Selection of Drought-Tolerant Pasture Species under Varying Soil and Moisture Conditions in Bauchi State, Nigeria

Livestock production is an agricultural system that serves as humanity’s protein and calorie source. Its production is the main economic stay for some people and a complementary source for others. However, land misappropriation and draught constrain the sustainable production of pasture for feeding livestock. Further aggravated by farmer/herder clashes and wetlands extinction. The need for an experiment for the selection of the best pasture species in the Sudan Savannah region that can thrive well under diverse soil textures and moisture status becomes imperative. This trial was conducted in the screen house of Babcock University, objectively to test the performances of Sorghum almum, Andropogon gayanus, Brachiaria mulato and Centrosema pascuorum under Sand, Sandy Clay Loam, and Sandy Loam textures and four water regimes: 100%, 75%, 50%, and 25%. Standard agronomic recommendations were practised throughout the experiment. Data collected included plant height, fresh and dry shoot and root weights, number of leaves, and leaf length. Data generated were analyzed using ANOVA. According to the results, Sandy Loam soil (Soil type from Gamawa) was the best for supporting all the pasture species, followed by Sandy Clay Loam (Soil type from Zaki). S. almum outperformed other pasture species significantly irrespective of soil textural type and water stress level treatments, followed by B. mulato . For water levels; 100% and 75% had the most promising biomass outcome. Based on the results, a 75% water regime which represents 25% deficit of the actual crop water requirement is recommended for the production of the tested pastures in the area.


Introduction
Rearing and grazing livestock on grasslands has been a global practice in the history of animal production (FAO, 2009). In Nigeria, the practice of pastoral, seminomadic, and controlled livestock management systems are well pronounced (Awoyomi et al., 2022). A year-round supply of herbages of high dry matter and optimal calories has been a major constraint for pasture management (Gill and Tuteja, 2010). Short rainy season and frequent dry spells pose major constraints for pasture production in the Northern part of Nigeria (Idris et al., 2019b). This is a result of low annual precipitation and poor soil properties which dictate the vegetation characteristics of the area (Idris et al., 2019a). Low precipitation facilitates drought and subsequently desertification (Nkechi et al., 2016). Loss of wetlands and floral diversity has been due to low precipitation and rising temperature (Abdulkadir et al., 2017). Applying water through rain or irrigation below the actual crop water requirement could result in water stress (Ibrahim et al., 2020), and this could negatively affect plant growth and development as well as the economic yield, depending on the crop tolerance to water stress and the development stage at which the stress occurs (Culman et al., 2019). Soil water deficit poses threat to crop biomass and pasture species production critically required for the optimum management of livestock, particularly in the Sudan savannah zone of Nigeria (Idris et al., 2019b). Numerous negative effects may be manifested through tempered physiological and physical processes which in turn affect crop phenology and quality and quantity of biomass and yield (Zwicke et al., 2015). Gill and Tuteja, (2010) reported that water stress conditions resulted in death and plasmolysis of meristematic cells and lipid peroxidation. Wallace et al., (2016) complimented that water stress inhibits the reaction of oxygen, thereby hampering photosynthetic functions to cause poor pasture performances. Various physiological mechanisms that enhance crop tolerance to soil water stress had been reported (Khaleghi et al., 2019). Per et al., (2017) discovered that, when certain soluble osmolytes within plant tissues accumulate, they enable plant cell membranes to become rigid and resistant to water stress. Studies had shown that positive adjustment of enzyme functionalities within the plant tissues can facilitate their level of water stress tolerance (Nikoleta-Kleio et al., 2020). Various methods and tools, including telemetric systems, evaporation pan and Theta Probe are being employed for the determination of crop water requirement and tolerance through precise investigation of soil water content at various depths (De Lara et al., 2018). However, most water stress studies considered only environmental and soil moisture characteristics while paying less emphasis on the physiological processes and genetic traits of plants (Sarker et al., 2019). Many works on water requirement and deficit irrigation have been conducted on staple crops, however, very few researches captured pasture to identify and select plant species that are tolerant to water deficit conditions for a sustained livestock feed supply. The objective of this research is to identify and select pasture species that are resilient to water stress in the desert encroached area of Bauchi State, Nigeria under Sand, Sandy Clay Loam, and Sandy Loam textures.

