Evaluation of biomass yield of selected fodder species in relation to frequency of harvest alongside defoliation heights in Western Kenya

Napier grass (Pennisetum purpureum Schumach), the most preferred fodder species for dairy production in East and Central Africa, is under threat from stunt disease caused by Candidatus Phytoplasma oryzae (Ns-phytoplasma). The disease reduces forage yield by 40 to 90%. Two alternative fodder grasses, Guinea grass (Panicum maximum Jacq) and Guatemala grass (Tripsacum laxum Scrib and Merr) and a new stunt disease tolerant Napier cv Ouma 3 were studied to determine their biomass production potential when subjected to three intervals harvest (4, 8, 12 weeks) alongside three basal heights of defoliations (5, 10 and 15 cm). The study was conducted at KARI, Kakamega (high rainfall zone) and KARI, Alupe (low rainfall zone) in Kenya. A split-split plot design with 3×3×3 factorial treatment arrangements was used. A 4-weekly interval of harvesting alongside defoliation heights of 10 and 15 cm significantly increased biomass yield in Napier cv Ouma (38.47 and 33.90 T/ha/year respectively) compared to Guatemala grass and Panicum maximum which yielded 23.3 and 27.4 T/ha/year respectively. There is appositive correlation between yield and morphological parameters with canopy diameter, plant height and number of leaves highly correlating in biomass yield.


INTRODUCTION
In Sub-Saharan Africa improved grasses and legumes have been recommended for livestock production due to their high dry matter yield as well as nutritive value (Onyeonagu and Asiegbu, 2013).Napier grass (Pennisetum purpureum Schumach) is the most preferred fodder species for dairy production in East and Central Africa.It can provide a continual supply of green forage throughout the year and it fits in intensive small-scale farming.It is the dominant grass in zero-grazing systems and can out-yield many other grasses such as Guinea grass (Panicum maximum) and Rhodes grass (Chloris gayana) (Orodho, 2006).It can withstand repeated cutting, and four to six cuts is able to yield 50 to 150 tons per hectare per year (Orodho, 2006).It is the main feed for dairy cows supplemented by crop residues in western Kenya (ICRAF, 1997).However, the threat of Napier Stunt Disease against Napier grass in western Kenya inspired the development a Napier cultivars that is tolerant to the disease hence the need to understand its physiological response to defoliation stress and biomass productivity compared with selected alternative fodder species.
The selected alternative fodder species have been found to be tolerant to Stunt disease (Wamalwa, 2013) but their adoption in the dairy farming system of Western Kenya is little understood, possibly because of greater emphasis previously placed on Napier grass.Andropogon gayanus, chloris gayana, Sorghum almum, Panicum maximum and Guatemala tripsacum are viewed as potential alternative fodder grasses to Napier grass in the dairy industry in Western Kenya (Orodho, 2006).Tripsacum laxum and Panicum maximum have the potential of producing up to 18 to 22 tons/ha/year and 9 to 10 t/ha/year respectively with protein concentration of 7 to 10% (Sebastian et al., 2013).
Harvesting of forages without due consideration of frequency of harvest and defoliation height may affect the re-growth of the remaining stubble residues.Studies have shown variation in species tolerance to frequency of harvest and intensity of defoliation arising from differences in growth habits and root systems (Onyeonagu and Asiegbu, 2013) Scanty information exist in Western Kenya regarding the relation between biomass yield, frequency of harvest and defoliation height for Guatemala and Panicum species (Orodho, 2006) as well as newly identified stunt disease tolerant Napier cv Ouma.
Furthermore, knowledge on stage of harvest as well as frequency of harvest is important because the quality and quantity of forage species for animal feed is partially contributed by these factors (Ball et al., 2001).The objective of the study was to determine biomass production of the alternative fodder species and Napier cv Ouma in relation to frequencies of harvest alongside defoliation heights in western Kenya.

