THE IMPACT OF LIMING ON BIODIVERSITY IN EMBU TEA ZONE LANDSCAPES : A CASE STUDY OF KAVUTIRI

Biodiversity in agro-ecological zone UM1 on southern slopes of Mt. Kenya commonly termed as tea zones has been declining due to many factors among them soil acidification. In Embu tea zone specifically Kavutiri area, acidification has been increased by intensive agriculture without proper soil management. Soils have developed high acidity level ranging between pH 4.2 – pH 4.6. Whereas lime has been used in the past to reduce the soil acidity in many parts of the world, it has not been experimented for flora and fauna diversity in Kenyan acid soils. This research was conducted to test the effect of soil liming on biodiversity in the acidic soils of Kavutiri area of Embu County. A randomized complete block design with four blocks and four treatment plots per block was laid out. The treatments comprised rates of lime which were broadcasted on plots as follows; 0 (L0), 2.4t/ha (L1), 6t/ha (L2), 8t/ha (L3). Soil, flora and fauna data were sampled 9 months after liming. Soil parameters that increased with increase in liming from L0 to L3 were pH, BS%, available phosphorus and exchangeable bases (Ca, Mg, K and Na). However, ECEC, extractable Al and Mn decreased from L0 to L3. Flora characteristics within the treatment plots differed with particular species. Fauna diversity increased with increase in flora diversity. Limed plots had significantly (p<0.05) more benefits to biodiversity than the control plots. Liming level L2 gave the most recommendable results.


Introduction
Biodiversity in agro-ecological zone UM1 (Sombroek et al., 1980) on southern slopes of Mt.Kenya commonly termed as tea zones (Unilever Sustainable Agriculture Advisory Board, 2003) has been declining due to many factors among them soil acidification (Susan and Fahrig, 1995).In Embu tea zone specifically Kavutiri area, acidification has been increased by intensive agriculture without proper soil management (Warren and Kihanda, 2001).Excessive rain that leaches the soil and depletes calcium and magnesium, organic matter and application of acidic nitrogen fertilizers have not made the situation any better.Kavutiri soils have developed high acidity level ranging between pH 4.2 -pH 4.6 (Warren and Kihanda, 2001).Few species like Pteridium aquilinum, Tagetes minuta and Penisetum sp were very abundant and hence dominated the site before experiment.These species are tolerant to acidity.
Soil acidification disrupts flora and fauna diversity due to toxicity of elements such as aluminum and manganese and unavailability of nutrients such as phosphorus, magnesium and calcium (McKenzie, 2003).While a few species like Pteridium aquilinum,Tagetes minuta and Penisetum sp tend to adapt to acidic soils, many others species either disappear completely or their numbers are reduced drastically leading to instability in ecosystem (William, 1998).To have a stable ecosystem and sustainable ecology, high abundance alone is not enough.High richness and evenness are equally needed (Graeme, 2007).
Whereas lime has been used in the past to reduce the soil acidity in many parts of the world (Smith et al., 1994), it has not been experimented for flora and fauna diversity in Kenyan acid soils.Could liming be the key to the ever narrowing diversity in acidic soils of Kavutiri?This research was conducted to test the effect of soil liming on biodiversity in the acidic soils of Kavutiri area of Embu County.

Study Site
The experiment was carried out in Embu County, Kavutiri area, Agro-Ecological Zone (AEZ) UM1 (Sombroek et al., 1980).It is located about 20 km north of Kenya Agricultural Research Institute, Embu at an elevation of 1700m.The coordinates are 0° 24' S, 37 o 29' E, and slope of 6.53%.The area has a mean annual rainfall of 1736mm with mean monthly maximum and minimum temperature ranges of 19.9-27.8 °C and 11.5-14.3 °C respectively (Warren and Kihanda, 2001).The soil type is ando-humic Nitsols (FAO- UNESCO, 1988).It has a bulk density of 0.77g/cm 3 and Al saturation of 74.45%.Though clayey in texture, it has loamy physical characteristics due to high content of kaolinitic and hydrous oxides (Mugai et al., 2008).

Experimental design and treatment
Land was normally prepared and a land measuring 32m x 28m was measured.A randomized complete block design with four blocks and four treatment plots per block was then laid out.The plot dimension was 4m x 4m.The distance between plots within a block was 2m and distance between blocks was 3m.An allowance of 2m was given for the paths around the whole experimental area.Each plot was clearly demarcated using wooden pegs.
Two weeks after liming, maize and beans crops were planted to simulate the cropping practices.The plots were weeded twice in 3 rd and 10 th weeks and subsequently reserved for colonization of flora and fauna for another nine months.

