Influence of Shea tree (Vitellaria paradoxa) on maize and soybean production

Vitellaria paradoxa provides many benefits to farmers within the Shea belt. However, increased threats to it necessitate its conservation, and one common approach is the practice of agroforestry. A number of studies have shown that Shea tree has influence on crop production, and yet, some of these studies were done using single season experiments or bioassays using mature Shea tree components. In this study, the seasonal influence of young and mature Shea trees on Maize and Soybean yields was investigated using field experiments in Otuke district of northern Uganda, where, Shea tree parklands are dominant and Maize and Soybean are used for food security and income. Our results show that there are differential responses of maize and soybean yield to rainy seasons and physiological variations of Vitellaria paradoxa with age. We find yield reduction for maize more pronounced than yield reduction for soybeans under different Shea age (Mature and Young) for two rainy seasons. We attribute the variance to the differential maize and soybean responses to Vitellaria paradoxa shading and its differential allelopathic inhibition of these crops. We recommend that Soybeans should be preferred to maize when planting under Shea canopy.


Introduction
The Shea tree (Vitellaria paradoxa) is an important parkland tree species indigenous to Africa, specifically occupying the Sudano-sahelian regions stretching from West Africa, across Central Africa to East Africa [1]. In Uganda, the Shea belt that occupies parts of Eastern and Northern Uganda predominantly has Vitellaria paradoxa sub species nilotica [2]. The tree produces nuts that are processed to obtain Shea butter, which are of high economic value [3]. Shea butter has a wide range of uses including in the cosmetic and pharmaceutical industries [4], and it contains important fatty acids including palmitic, stearic, oleic, linoleic and arachidic acids [5].
Farmers in the Shea belt usually collect Shea nuts for domestic consumption and income and Shea butter is becoming an important foreign exchange commodity for countries where it is found [6]. A number of scholars and stakeholders agree that there is need to conserve Vitellaria paradoxa due to its economic potential and threats to it [7]. These threats include a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 Editor: Randall P. Niedz, United States Department of Agriculture, UNITED STATES burning the tree for charcoal, and large scale clearing of the Shea tree to pave way for mechanized agricultural production [1,4,7,8]. Also, increasing population pressure in the Shea belt means that there is no longer room for natural regeneration of the Shea tree [9].
As a result, several conservation approaches are being encouraged by stakeholders including among others, intercropping Shea tree with annual agricultural crops. It is recommended that crops grown under Shea trees are shade tolerant [1]. More so, an increasing number of studies find that Shea tree has inhibitory effects on certain crops [1,[10][11][12]. The magnitude of the inhibition ranges from those that are not statistically significant and do not cause significant differences in yields, to those that are statistically significant leading to significant reduction in yields. The influence of planting season and crop productivity under Shea has not been widely reported. Majority of the studies on inhibitory effects of Shea tree are single season studies such as those of Aleem et al., 2014, [11] or bioassays that involve planting the experimental crops in soils incorporated with extracts of mature Shea tree components such as those of Folarin et al., 2015, [12]. It is widely accepted that variations in seasons have significant influence on crop productivity [1]. Physiological stages of Shea is also expected to have significant influence on crop production. These influences are expected to vary with the crop in question, for instance, Alamu and Aleem, 2014, [10]; Aleem et al., 2014 [11] reported differential responses of cowpea and maize to Shea tree. This study was therefore done to investigate the seasonal variations in the influence of Vitellaria paradoxa on production of Zea mays L. (Maize) and Glycine max (L.) Merrill. (Soybean) and investigate the influence of the physiological stage of the Shea on production of the two annual crops.
The two crops have been chosen for this study for two main reasons. First, both maize and soy beans are very important food security and income generating crops in the Uganda Shea belt. Secondly, these two crops were chosen for the study, due to differences in their physiology which is a big factor in their individual responses to the influence of Vitellaria paradoxa. Maize, for instance is a C4 plant, while soybean is a C3 plant. This means that the crops follow different photosynthetic pathways and are expected to exhibit different responses to the influence of Vitellaria paradoxa on their yields. We thus expect seasonal differences in the influence of Vitellaria paradoxa on both maize and soybeans, and a prominent influence of Vitellaria paradoxa on maize which, as a C4 plant is likely to be more responsive to variation in light intensity than soybean. We also expect different responses of both maize and soybean to different physiological stages of Vitellaria paradoxa, and that the influence of Vitellaria paradoxa on both maize and soybeans would be more pronounced for the mature Shea garden than young Shea garden.