MATERIALS AND METHODS
The study was conducted at KALRO Kakamega and Alupe sites of western region of Kenya (Figure 1).Kakamega and Alupe sites represent the Low Midlands 2 (LM 2) and Low Midlands 3 (LM 3) respectively with an altitude of approximately 1430 m and 1330 asl respectively (Jaetzold et al., 2005).The amount of rainfall received during the study period in 2012 LR and 2012 SR was 548 Mean annual minimum temperature was 13°C while corresponding maximum temperature was 29.6°C at Kakamega site and Alupe site the minimum was 15.3°C and maximum 31.4°C.The general soil texture for Kakamega site was sandy loam while at Alupe site was sandy clay loam soil.The soil pH for Kakamega and Alupe sites were 5.1 and 4.94, respectively which were classified as slightly acidic based on the critical value levels (Okalebo et al., 2002).The organic carbon levels of soils at Kakamega and Alupe sites were 3.4 and 2.5% respectively.This implies that soil at Kakamega site is classified as highly organic carbon and Alupe site as moderate (Okalebo et al., 2002).Nitrogen content in the soil at Kakamega site was 0.2% while that of Alupe was 0.12%.The two sites contain moderate levels of Nitrogen content in the soil as classified by Okalebo et al. (2002).
The experimental design was randomized complete block design arranged in a split-split with three replications (blocks).This was a three-factor experiment where three levels of precision were required for the various effects (Gomez and Gomez, 1984).The main-plot factor was the frequency of harvest and the sub-plot factor was defoliation height while the sub-sub-plot factor was the species.There were 27 treatments replicated three times, which consisted of three frequencies of harvest (4, 8 and 12 -weekly interval of harvest), three defoliation heights (5, 10 and 15 cm) and three fodder species (Panicum maximum Schum), Guatemala tripsacum and Napier grass cv Ouma).
Three rooted-splits of Panicum maximum and Guatemala grass and one rooted split of Napier cv Ouma (Ramadhan et al., 2012) were uprooted from the parent field and planted in prepared holes of 15 cm depth on a plot size of 2×2 m at 0.5 to 0.5 m spacing for Panicum and Guatemala grass while Napier grass was 1 to 1 m to maintain optimum plant density (Muia et al., 1999).Fertilizer was applied at the rate of 60 kg/ha of N and 9.3% of P top-traced with 100 kg of CAN immediately after defoliation to minimize the local soil nutrients influence on the performance of the fodder species.Biomass yield was determined by hand clipping three and nine internal stools of Napier grass and Panicum and Guatemala respectively at their respective defoliation heights and frequency harvest.At 4-weekly interval of harvest , three defoliation heights of 5, 10 and 15 cm for each forage grass were clipped manually using secateurs and field weights measured using electronic balance.The clipped samples were chopped into small pieces (about 3 cm lengths) weighing about 500 g.The samples were oven dried at 60°C for 48 h to determine percentage dry matter.The same procedure was conducted for samples harvested at 8-weekly and 12 weekly intervals.Statistical analysis was done using the statistical analysis system (SAS).The data were subjected to analysis of variance (ANOVA) and the means were separated using Duncan's Multiple Range Test (DMRT) at the 5% level of significance.Dry matter yield was calculated using the following formula:

