Effect of Soil Water Environment and Simulated Savanna Climate on Growth and Mineral Nutrition in Jatropha curcas L

This study aimed to determine the optimal soil water conditions for J. curcas L. cultivation under the bi-annual rain seasons of the African Savannah climate. Changes in J. curcas L. seedling biomass and mineral nutrient uptake and movement patterns were examined under soil water matric potentials of -2.5, -2.9, -3.4 and -4.0kPa, and two rain seasons separated by a dry spell. The biannual savannah rain conditions were simulated using automated buried-type tensiometer system. During the first rain season (irrigation), -3.4kPa soil water matric potential showed the highest water-use efficiency (2.44 g dry weight L) and biomass yield (57.14g). Seedlings were variously affected by the subsequent dry spell, including total defoliation and stunting. On re-irrigation, seedling recovery (biomass and water use efficiency) was less as compared to the first rain season. Among the nutrients monitored, N, P and K were the most predominantly uptaken, with N showing the closest correlation with seedling growth. Soil matric water potential of -3.4 kPa is the optimum soil water environment for J. curcas cultivation in the savannah conditions; additionally, irrigation during the dry season is essential for a stable harvest.


Introduction
Jatropha curcas L. belongs to the spurge family (Euphorbiaceae) and is an important biodiesel crop cultivated in different parts of America, Africa and Asia (Openshaw, 2000).It is sometimes categorized as a deciduous stem-succulent tree (Maes et al., 2009b), and commonly used for several purposes such as; hedge to protect fields, green manure, oil-soap production, pharmaceuticals and biodiesel feedstock.It has a high potential for socio-economic development in terms of job creation for communities around plantations (Heller, 1996;Kumar & Sharma, 2008).

~ 14 ~
In sub-Saharan Africa, J. curcas has recently been adopted as one of the most suitable crops for biodiesel production (King et al., 2009;Jingura, 2011;Jingura, Matengaifa, Musademba & Musiyiwa, 2011).It can be grown as a single stand on abandoned farmland, or used in marginal/waste land reclamation without competing for land currently used for food production.Moreover, it can adapt to poor soil fertility and drought/arid conditions (Augustus, Jayabalan & Seiler, 2002;Wang et al., 2011).Water requirement for the crop is extremely low and can stand long periods of drought by shading most of its leaves to reduce transpiration loss (Reyadh, 2002).
A recent survey in Sub Saharan Africa however, revealed that farmers continuously irrigate J. curcas in both arid and semi-arid conditions and during both rainy and dry seasons of the year in order to maintain good yields.This is done to mitigate leaf shading which is triggered by water stress, especially during the dry season.The areas under J. curcas cultivation receive an annual rainfall requirement of about 944 mm (Maes, Trabucco, Achten & Muys, 2009a), and therefore require additional water supply during the dry season if good yields are to be obtained.Sustainable J. curcas production in these areas is uncertain unless water supply is supplemented during the course of cultivation.Assessment of different areas for optimal field water conditions might be necessary for sustainable J. curcas cultivation during both rain and dry seasons of the year.Meanwhile, some recent studies reported a variation in the yield and oil component of J. curcas seed as influenced by irrigation water levels (Kheira & Atta, 2009).Another study evaluated J. curcas seedling biomass production and distribution under different irrigation water scheduling under laboratory and field conditions (Behera, Srivastava, Tripathi, Singh & Singh, 2010; Maes, Achten, Reubens & Muys, 2011).
Additionally, Kheira and Atta (2009) estimated the weekly average water requirement for optimal J. curcas yield to be about 6 L. However, so far no studies have focused on optimum soil water potential for J. curcas growth.The aim of this study therefore, was to evaluate J. curcas performance under different soil water potential during both rain and dry seasons.Results of this study would be useful in guiding irrigation scheduling for sustainable J. curcas production in the arid and semi-arid regions of sub-Saharan Africa.

Nursery Cultivation
J. curcas seeds imported from Tanzania were sown in sandy soil in cell molding trays under greenhouse conditions at Tottori University, Japan.The textural composition of sandy soil was 95% sand, 1.3% silt, and 3.7% clay and soil texture was sandy.The wilting point and field capacity of this soil were 0.027 cm 3 cm -3 and 0.08 cm 3 cm -3 (these correspond to matric potentials of -1600 and -5.5 kPa), respectively.The chemical properties of sandy soil are as described by Uzoma et al., (2011).After 6 weeks, J. curcas seedlings were transplanted to lager pots (30cm height, 12.5L) still in a sandy soil, and field capacity and nutrient supply maintained by automated fertilization (nutrients; total N: 1.52, P 2 O 5 : 0.70, K 2 O: 2.36 g plant -1 ) for a period of 62 days.