Study design and data collection
Study area. The study was conducted in the first and second rainy (planting) seasons in Opejal parish (N 2031'46.145, E 3306'48.203) located in Okwang sub County, Otuke district, in Northern Uganda. It has a rainfall pattern with two rainy seasons from late March to May and July to November, with a long dry spell stretching from December to early March. The Average annual rainfall for the district varies between 1000 mm-1600 mm. This rainfall is suitable for the production of both maize and soybeans. The mean temperature is between 22˚C-26˚C. However, temperatures may be as high as 40˚C during certain periods of the long dry season.
The natural vegetation is mainly savannah woodland with scattered trees dominated by Shea trees (Vitellaria paradoxa). Other prominent tree species includes Terminalia, Cambretum spp, Ficus spp, Accacia spp and Phoenixma linareclinata. Otuke district is generally flat or gently undulating with an altitude between 900 meters to 1500 meters above sea level, although much of the district lies above 1020 meters above sea level. The largest part of the district comprises of remnants of lowland surface, and is generally well drained except for some peripheral areas, that are occupied by poorly drained swamps. The district lies mostly in the Aswa water catchment that drains her wetlands in the south and west into river Aswa and Moroto respectively.
Experimental design and data collection. This is to confirm that this experiment was done on private land and that permission was sought and granted from the land owner. This study also did not involve any endangered or protected species. The experiments involved one factor with three levels of Mature Shea tree garden, Young Shea tree garden and a control garden that had no Shea tree. For purposes of this study, mature Shea trees were trees that were already producing nuts, while the young Shea trees had never produced nuts. The experiments involved planting both maize and soybeans independently in replicates of (1) four mature Shea gardens, (2) four young Shea gardens and (3) four Control gardens. Each Shea and control garden measured 10 x 15 meters. These were divided into sub-plots of 2.5 x 2.5 meters. In each garden, a total of four sub-plots were planted with maize and another four sub-plots planted with soybeans in an alternating design (Fig 1). The planted plots were alternated with rest plots. This gives a total of 16 sub-plots for each level of factor under maize and another 16 sub-plots for each level of factor under soybeans respectively. In season two, the alternating rest plots were planted alternately with maize and soybeans.
The maize variety Longe 10H and soybean variety Maksoy 3N were used in this experiment. These varieties were chosen due to their drought resistance and high yielding potentials and are varieties that are recommended for the region. The agronomic practices undertaken for both maize and soybeans followed the standard practices that farmers in Otuke district follow. No inorganic fertilizers were used for all the two seasons. The crops were weeded twice as required. Maize and soybean yields after harvest, threshing and drying were weighed and recorded in grams for further analysis.

Data analysis
The maize and soybeans yield data from both planting seasons were analyzed using Microsoft office excel 10 and SPSS version 20. The yield data was reported in kilograms per hectare. The data was checked for normality using the Shapiro-Wilk test for normal data. This was applied to each level of the factor. The null hypotheses that the data was normal could not be rejected. The data was then subjected to one way Analysis of Variance (ANOVA) and the Duncan Multiple Range test (DMRT) to test the difference in yields at 5% level of significance. The analysis did not look at interactions between season and Shea tree level of shading (factor). This analysis compared differences between maize and soybean yields in season 1 and season 2, under young and mature Shea trees. A yield decline index was also constructed to compare the yield difference between maize and soybean. The index was constructed by taking the yield from the control experiment as the base and comparing it with the yield from Young and Mature Shea as shown in the equation below.

Yield decline Index ¼
Mean yield under shea Mean yield under control x 100

Experimental maize and soybean yields
The results show variations in yield of maize and soybeans under the different ages of Shea (Fig 2). Analysis of Variance found significant difference in mean yields for maize and soybean in the two seasons ( Table 1, Table 2). Further analysis of data (S1 Table) shows significant differences in mean yields of maize under all ages of Shea, while there was no significant difference of soybean yield under mature and young Shea trees. The mean soybean yields were however significantly different between those under control and under Shea trees, and the results for season one was consistent with those of season two. However, there seems to be seasonal variations in yields (Fig 2).

Comparing maize and soybean yield response to Shea trees
Comparison of maize and soybean yields from the field experiments found that, overall, intercropping with mature Shea tree had the highest yield reduction for both maize and soybeans as compared to intercropping with young Shea. This reduction was more pronounced in the maize than in the soybeans, for instance, the average maize yield in season one for the mature Shea tree garden was only 23% of the average maize yield from the control garden while the average soybean yield in season one for the mature Shea tree garden was 42% of the average soybean yield from the control garden (Table 3). Similar results are seen for both maize and soybeans in season two.