Sampling
Soil, flora and fauna data were sampled 9 months after liming.Due to the differences in methods of sampling, the methods used are described separately.

Soil sampling and analysis
Three soil sub-samples from every plot were collected by augering at 0-15cm depth after which a composite sample for every plot was obtained.Soil pH was determined by a pH meter (EYELA model pH M2000) in water 1:2.5 and 0.01M CaCl 2 1:2.5 suspensions (Okalebo et al., 1993).Exchangeable cations (Ca 2+ , Mg 2+ , K + and Na + ) were determined after extraction by ammonium acetate buffered at pH 7.0 (Thomas, 1982).The K + and Na + concentrations in soil extracts were read on 410 flame photometer while Ca 2+ and Mg 2+ concentrations in soil extracts were read using atomic absorption spectrophotometer (AAS) Perkin-Elmer Model 403.Available phosphorus in the soil was determined using Bray P1 method (Gary and John, 2009).The total sum of exchangeable bases (Ca 2+ +Mg 2+ +K + +Na + ) and exchangeable acidity (Al 3+ ) gave the effective cation exchangeable capacity (ECEC) (Okalebo et al., 1993).Percentage base saturation was calculated by dividing the sum of exchangeable bases with the sum of cations including aluminium.Mn, was extracted with 0.1M HCl (Okalebo et al., 1993) and its concentration in soil extracts was read on AAS (Perkin-Elmer Model 403).

Flora sampling
A quadrant of 1m x 1m was used to guide the sampling within the plots.A quadrant was laid starting from the right bottom side of the plot and continued to the rest of the plot to ensure nothing was left out or repeated.

Fauna sampling
Non-flying organisms' data was collected by counting organisms in a quadrant of 1m x 1m, that was laid starting from the right bottom side of the plot and continued to the rest of the plot similar as was done for flora.Flying organisms' data was collected by the person wearing an overall for protection then lying against the ground surface, perpendicular to the center of the plot, three times in a day (morning, noon and evening) in order to cover as many species as possible.This was done for three minutes in each plot.Concentration and hence visual acuity may have declined if the sampling period exceeded approximately 3 minutes (David, 1998).
Looking directly at the plot through binoculars, organisms were counted as they pass though the focal volume visible through the binoculars.With the binoculars set to a known focal distance, all organisms detected during the count period were tabulated and when the alarm sounded, counting stopped.Because the magnification and focal distance of the binoculars used determined the minimum size of organisms seen, 2m from the plot was established as the range of distance within which small organisms such as flies were visible.This was done by choosing an object approximately the size of the smallest organism expected to be counted.In the study, the interest was to know the abundance of all the flying organisms relevant to the plot of 4m by 4 m, hence, when calibrating binoculars, 0.15 cm was chosen.

Biodiversity measurement
Four measurements were used to express biodiversity.They included; species richness (S), abundance, evenness (J) and Shannon's diversity index (H').Abundance is the quantity of species in a given unit area.It was obtained by counting the species in a given area (Beals, 1998).
Richness indicates the number of different species in a given area (Butler and Chazdon, 1998).It was obtained by counting different types of species in a given area.In many circumstances the species richness number is enough to allow comparisons, especially between geographic areas or habitats (Butler and Chazdon, 1998).But in this experiment, species richness alone could have been misleading.Species richness does not give details of how many times a species will occur in a plot.Think about two treatments, each with the same number of species but where one has one very common species with only a few individuals of the other species while the other community has equal numbers of each species.Intuitively we recognize that the plot where the individuals are spread out equally among the species has a more diverse community than the one where almost all of the individuals belong to one species (Beals, 1998).This richness drawback was corrected through determination of species evenness.
Species evenness defines the number of individuals from each species in an area (Emerson and Brent, 2005).It was used in this experiment to specify how close in numbers each species in given area were.Usually, evenness is constrained between 0 and 1.When all species are found in all plots as equally as possible, there is less variation in communities and the evenness is very close to 1 (high) and when one or very few species dominate the others the evenness value is close to 0 (low) (Beals, 1998).However, evenness by itself had a disadvantage because it only measured equitability but lost the information on species richness.As a result, in this experiment a more useful biodiversity measurement that combined species richness and evenness (Graeme, 2007) was included.This was Shannon's diversity index.
Shannon's diversity index (H') indicates the number of different species in an area as weighted by abundance in the same area (Beals, 1998).The index is given by the formula: Where (∑) is the summation sign, (pi) is proportion of each species to the whole species in the treatment plot, (nl) is the natural log, J is evenness and (S) is the species richness.
This index meant that for each species, the proportion (p) to the whole species in an area was determined, and then multiplied by the natural log (ln) of itself, this gave p ln p.This was done for each species (i) in the treatment plot and all of the resulting numbers were added up (∑).Since the resulting number comes out negative, it is normally turned into a positive number for convenience by multiplying with (-) (Beals, 1998).
These diversity parameters (species abundance, richness, evenness and Shannon's diversity index) were put in Ms excel program using Hutcheson (1970) methodology to simplify the computations.As explained in Zar, (1996), this method also estimate the variance of the diversity hence allowing the treated plots to be statistically compared with the control at α = 0.05.Means were separated using LSD.