Study area, Soil Sampling and Seed Sourcing
Babcock University is located in Ikenna local government, Ogun State. The area has distinct dry and wet seasons, with an average temperature of 27.1°C; relative humidity of 74.4%; and rainfall amount of 185.4 mm per annum. Twenty-one soil samples were randomly collected each from; Azare (lat. 11°40'42'' N, long. 10°11'31'' E); Zaki (lat. 12°17'57'' N, long. 10°18'32'' E), and Gamawa (lat. 12°8'14'' N, long. 10°32'19'' E) areas of Bauchi State. The samples were collected during the dry season at a depth of 0 -20 cm, bulked to form composite samples for each location. Equal weight (5.0 kg) of the soils were accordingly transferred into 4-litre capacity experiment pots bearing a dimension of 5.5 cm height and 18.6 cm diameter equivalent to total area of 0.0865m 2 . The pasture seeds were sourced from the National Animal Production Research Institute, Ahmadu Bello University, Shika, Zaria. .

Soil physicochemical analyses
Physicochemical properties of the soil determined include; soil particle size, organic carbon, soil pH, available phosphorus, exchangeable bases, exchangeable acidity and total nitrogen according to Okalebo et al. (2012).

Agronomic Management and Experimental Design
The experiment was arranged in a Completely Randomized Design (CRD) with four pasture species, three soil textures and four water stress levels as the superimposed treatments. Each treatment combination was replicated three times. The pasture seeds were sown by broadcasting using standard rates. The rates were 80; 10, 35, and 30 kg/ha downscaled as 0.692 g/pot, 0.0865 g/pot, 0.303 g/pot, and 0.258 g/pot for A. gayanus, S. almum, B. mulato and C. pascuorum respectively. After germination, all the pots were watered adequately for one week before subjecting them to four varied water stress levels. The water stress levels; were 100%, 75%, 50% and 25% actual crop evapotranspiration (ETa) which is given in equation 1. Available soil moisture status was determined in-situ using ThetaProbe -HH2 moisture meter, Delta -T Device model for computing irrigation supply schedule:  (NPK 15:15:15) was basally applied at a standard rate of 150 kg/ha (1.29g/pot equivalent) to each of the experimental pots except for C. pascuorum which is a legume and was fertilized with 100 kg/ha downscaled as 0.86g/pot equivalence. Data collection was achieved for plant height, leaf length and the number of leaves at the end of each week from week 1 to week 5 using the meter rule and manual counting respectively. Fresh shoot and root weights were recorded using an electronic weighing scale during the fourth and fifth week after planting. Dry shoot and root weights were calculated after oven drying at a temperature of 65 ₒ to a constant weight.