Experimental Treatments
J. curcas seedlings were cultivated under three different sets of simulated climatic conditions representing the short rainy (40 days), short dry (46 days) and long rainy (158days) seasons of Sub Saharan Africa.Detailed cultivation conditions including irrigation regimes and amount of cumulated irrigation each treatment during the entire experiment are given in Figure 1.
Irrigation was terminated at day 40, for short dry season simulation experiment (d).The treatment permutations in this season were represented as: W 1d , W 2d , W 3d and W 4d , with each pot fitted with a TDR sensor (Campbell Scientific Inc., USA) to measure volumetric soil water content.On day 86, J. curcas seedlings, which had survived the dry period were irrigated under-3.4kPa(W 3 ) again using the same irrigation system as earlier described for the next experimental season; the long re-irrigation (long rainy season).The treatment permutations were set at follows: W 1d3: -3.4kPa, W 2d3: -3.4kPa, W 3d3: -3.4kPa and W 4d3: -3.4kPa.The-3.4kPamatric potential was adopted here as it had showed the highest biomass yield in the first irrigation experiment.All treatments were replicated 5 times.During the entire experiment, air temperature was kept at 16-25 °C and relative air humidity (RAH) at more than 50%.

Sampling during Seedling Growth
Seedling leaf number, along with the fallen leaves were taken on a weekly basis for the entire cultivation period.Fresh biomass weight (including fallen leaves, stems, branches and roots was taken immediately after sample collection at 0, 40, 86 and 244 days.Total leaf area of each plant was recorded using an automatic area meter (AAC-410, HAYASHI DENKO CO., LTD.Japan).The different samples were then dried at 80℃ for 48 h and dry weight recorded.

Plant Analysis
The dried biomass parts were ground into a fine powder prior to further chemical analysis.Total mineral nutrient contents potassium (K) and phosphorus (P) were measured after digestion with a mixture of hydrogen peroxide and concentrated sulfuric acid (99%).Potassium was determined by atomic absorption spectrophotometry on spectrophotometer (Z-2300, Hitachi, Tokyo, Japan).Phosphorus determined by the molybdenum-yellow method using a UV-visible spectrophotometer (U-2001, Hitachi, Tokyo, Japan).Total nitrogen content (N) was measured using an automated CN analyzer (Macro corder, JM1000CN, Yanaco, Kyoto, Japan).

Statistical Analysis
Each experimental data was subjected to statistical analyses using R version 2.12.0 (The R Foundation for Statistical Computing).The differences were significance by ANOVA (P ＜ 0.05), the treatment means were separated using Tukey's multiple-range test.The linear regression analyses were used to check if the residuals showed a trend.

Irrigation
Figure 1 shows the amount of cumulative irrigation water used for J. curcas seedling growth during the three different seasons.The W 1 soil matric potential treatment (-2.5 kPa) required the highest amount at 24.78 L plant -1 , while W 2 , W 3 and W 4 treatments followed with 19.41, 16.94 and 15.09 L plant -1 respectively.During re-irrigation at the end of the cultivation cycle, irrigation water consumption was 42.09 L plant -1 (0.17 L day -1 ) in W 1d3 treatment (-3.4 kPa in re-irrigation period) and 36.72,34.25 and 32.40 L plant -1 for W 2d3 , W 3d3 and W 4d3 treatments respectively.~ 16 ~

Biomass Yield
Aboveground biomass yield of J. curcas seedlings is shown in Figure 2. At day 40, leaf count and leaf area were highest in W 3 treatment at 27.4 ± 1.9 and 2209 ± 113 cm 2 respectively.Leaf number declined immediately after cessation of irrigation (dry season), and on day 77, all leaves in W 4d treatment had been completely shed (volumetric soil water content was 4.95%).Leaves in the W 2d and W 1d treatments defoliated on days 78 and 84 (volumetric soil water content: 7.59% and 11.06%) respectively.Only W 3d treatment retained its leaves through the dry season (volumetric soil water content was 6.67%).Leaves in all treatments began to recover in the middle of re-irrigation.
Seedling height increased steadily under irrigation period with the highest of 69.1 ± 4.9 cm occurring in W 3 treatment (Figure 2b).Rate of height gain declined during the dry period, likely due to defoliation.The trunk diameter kept increasing in the first part of the dry experiment (Figure 2c), though on re-irrigation, trunk diameter remained more or less the same.A similar trend was observed in height.

Dry Weight
Table 1 shows dry weight of the various seedling parts under the different soil water potential and irrigation conditions.As seen earlier, dry weight yield was highest in the W 3 treatment at 57.14 ± 4.44 g, and so was water use efficiency (WUE) at 2.44 ± 0.26 g DW L -1 .On re-irrigation after the dry season, WUE tended to increase with decrease in amount of irrigation water.However, these values were less than those of the first irrigation period.Additionally, total dry weight under reirrigation period was lower than that of initial irrigation period.

Mineral Nutrient Content
The mineral nutrient content N, P and K of the seedlings generally increased with cultivation time, but concentration varied in the different plant parts (Table 2).Nitrogen was highest in all parts under the -3.4 kPa matric water potential treatment (W 3 , W 3d and W 3d3 ) during the entire cultivation cycle.Meanwhile, P content was observed to decrease during re-irrigation; K, uptake was highest and tended to increase with decreasing irrigation.

