumes species.

Conclusions
In conclusions, the purpose of this paper was to identify suitable planting date for sowing tepary bean in Botswana.Although specific dates are not mentioned, the study managed to reveal that different planting dates affected the yield and agronomic characters of tepary bean genotypes.Sowing dates for December and January are identified as options for tepary bean farmers in southern part of Botswana.The fact that some yields were realized under severe drought environment identifies tepary bean as a climate smart crop suitable for production in the semi-arid environment of Botswana.Our study also buttresses the need for good meteorological forecast for farmers to make informed decisions before planting.

Table 2 .
Effects of three planting date on the yield and related traits on tepary genotypes combined across two seasons2012-13 and  2013-2014

Table 3 .
Yield and related components of nine tepary beans planted on three planting dates in Sebele, Botswana in 2012-2013 and 2013-2014 seasons.

Seasons 100-Seed Weight (g) Shoot dry weight (g) Seed Yield (kg/ha) Crop lost % Plant population
Mean with the same letter(s) in a column are not significantly different at 5% level of significance according to Duncan's multiple range tests.

Table 4 .
Mean of yield and related components for the three planting dates averaged over nine tepary bean genotypes for two seasons in Sebele, Gaborone.

Table 5 .
Mean square analysis of variance for yield and related components for nine genotypes evaluated at three planting dates in Sebele in 2012-2013 and 2013-2014 crop season.

Table 6 .
Pearson correlation coefficient among the six traits based on nine tepary bean genotypes grown over three planting dates for two seasons2012-13 and 2013-2014 in Sebele, Gaborone.