FINDINGS AND RECOMMENDATIONS
Significant interactions (p≤0.05) between frequency of harvest, defoliation heights and fodder species were observed in cumulative biomass yield at Kakamega and Alupe sites.Napier grass cv Ouma significantly (p≤0.05)out-yielded Guatemala and Panicum when harvested at a 4-weekly interval alongside defoliation height of 10 cm above the ground followed by the same species harvested at 4-weekly interval but defoliated at 15 cm stubble height (34 T/ha/year) at both Kakamega and Alupe sites (38.5 and 34 t/ha/year respectively) (Table 1  and 2).The result is in agreement with results of  Muyekho et al. (2003), where dry matter yield of recommended Napier grass varieties (Bana grass, French Cameroon and Clone out-yielded Guatemala grass when harvested at 4-weekly interval along with defoliation height of 10 cm above the ground This could be explained by cumulative biomass yield after several repeated harvests (12 times) for the 4-weekly intervals compared to 8-weeklyinterval which was repeatedly harvested 6 times and 12-weekly intervals which had three repeated harvests within the experimental period of two years.
The trend of increased dry matter yield with interval of harvest in this study is in agreement with the findings of Saddul et al. (2004) who obtained increased biomass yield with increased intervals of harvest.Furthermore, Hsu et al. (2005) established that forage yield and quality of Nile grass (Acroceras macrum Stapf) and Pangola grass (Digitaria eriantha Steud) increased yield with increased frequencies of cutting.Conversely extended interval of harvest similar to 12-weekly and 8-weekly interval yielded low mainly because there was less regrowth and tillering in our present study.
This result is in agreement of Hoglind et al. (2005) who reported that more frequent harvest promoted re-growth and tillering in fodder grasses than less frequent harvesting.Ruiz et al. (2012) attributed to less biomass yield as a result of longer intervals between the harvests to the aging of the leaves and a great number of them fall down due to senescence.Njarui et al. (2008) found Napier grass yielding more than P. maximum due differences tillering ability during re-growth after defoliation.
Stichler and Bade (2002) found the stage of plant growth important in determining the biomass yield.In his research on frequency and cutting heights on biomass production of Tithonia diversifolia, Hsu et al. (2005) reported that plants cut at 5 cm and more frequent performed least in terms of biomass yield and this was associated with leafing and tillering ability since plants cut at this height and frequency have fewer nutrient reserves in the stems for the next re-growth.
The findings in this current study suggest Panicum has a potential of being an alternative fodder to Napier grass cv Ouma in western Kenya when harvested at 4-weekly interval alongside defoliation basal height of 10 cm.Apart from high yielding when forage is harvested at the frequency of 4-weeks, Stichler and Bade (2002) observed that the forage is more palatable and large quantities is used by grazing animals.Although Napier grass cultivar Ouma is more tolerant to stunt disease (Khan personal communication), this study has shown that the biomass yield is comparable with other Napier species that are susceptible to stunt and smut diseases (Wamalwa, 2013;Muyekho et al., 2006).However, at Alupe site Napier grass harvested at 4-weekly interval along with defoliation height of 10 cm above the ground yielded the highest biomass (34.98 t/ha/year).This was below the yield observed at Kakamega site.This could be attributed to variation in climatic conditions between Alupe and Kakamega.Saddul et al. (2004) reported similar findings of forage yields variations between locations due to differences in climatic patterns.
In the current study, Kakamega site experienced relatively high cumulative rainfall (1294.99mm/year) compared to 1175 mm/year in Alupe and therefore may have stimulated vigorous tillering ability, leaf numbers, wider canopy formation and stool diameter as is demonstrated in the highly positive correlation between these parameters and biomass yield (Tables 3,4 and 5).Breshears and Bainers (1999) reported similar findings that biomass yield of a plant progresses with available soil moisture and diminishes with the fall of moisture below field capacity and ceases at the permanent wilting percentage.Cameron (2001) further reported that soil   water stress may lead to limited leaf area development and consequently reduce dry matter yield.

Conclusion
The study has demonstrated that Napier cultivar cv Ouma yielded significantly higher DM than Panicum and Guatemala at all the three frequencies of harvesting alongside defoliation heights.However, Panicum harvested at 4-weekly interval alongside defoliation height of 10 cm above the ground produced the highest DM yield that was not significantly different from Napier cv Ouma harvested at the frequency of 4-weeks relative with 10 cm defoliation height.This finding demonstrated that Panicum species out yielded Guatemala in western Kenya and therefore, is a potential candidate of alternative fodder grasses in the event that a solution for stunt disease on Napier cv Ouma is delayed.Furthermore, fodder species harvested at 4-weekly interval cumulatively alongside defoliation height of 10 cm above the ground yielded highest dry matter compared to other intervals of harvest along with basal heights of 5 and 15 cm across the experimental sites.

Table 1 .
Interaction between frequency of harvest, defoliation height and species on cumulative dry matter yield at Kakamega site.

Frequency of harvest (weeks) Defoliation height (cm) Dry matter yield t/ha/year Guatemala Napier cv ouma Panicum
DMRT0.05 = 1.66;CV% = 5.34; NoMeans marked by different letters are significantly different at p<0.05 significance level.mm and 186 mm (Total 891.1 mm) at Kakamega and 186 mm and 460 mm (Total 646 mm) at Alupe.In 2013, Kakamega recorded 1064.3 in LR and 634.6 in SR (total 1698.9mm) while Alupe recorded 1190 in LR and 515 in SR (total 1705mm)."More than 75% of the precipitation was received in long rain season (March-August).

Table 2 .
Interaction between frequency of harvest, defoliation height and species on cumulative dry matter yield at Alupe site.

Table 3 .
Correlation coefficient (r) between biomass yield and yield components for Guatemala species.

Table 4 .
Correlation coefficient (r) between biomass yield and yield components for Napier species.

Table 5 .
Correlation coefficient (r) between biomass yield and yield components for Panicum species.