Biomass Yield and Mineral Nutrient Content
Relationship between the total dry biomass and N, P and K uptake during the entire experiment is shown in Figure 3.The linear regression analyses showed significant differences, especially total dry biomass for N uptake (P ＜ 0.01, R 2 = 0.709).

Optimal Soil Water Environment
Biomass yield, water use efficiency and mineral nutrient uptake were highest under a soil water environment of -3.4 kPa (volumetric soil water content: 10%) during the first rain season.Average water consumption per week of this treatment was 3 L in the short rainy season; this is however, less than the 6 L week -1 reported by Kheira and Atta (2009) throughout the growing season.Under the other soil water environment conditions of -2.5 and -2.9 kPa volumetric soil water content > 13%, the soil aeration was greatly reduced (ventilation of the soil of these treatments was lost) thus the decrease in dry biomass, WUE and mineral nutrient uptake (Table 1, 2).This result is similar to a report by Heller (1996) in which J. curcas grew well on soil of good ventilation and drainage.

Response to Water Stress
Although biomass yield in the dry and re-irrigation experiments did not show significant differences among the treatments (Table 1), leaves in the -3.4 kPa were not completely shed (Figure 2).Moreover, Table 1 shows stem water loss rate was lowest (1.56 ± 0.41 %) here throughout the experiment period.This is in agreement with a study by Maes et al. (2009b) which stipulates the role of the J. curcas succulent stem in the water economy as balancing the small water losses through the leaves during drought.This result showed that proper management of this soil water matric potential enhanced drought tolerance.In the re-irrigation experiment, the cultivation days were 3 times more as compared to the initial irrigation experiment, though biomass yield and WUE of the latter were 1.1 ~ 1.5 and 2.2 ~ 3.6 times as much as those of the former respectively.Moreover, the leaves were completely shed in the dry spell, and it took time to sprout back on reirrigation.Irrigation in the dry season was therefore important to maintain the above ground biomass of the plants.This supports findings by Maes et al. (2009b) that productivity of plantations in arid and semi-arid regions could decline unless additional irrigation was done.
~ 20 ~ During the dry season, it was observed that J. curcas increased mineral nutrient storage in the stem part (Table 2).This behavior is considered as a strategy of the succulent stem to minimize defoliation, and prepare for budding in the next rainy season.

Mineral Nutrient Uptake
Total K uptake and retention was highest (81.5 ~ 99.7% of the amount supplied), followed by N and P at 30.7 ~ 53.3 and 30.4 ~ 69.8%, respectively.Potassium was particularly predominantly high in leaves and stems, thereby making it easy to lose if branches/leaves were to be lost; for instance K content in fallen leaves was about 30.4 mg g -1 dry weight.It is therefore necessary to provide supplementary supply of this nutrient for long-term cultivation of J. curcas in semi-arid regions.
Meanwhile, there is potential for utilization of fallen leaves as organic sources of fertilizer in J. curcas plantations.Dange, Suthar and Reddy (2006) evaluated J. curcas leaves as organic manure, and found them effective.Additionally, J. curcas was shown to respond better to organic manure than mineral fertilizers on degraded soils (Francis, Edinger & Becker, 2005).Seed cake and seed hull of this tree are also good sources of compost (Sharma, Pandey & Lata, 2009;Das et al., 2011); these are rich in essential mineral nutrients N, P and K.
Although K uptake was highest, it is the N uptake that correlated most with seedling growth (Figure 3).This is in agreement with a previous study that showed N to increase total dry biomass, whole plant water storage capacity, total evapotranspiration, irrigation water use efficiency and crop water use efficiency as compared to no N treatment (Yin et al., 2012a; Yang, Li, Zhang & Liu, 2013).Tikkoo, Yadav and Kaushik (2013) had also showed seed yield and seed oil content to increase with amount of N fertilizer applied.Thus we suggested that N is the most important mineral nutrient in J. curcas cultivation.This is however in contrast to another study which suggested that the effect of N fertilization on the accumulation of osmolytes in the plant depended on drought stress level and N fertilization rate, and that applying a high rate of N fertilizer inhibited osmolytes accumulation under soil water stress conditions (Yin et al., 2012b).

Conclusion
Basing on growth, water use efficiency and mineral nutrient uptake results, we conclude that the best soil water environment for J. curcas cultivation is -3.4 kPa (volumetric soil water content: 10%).This soil water condition prevented total leaf-shedding by saving water and mineral nutrients in the stem part; thus, soil water environment enhances drought tolerance.On the other hand, amount of K uptake was highest, however, it is N uptake that correlated most with growth.Therefore, we suggested that N is the most important mineral nutrient in J. curcas cultivation.

Figure 1 .
Figure 1.Cumulative irrigation for the entire experiment

Figure 2 .
Figure 2. Changes in the number of leaves (a), height (b) and trunk diameter (c) of J. curcas seedlings

Table 1 .Figure 3 .
Figure 3. Regression analysis of total matter yield from N, P and K uptake in J. curcas seedlings