Advantage of Intercropping Roselle with Common Bean over Sole Cropping in Ethiopia

This experiment was conducted in Ethiopia, during 2017/2018 dry season to determine the roselle-common bean advantage over sole cropping through competition indices. The experiment was carried out in a RCBD with a 2x4 factorial arrangement of two roselle varieties with four planting densities of common bean plus three sole cropping. It was replicated three times having 11 treatments (sole cropping of two roselle varieties (Hibiscus-Sudan and Hibiscus-Jamaica) and a common bean variety (Ibbado) and intercropping of the two roselle varieties with four common bean planting densities combinations (25%, 50%, 75%, and 100%)). Data analysis was done using SAS software version 9.3 and the differences between means were assessed using Duncan`s Multiple Range Test at 5% probability level. Partial land equivalent ratio (LER) analysis showed that there was no significant difference between two roselle varieties while partial LER of common was significantly higher when intercropped with Hibiscus-Sudan (0.51) than with Hibiscus-Jamaica (0.38). Both roselle varieties and common bean planting density did not influence total LER and monetary advantage index (MAI). However, intercropping Hibiscus-Sudan variety resulted higher yield advantage (35%) and monetary advantage (17,946 ETB) and combination of common bean with roselle at 100% planting density resulted 39% yield advantage and 21,410 ETB. Therefore, we recommend combination of Hibiscus-Sudan and 100% planting density of common bean for growers in the same agroecology with the study area for better yield and economic advantage. system and weak research-extension linkage are among the main of roselle in of the in order to the of this crop provide a and foreign identification of a certain cropping system, like intercropping with legume crops common bean, be one of the best options. Therefore, identification and utilization of a suitable cropping system for different roselle varieties can play a key role to increase productivity of the crop and land use efficiency in potential growing areas. Hence, this study was initiated to evaluate productivity of two different roselle varieties intercropped with different planting densities of common bean using competition indices.


Introduction
Intercropping which is growing two or more crops on the same land unit in a particular growing season (Sullivan, 2003) seen as advantageous in the world, particularly in developing world for enhancing yield per unit area to answer food and feed requests of growing population. Despite of some limitations, intercropping has various advantages over sole cropping, such as, it increases yield per unit area (Ali et al., 2015), avoids a risk of total yield loss as when one crop fails the producer may harvest the other crop (Heydari et al., 2016), and improves household income of the producers (Zhang et al., 2015). Another important aspect of intercropping is the reduction of the negative impact of weed, insect, and disease incidences compared to sole cropping (Lithourgidis et al., 2011). The yield and economic advantage of intercropping might be due to efficient use of limited resources like arable land. In Ethiopia, a country where population has been increasing and where the size of arable land was being reduced to due to environmental and urbanization, intercropping is assumed important practice to increase yield per unit area bean (Hirpa, 2014;Adafre, 2016). Studies reported the advantage of intercropping through combinations of different crop species, including medicinal plants in the country (Lulie and Bogale, 2014). Roselle (Hibiscus sabdariffa L.), belonging to the family Malvaceae, is one of the important medicinal plants grown in tropical and subtropical regions mainly for its calyx (Plotto et al., 2004). In addition to its health benefit, roselle has industrial and nutritional values (Mahadevan and Kamboj, 2009). Compatibility and higher intercrop productivity of roselle with legume crops were reported (Heydari et al., 2016;Gendy et al., 2017). However, the performance of roselle in intercropping system has not been well known in Ethiopia, except some inter and intra row spacing determination (Gebremedin, 2015) and cost-benefit studies during direct sowing and transplanting roselle varieties (Girma et al., 2014). Cultivation of roselle is restricted to few growers though it is an important medicinal plant in the country. Lack of appropriate cropping system and weak research-extension linkage are among the main production constraints of roselle in Sidama zone of the Southern Nations, Nationalities, and Peoples Region (SNNPR) of Ethiopia. Hence, in order to expand the production of this crop and provide a quality product to the domestic and foreign markets, identification of a certain cropping system, like intercropping with legume crops such as common bean, can be one of the best options. Therefore, identification and utilization of a suitable cropping system for different roselle varieties can play a key role to increase productivity of the crop and land use efficiency in potential growing areas. Hence, this study was initiated to evaluate productivity of two different roselle varieties intercropped with different planting densities of common bean using competition indices.