Statistical analyses
Data collected were subjected to Analysis of Variance using Statistical Analysis Systems (SAS). Means with statistical differences were separated using Duncan Multiple Range Test at P ≤ 0.05 significance level. Table 2 shows the effect of soil texture, pasture species and water stress level on plant height (PLH). Mean values of PLH across the five weeks for soil textural type factor ranged from 32.02 to 61.46 cm. The lowest was recorded in the first week and the highest in the fifth week. The PLH for pasture species ranged from 23.25 to 70.03 cm. Concerning the water stress level, it ranged from 31.53 to 71.23 cm for all the weeks. The first order interaction i.e., soil texture (ST) × pasture species (PS) and ST × water stress (WS), PLH was significant in the second, third, and fifth week. For soil type, there was no significant difference in PLH in the first and fourth week. However, in the second week, the PLH under Sandy Clay Loam was significantly higher compared to that under Sand and Sandy loam soil types. Sandy Clay Loam and Sandy loam soil types were not significantly different from each other. Nevertheless, PLH in Sandy loam showed a highly significant difference compared to Sand and Sandy Clay Loam soil textures which were also not significantly different from each other in the fifth week. For the effect of pasture species on PLH, Sorghum almum (S. almum) recorded significantly higher values across the weeks than other species. Contrarily, Centrosema pascuorum (C. pascuorum) performed significantly lower for PLH parameter across the weeks. For the water stress factor, 100% (WS4) which was the control caused significantly higher PLH across all the weeks compared to the other levels. Also, the water stress level at 25% (WS3) led to a significant increase in PLH compared to other deficit levels for all the weeks. This study revealed that PLH was higher under SCL compared to other soil textures which could be due to the high level of nitrogen observed in the SCL, which is in affirmation with Malik et al. (2014), who reported nitrogen's positive effect on rice PLH. It was also observed that S. almum and Brachiaria mulato (B. mulato) maintained a higher PLH across the weeks despite water stress levels. This corroborated the finding of Schneider et al., (2018) who observed an increase in the activity of pasture plants under water stress, due to their ability to maintain normal metabolic processes.  Table 3 shows the effect of soil textural type, pasture species and water stress level on the number of leaves (NL). For ST treatment, the NL increased from the first week (max 4.77) to the second week (max. 5.48) and shoot up in the third week (max. 6.21) before it progressively declined in the fourth week (max. 5.65). This means that it followed a sinusoidal pattern from the first week to the fifth week. This trend was not different for PS and WS factors. The highest NL for WS treatment was observed under 0% water stress level (WS4) having obtained 5.20 leaf count in week 1 and 7.53 in week 3. Contrarily 75% (WS1) obtained 3.94 in week 1 and 4.75 in week 3. All interactions showed a significant difference except in ST x WS week 1; PS x WS week 2 and 5 and ST x PS x WS week 5. From the results, there wasn't any significant difference observed in NL from week 1 to 4 based on ST variation. However, NL was observed to be significantly lower in SCL compared to SL in week 5. The number of leaves was significantly higher in C. pascuorum across the weeks studied compared to other pastures. The water stress level at 0% influenced higher NL across all the weeks compared to other stress levels. It was deduced that a 25% deficit also significantly influenced NL compared to other water stress levels. A significant reduction in the NL at 25% water stress observed may not be unconnected with the water stress adoption mechanism by the pastures. Lambers et al. (2008) stated that plants respond to water deficit through various mechanisms including shading of leaves, early maturing, and development of fewer leaves. The outcome is in harmony with Pandit et al. (2016), who reported low NL due to water deficiency. * * * * ns WS1 = 75% water stress, WS2 = 50% water stress, WS3 = 25% water stress, WS4 = No water stress, * = significance at 5% LOS, SE = Standard error. 3 Different letters indicate significant differences among treatment means with the same column at P < 0.05 probability level, ns = no significant difference at P < 0.05 probability level. Table 4 shows the effect of soil textural type, pasture species and different water stress levels on leaf length (LL), fresh shoot weight (FSW) and fresh root weight (FRW). The LL depicted an abrupt trend from week 4 to week 5 across the various treatments.