Comparing seasonal maize and soybean yield under Mature/Young Shea and control
Analysis of effect of season on the influence of Vitellaria paradoxa provides evidence of differential responses of maize and soybeans to Shea age. Specifically, maize yield under Mature Shea and Control gardens and soybean yield under mature Shea were significantly different for the two planting seasons (Table 4).

Discussion
Agriculture is the main source of livelihood in Otuke district with crops cultivated including; rice, groundnuts, sesame, soybeans, sorghum, beans, millets, pigeon peas and maize. This study investigates the influence of seasonal and physiological stages of Vitellaria paradoxa on maize and soybean yields in Otuke district in northern Uganda. We find yield differences between the treatments, specifically; maize yield significantly different for all the treatments for the two seasons, while soybean yields not significantly different for mature Shea and young Shea treatments. We attribute this yield differences between the treatments to a number of factors; firstly, the response of maize and soybeans to different levels of shading as reported by [13]. The difference between maize and soybean yield can also be attributed to the physiological differences between the two crops. For instance, Maize which is a C4 plant is very responsive to light intensity and temperature [14]. A C4 plant's photosynthetic pathway uses different enzymes from those used by C3 plants. C4 plants are often called tropical or warm season plants. They reduce carbon dioxide captured photosynthesis to useable component by first converting carbon dioxide to oxaloacetate, which is a 4-carbon acid [15]. Boffa, 2015 [1] indicated that C3 plants are less affected by Shea tree shading than C4 plants. This explains why our results show that the effect of Vitellaria paradoxa on Maize (a C4 plant) is more pronounced than in soybeans (a C3 plant). We also attribute the yield decline due to Vitellaria paradoxa, to the presence of phytochemicals referred to as allelo chemical that have inhibitory effects on the growth and production of other plants. For instance, Aleem et al., 2014 [11] and Alamu and Aleem, 2014 [10] in separate studies of cowpea and maize respectively reported that Vitellaria paradoxa had allelopathic effects on Cowpea and Maize. However, the influence of Vitellaria paradoxa on maize was more pronounced than in cowpea.
We also find that there were seasonal variations in the influence of Vitellaria paradoxa on maize and soybeans. Maize and soybeans yield in the second planting season were relatively higher than yields from the first season of planting. However, there was no significant influence of season on both maize and soybean yields under young Shea and soybean under the control. Intercropping with Mature Shea showed significant (p<0.05) yield difference for both maize and soybean. This can be attributed to the seasonal variation in the chemical composition of Vitellaria paradoxa. Boffa, 2015 [1] reported that Vitellaria paradoxa exhibits  [16] reported that Shea population structure and regeneration status depends partly on land management regimes. This means Shea tree chemical components that have inhibitory effects on growth and production of Maize and Soybeans vary across time and place. The differences in yields can therefore be attributed to differences in seasonal water availability.

Limitations of the study
This study was carried out in two planting seasons. The results are therefore limited to situations similar to those of these two seasons. In case of very high weather variability to levels significantly different from these seasons, these results may not be applicable.

Conclusion
In this study, we investigated the influence of Vitellaria paradoxa on the yield of Z. mays and G. Max. Specifically, we expected to find variances in the seasonal influences of Vitellaria paradoxa, and influence of mature Shea and young Shea on these two crops. This would be because, seasonal differences would lead to differences in crop responses under Vitellaria paradoxa, and that differences in the physiology of both maize and soybeans would lead to varying responses of these crops to Vitellaria paradoxa under different physiological conditions and season. In line with our expectations, the study found that V. paradoxa influences maize and soybeans differently. Specifically, the influence was more pronounced in maize than in soybeans in both seasons. This can be attributed to the response of maize and soybean to shading, and phytotoxic inhibition by other plants. The response of these crops can be attributed to their physiology. We also find that mature Shea tree exhibited seasonal influence on both maize and soybean. This was attributed to the variance in the chemical composition of V. paradoxa especially with respect to allelo chemicals.
Our study provides a basis to recommend preference to planting of soy beans under Shea tree canopy as opposed to planting maize. This will help in furthering the effort to conserve the Shea tree (V. paradoxa), given its profound economic value.
Supporting information S1