Soil characteristics
Table 1 shows chemical properties in the control and limed plots after nine months.pH determined in water and CaCl 2 suspensions increased with liming from L1 to L3 (Table 1).Increase in pH after liming was expected due to presence of CO 3 2-in dolomite.Other soil parameters that increased with increase in liming from L0 to L3 were BS%, available phosphorus and exchangeable bases (Ca 2+ , Mg 2+ , K + and Na + ) (Table 1).The increase in base saturation was due to the supply of exchangeable cations by the lime.The increase in Ca 2+ and Mg 2+ from L0 to L3 was because the two were the main constituents of dolomite and these were deposited during soil liming.K + and Na + increase may have been the result of impurities in the dolomite (Mitchell, 1999).However, ECEC, extractable Al and Mn decreased from L0 to L3 (Table 1).This decrease in ECEC was attributed to decrease in extractable Al which surpassed the increase in Ca 2+ and Mg 2+ .The extractable Mn 2+ and exchangeable Al 3+ decreased with increase in liming, a trend that agreed with the statement of Wild (1995) that the solubility and mobility of metal trace elements in the soil decreases as soil acidity decreases.This is due to their decreased solubility of their oxides.The increase in available P from 18.0 to 49.4 for L0 and L3 respectively was attributed to mobilization of fixed P as pH increased (Wild, 1995).
While liming to pH 6.5 may form calcium triphosphate once again immobilizing phosphorus (Wild, 1995), the highest liming rate in the experiment which was 8t/ha (L3) did not raise pH to even beyond 6.0 probably due to high aluminum ions in Kavutiri soils.Aluminium ions are known to cushion soil from pH changes (Douglas, 1995).

Flora diversity characteristics
Table 2 shows the effect of liming levels on flora populations and the resulting flora abundance, richness and evenness after nine months.Flora characteristics within the treatment plots differed with particular species.This is probably because different plants have different survival mechanisms and hence respond to ecological changes differently (Emerson and Brent, 2005).Ageratum conyzoides and Sonchus arvensis abundance increased with increase in liming from L0 to L3 probably due to improved nutrition (Mitchell, 1999).Echium sp., Medicago sativa, Mwaura ciau, Oxygonium sinuatum and Viola sp were absent in L0 but colonized L1 and increased with increase in liming indicating that liming may have eliminated the factors that were suppressing their growth (Smith et al., 1994).Fallopia convolvulus, 'Muvururo', 'Muvuria ndudi', Stellaria media and Polygonum aviculare were absent in L0 and L1 but colonized L2 and L3 probably indicating that the factor that was suppressing their growth needed higher liming level than L1 and hence it took liming level L2 to trigger their emergence (Smith et al., 1994).
Setaria pumila also colonized plots from L2 only to decrease at L3 probably showing that this species needed a narrow range of liming level conditions to thrive below or above which it could not grow well (Butler and Chazdon, 1998) 2) as well as to elements such as aluminium agreeing with earlier findings (Mugai et al., 2008).However, the four; Ageratina adenophora, Pteridium aquilinum, Oxalis latifolia, Penisetum sp reduced due to liming probably due to enhanced competition from other species (Hutchinson 1959) that appeared following liming to their optimum pH levels.
General increase in flora abundance from 165 (L0) to 249 (L2), richness from 14 (L0) to 25 (L2) and evenness from 0.67 (L0) to 0.94 (L3) indicated at the bottom of Table 2 was due to the ability of liming to stabilize the nutrients and pH that had become low in acidic soils (Brett et al., 2005).In this case liming raised Mg and Ca, mobilized P and helped in reducing toxic microelements such as Mn and Al (Gordon, 1987).This provided a level platform for essential elements to be absorbed by plants leading to stability and high species evenness (Maneepitak, 2007) in L2 and L3.
Low evenness in L0 was caused by high abundance of some species such as Pteridium aquilinum and high suppression of others such as Spergula arvensis (Table 2).This trend threatens an ecosystem because in the long run suppressed species may eventually become extinct (Susan and Fahrig, 1995).Liming to L2 and L3 successfully averted such threats (Table 2).Dominating species reduced as liming proceeded to L3.This was advantageous because the conditions became favorable enough for species that were getting wiped out to re-inhabit more niches (Slattery, 2002) as was observed in the treated plots (Table 2).
The flora diversity index of the treated plots differed significantly (p≤ 0.05) from the control (Fig 1).L2 and L3 gave the highest diversity index score of 3.02 and 3.01 respectively (Fig 1).Diversity in L2 and L3 did not differ significantly from each other in 9 months of the experiment.The broadening of flora diversity from 1.77 in L0 to a high of 3.02 in L2 and L3 indicate that liming was reversing the toxicity effects of elements such as Al and Mn as well as mobilizing fixed P (Mitchell, 1999).Al 3+ and Mn 2+ are more soluble in low pH such that they get to toxic levels and are the major causes of plants failure in acid soils (Mitchell, 1999).