MATERIALS AND METHODS
The experiment was conducted during 2017/2018 cropping season at Hawassa, in Southern Ethiopia. This study site is located at 7 0 05' North latitude, 39 0 29' East longitude and at an altitude of 1652 m a.s.l. It receives a bimodal rainfall with short and long rainy seasons from March to April and June to August, respectively and the average annual precipitation ranges from 1000 to 1800 mm. The minimum and maximum mean temperatures of the area are 13 and 27 °C, respectively. The soil textural class of the area is sandy loam with a pH of 7.2 (Dessie & Kleman, 2007). Two roselle varieties (WG-Hibiscus-Sudan and WG-Hibiscus-Jamaica) and a common bean variety (Ibbado) were used for the study.
The treatments were arranged in factorial Randomized Complete Block Design (RCBD) with three replications. The experiment consisted of a total of 11 treatments (2 roselle varieties × 4 planting densities of common bean plus 3 sole plots). The size of each plot was 8.64 m 2 (3.6 m x 2.4m), and pathways between plots and blocks were 1 and 2 m, respectively. Row and plant spacing for roselle were 60cm and 30cm respectively while that of common bean were 40cm and 10cm for sole cropping and for 100% mixture with roselle. Plant spacing used for 75%, 50%, and 25% common bean planting densities were 13cm, 20cm and 40cm respectively. Roselle and common bean seeds were sown on separate rows with roselle sown first and followed by common bean after 30 days of roselle. All necessary management practices were applied as required uniformly.
Roselle calyx was harvested from each net plot area, sun dried and converted to per hectare to record dry calyx yield per hectare. Common bean seed yield was recorded after weighing seeds of the whole central plants (by excluding the border rows and plants) using digital balance (model YP 20002) and after the seeds were adjusted to 10% moisture content using a digital moisture tester (model M-3G) according to the following formula (Hellevang, 1995).
Adjusted yield = [(100-AM)/(100-SM)] X Obtained yield, Where, A M = actual moisture and SM =standard moisture. Intercropping productivity was evaluated using the following competition indices. Land equivalent ratio (LER): The yield advantage of roselle and common bean intercropping was determined by calculating LER, which is the amount of land required under sole cropping to obtain the same yield as produced in the intercrop (Mead and Willey, 1980) as cited by Bantie et al. (2014). When LER measures 1.0 or greater than 1.0, it means that an intercropping is more advantageous than sole cropping in utilizing resources. Hence, the effectiveness of roselle and common bean intercropping for using available resources was evaluated using the following formula. Local market prices of a kg of dry calyx of roselle and a kg of common bean seed were used for monetary advantage determination. The average prices for a kg of a dry calyx of roselle (70 ETB) and common bean (11 ETB) were taken from local markets in Hawassa and Shashamane towns.
intercropped with Hibiscus-Jamaica (Table 2). The lower partial land equivalent ratio, which was less than 0.5, showed the disadvantageous of intercropping, and thus, according to the present finding, common bean intercropping with variety Hibiscus-Jamaica was found to be inappropriate. This could be due to the competitive ability of the roselle variety for light, water and nutrients. On the other hand, the higher partial land equivalent ratio (0.51) of common bean from intercropping with Hibiscus-Sudan variety indicated as yield of common bean was more favored than the case of intercropping with Hibiscus-Jamaica. Furthermore, as common bean planting density increased from 25% to 100%, partial land equivalent ratio of common bean increased from 0.32 to 0.56, which, might be due to better resource utilization (Table 2). A similar increase in partial LER of common bean with increases in planting density has been reported for intercropping common bean with maize (Lulie et al., 2016).

Total land equivalent ratio
The combined (total) land equivalent ratio of roselle and common bean intercropping was not significantly influenced (P > 0.05) by the interaction between main factors, by roselle variety and common bean planting density (Table 1). However, numerically, total LER was higher for intercropping common bean with variety Hibiscus-Sudan (1.35) compared to the result obtained for intercropping common bean with variety Hibiscus-Jamaica (1.30) ( Table 2). This showed that intercropping Hibiscus-Sudan with common bean was more advantageous compared to Hibiscus-Jamaica. Furthermore, as common bean planting density increased from 25% to 100%, there was an increasing trend in total LER from 1.29 to 1.39 (Table 2). This was in agreement with the reports of Lulie et al. (2016) and Adafre (2016) who observed increment in total LER as common bean planting density increased from 25% to 100%. Based on the values of total LER, advantage of intercropping roselle with pigeon pea (Pushpa et al., 2017), maize with faba bean (Rezaei-Chianeh et al., 2011) and cotton with cow pea (Aesim et al., 2008) has been reported. 50.17 NS=not significant; *, **, and *** significant at P≤0.05, P≤0.01 and P≤0.001 probability levels respectively; Rep= Replication; DF = degree of freedom; LER=Land equivalent ratio; MAI=Monetary advantage index.

Monetary advantage index (MAI)
Monetary advantage index of roselle-common bean intercropping was not affected by variety, planting density and their interaction (p > 0.05) (Table 1). However, numerically the highest monetary advantage (17, 946 ETB) was obtained from Hibiscus-Sudan than from Hibiscus-Jamaica (17, 283 ETB) (Table 2). Besides, as planting density increased from 25% to 100%, MAI increased from15, 525 ETB to 21, 410 ETB (Table 2). This could be due to difference in the amount of yield of common bean.