Results and discussion
For soil textures, LL in week 5, (34.92 -35.63 cm) was numerically but inconsistently higher than in week 4 (32.11 -36.13 cm). For PS factor however, the difference was obvious as B. mulato had 40.97 and 48.28 cm LL in fourth and fifth week respectively. In a sharp contrast C. pascuorum had 9.92 and 8.67 in week 4 and 5 respectively. In the same table, it can clearly be observed that the fresh shoot weight (FSW) and fresh root weight (FRW) in week 4 were higher than the dry shoot and dry root weights. This is expected due to water content contribution to their gross fresh weights. In week 4, the FSW and FRW ranged from 0.60 to 2.03 g/plant. In weeks 4, LL had values not significantly different from each other in Sand and SL soil types except for SCL. Higher values of FSW were observed in SL and SCL and are significantly higher than that of Sand textural class in week 4. Also, SCL influenced high FRW in week 4 compared to other soil types. In terms of dry biomass, SCL and SL influenced higher dry shoot weight compared to Sand in week 4 (Table  5). Dry shoot weights (DSW) and dry root weight (DRW) were consistently higher under SCL soil. This may be due to the high level of total nitrogen (N), available phosphorus (Av. P) and exchangeable Potassium (K) of this location which could have positively impacted on the DSW and DRW. This is in agreement with previous results, which reported that N, Av. P, and K significantly influence shoot and root development (Lasheen et al., 2021, Tshewang et al., 2020and Song et al., 2010. The pasture Sorghum almum (SA) showed significantly higher LL at week 4 compared to other pasture species. However, at week 5 higher mean values were observed in S. almum and B. mulato. Fresh shoot weight was statistically higher in S. almum compared to other pasture species at weeks 4.
At week 5, B. mulato showed significantly higher FSW compared to other pastures, however, S. almum still maintained a higher value with respect to FSW. The parameter of FRW in week 4 was observed to be significantly higher in S. almum compared to the others. The S. almum also had a higher value of FRW compared to other pasture species in week 5. Water stress level at 25% significantly influenced LL, FSW and FRW compared to other water stress levels except for the control (WS4). ns ns * * ns * WS1 = 75% water stress, WS2 = 50% water stress, WS3 = 25% water stress, WS4 = No water stress, * = significance at 5% LOS, ns = no significant difference, SE = Standard error. Different letters indicate significant differences among treatment means within the same column at P < 0.05 probability level. As shown in Table 6 A. gayanus interacting with 0% water stress level gave higher mean values for PLH at week 2 and 3 and significantly higher values in weeks 4 and 5. The same interaction also influenced high value of NL at week 1 and a significantly higher NL in week 3 and 4. The A. gayanus interacting with 25% deficit influenced higher values for PLH in week 2, 3 and 4. Centrosema pascuorum (C. pascuorum) interacting with a 25% deficit gave higher values of PLH than the other excluding the control in week 3. However, PLH observed in S. almum pasture showed higher yield compared to others across the weeks and NL was also positively influenced relative to others. WS1 = 75% water stress, WS2 = 50% water stress, WS3 = 25% water stress, WS4 = No water stress, SE = Standard error. Different letters indicate significant differences among treatment means within the same column at P < 0.05 probability level. Table 7 shows the interaction between pasture species and water stress level on the length of leaves (LL), fresh shoot weight (FSW), fresh root weight (FRW), and dry shoot weight (DSW) in week 4 and 5. Except for the control (0% water stress level), a 25% deficit influenced positive yield. However, B. mulato versus 25% deficit facilitated significantly higher values of LL index at week 4 and 5, and significantly increased FSW and FRW in week 4 compared to other water stress levels. A similar trend was observed when other pasture species interacted with water stress level at 25%. Nevertheless, B. mulato yield was the most positively influenced compared to other pastures. In general, at 25% water stress, all parameters evaluated were not severely affected compared to the severity observed at 75% deficit. This result confirmed previous work which discovered that moderate water deficit stress does not severely affect plants' performance. DaCosta and Huang (2007), and Santos et al., (2009) opined that pasture plants naturally adjust to damages caused by moderate water stress easily, while severe conditions affect them adversely.

Conclusion
Based on the results and discussion presented, this study inferred that Sorghum almum (S. almum) was the best-performing pasture species with respect to plant height, dry shoot and dry root weights. The number of leaves was highest in Centrosema pascuorum. Zero and 25% water stress levels on dry root weight in the week 4 had similar performance. Sandy loam had outperformed other soil textural types. Therefore, Sandy Loam soil is best for producing the test pasture species. Alternatively, Sandy Clay Loam can be suitably used since it has optimally supported good performance. Inference can be drawn that adding full water requirement, that is 0% to 25% water stress levels gave the best output, thus recommended for massive propagation of pasture species in the study areas.

Conflict of Interest
Regarding the publication of this manuscript, the authors declare that, there is no conflict of interest whatsoever.