Above ground fauna diversity characteristics
Fauna diversity increased with increase in flora diversity (Fig. 1), indicating that the above ground fauna responded to liming indirectly (Maneepitak, 2007).The flora was the primary producer to which fauna depended on (Brett et al., 2005).Hence, higher flora diversity led to higher fauna diversity.These were attributed to close relationship between fauna and flora (Susan and Fahrig, 1995).
Table 3 shows the fauna abundance, richness as well as evenness with respect to the liming levels after nine months.15 fauna species increased from L0 to L2 and did not show further significant increase or decrease to L3. Interestingly though, Mosquito family and Coenosia sp decreased with increase in liming from L0 to L3.These fauna trend could be attributed to their substantial level of dependence on bushy flora such as Ageratina adenophrora and Pteridium aquilinum where they were seeking shelter and which had also reduced following similar trend (Susan and Fahrig, 1995).
In general, species abundance increased with increase in liming from 218 at L0 to 701 at L2, but fallen to 698 at L3. Species richness increased from 21 at L0 to 30 at L2 and L3, while evenness increased from 0.68 at L0 to 0.95 at L2 and L3 (Table 3).Higher abundance, richness and evenness were registered for fauna than flora and was due to mobility characteristic of fauna (Ryder, 1986).
In line with this, fauna such as flies could be attracted from other ecology en route to limed plots where they could find their interests (meal or shelter) (Maneepitak, 2007) leading to higher diversity in treated plots.

Conclusions and Recommendations
Diversity index score accounted for abundance, richness and evenness hence it was a more informative estimate of biodiversity (Hutcheson, 1970) in experimental plots.Using the diversity index score, it was therefore noted that limed plots had greater biodiversity than control.This indicated that soil liming was able to stabilize the bionetwork with L2 and L3 thus providing highest biodiversity stability.
Upon liming, there was a related increase in abundance, diversity index score, richness and evenness.Limed plots had significantly (p<0.05)more benefits to biodiversity than control.Liming level L2 gave the most recommendable results.While species abundance show different species performed better at different lime levels, the Shannon diversity index in plots clearly indicated that there was no further improvement in biodiversity after liming level L2.More research is needed to determine why diversity ceased increasing after acidic soils were limed beyond pH 5.6.While biodiversity in L3 did not significantly deviate from that of L2, it was economically cheaper to apply L2 compared to L3. Therefore liming level L2 which was 6t/ha was recommended as the appropriate rate for conservation of sustainable biodiversity in acidic soils of Kavutiri.
Table 3.Effect of liming levels on fauna populations and the resulting flora abundance, richness and evenness after nine months.

Figure
Figure 1.Effect of liming on flora and fauna diversity after nine months based on 95% confidence intervals

Table 1 .
Soil chemical properties as influenced by liming after 9 months.

Table 2 .
Effect of liming levels on flora populations and the resulting flora abundance, richness and evenness after nine months.