Site Description
The study was carried out in Minjar Shenkora Woreda, North Shewa Zone of Amhara Regional State, Ethiopia which is located at about 135 km south-east of Addis Ababa, at 90º 6' and 90º 5' N and 39º 46' and 39º 26' East and has a total area of about 229,463 ha. The altitude of the study area ranges from 1400–2400 m.a.s.l. The topography of lands in the woreda is characterized by diverse geomorphological features. Data from the Woreda agricultural office indicate that plateau or flat plains features (65%), followed by 20% of the land area is mountainous, ravines (10%) and 5% other topographic features.
According to MSDARDO, Minjar Shenkora woreda has different soil types suitable to harvest various kinds of grains. The most dominant soil type in the study area is heavy clay soils known as vertisols and reddish-brown loam known as cambisol.
The woreda falls within three major agro-climatic zones, Dega (high altitude), Weyna Dega (Mid altitude) and Kola (low altitude). The highest mean annual rainfall of the study area within the last ten years was 1028 mm, whereas the lowest mean total was 162.8 mm. According to North Shewa Agricultural and Rural Development Bureau, Minjar Shenkora district has annual average temperature range between 13.210c and 23.020c.
The vegetation type is Acacia wooded grassland (Ib Friis, 2010). On most of the plain areas on which crop cultivation is dominant, Faidherbia albida, A. tortilis, A. seyal, A. nilotica, Croton macrostachyus, and Ziziphus mauritiana are scattered across with the farm plots which are the main components of agroforestry type of agricultural system with agricultural crops.
Cereals and pulses are among the commonly cultivated crops in the area for the purpose of household consumption and income through the sale. These include sorghum (Sorghum bicolor), wheat (Triticum aestivum), teff (Eragrostis teff), barley (Hordeum vulgare), onion (Allium cepa), pea (Pisum sativum), Chickpea (Cicer arietinum) and Horse bean (Vicia faba). Currently, due to the introduction of rainwater harvesting technology through ponds, farmers grow vegetables in small gardens as well as in the fields.
Materials
The equipment used for fieldwork should be accurate and durable to withstand the rigors of use under adverse conditions. The type of equipment required depended on the type of measurements (Pearson et al., 2005). The following materials were used for this study to collect available data meter tape for measuring distances of sample plot, hypsometer to measure height of tree, caliper to measure tree diameter, auger to take soil sample, cloth or paper bags to collect soil sample, core sampler to take soil sample for bulk density and Global Positioning System (GPS) to collect coordinate point of study site.
Sampling and data collection Methods
The Minjar Shenkora woreda is selected purposively as a study area by considering the extensive presence of parkland agroforestry practice. A preliminary reconnaissance survey had been conducted to identify the study area. Key informants i.e. development agents, elders and woreda natural resource experts were consulted to identify farmers and study site that contain parklands agroforestry practice on their lands-based accessibility, resource and time. The woreda is composed of 27 kebeles out of which parkland agroforestry practice is found in 21 kebeles. Moreover, the woreda has three agro-ecologic zones: dega with 6 kebeles; weyna dega with 11 kebeles; and kola with 10 kebeles. Based on this information from the woreda office of agriculture, two-stage stratified random sampling technique was used to select the unit of sampling for the study. In the first stage, a total of four kebeles: two kebeles (Bolo Giorgis and Bolo Slase) from weyna dega and two kebeles (Agirat and Korma Ager) from Kola agro-ecologies were selected randomly as specific study areas. In the second stage, by considering the list of farmers who owned farms with parkland agroforestry practice in the kebeles as a sampling frame, a total of 32 farm plot for the carbon stock determination, size 40 m*40 m (Talemos and Sebsebe., 2017; Tadesse, 2015) were randomly selected from Bolo Giorgis, Bolo Slase, Agirat and Korma Ager. The sample size accounts about 7% of the household and therefore considered adequate to balance reliability and cost. There are different mechanisms to determine the number of samples: the census for small populations, imitating a sample size of similar studies, using published tables, and applying formulas to calculate sample size (Israel, 1992). Among the listed methods sample size determination using similar studies was employed for the present study (Asefa and Worku, 2014; Tadesse, 2015).
The plot method was used that involves selecting plots of an appropriate size and number, laying them randomly in the selected strata (Tadesse, 2015). Plots can be marked at four corners in conspicuously (for example, by sinking available material below the ground and navigating to plot using a GPS. The plot size chosen was large enough to encompass the diversity of tree species on the smallholdings (Negash, 2013; Nair et al., 2008).
Carbon stock of woody species (dead trees, live trees), below-ground biomass (stumps plus coarse roots; >2 cm diameter and fine roots) and soil organic carbon were estimated. The farmland of sample households was used as a sample plot for inventory. Accordingly, woody species inventory was carried out on the farmlands of selected households located in the kebeles. In present study, woody species data were collected from 40 m × 40 m sample size quadrates (Nikiema, 2005), but the quadrat size in the study of the mentioned author is 50 m*100 m because of low density of trees on farmland in the study area, However; in our case as the density of trees on farmland is relatively high, a sample plot of 40 m*40 m was considered as an optimum plot size. The collected data were the name of species, tree diameter at breast height, tree height, tree diameter at stump height, soil sample, and location of the plot using GPS. All the woody species in each sample plot ≥ 5 cm DBH (diameter at breast height) were measured because below these DBH there are insignificant amount of biomass (Motuma et al., 2008). At every sampling point, a number of individuals per plot, DBH, height, and DSH of live trees were measured and recorded by using a measuring tape, caliper, and hypsometer.
At every sampling point from the selected study site, 20 m × 20 m subplots were taken for soil sampling from each corner and at the center of the plot. The most common depth for sampling is 30 cm but sometimes SOC is sampled to up to 1 m (David, 2013). The importance of sampling beyond the surface soil cannot be overemphasized while studying tree-based systems such as agroforestry, not only because tree roots extend to deeper soil horizons, but also because of the role of subsoil in long-term stabilization of C. Soil sample taken in the depth of 0–20 cm and 21–40 cm by using auger and core sampler. Two sets of soil samples were taken, one set for the determination of SOC contents and fine root (< 2 cm diameter) biomass and others for the determination of soil bulk density. In each case, samples of the 0–20 cm and 21–40 cm layers were taken from the four corners and center of each of 20 m × 20 m tree inventory plot and composited by layer while following (Negash and Starr, 2015). Two soil samples were taken from each sampling point after compositing the same depths together to get one representative soil sample. Four replication *eight plots for each replication * two soil depth, and hence a total of 64 soil samples were taken for soil carbon analysis.
Data Analysis
The non-destructive method was used for the estimation of carbon stock. The C content of tree biomass had been taken to 48%, the biomass weighted mean value for trees grown in agroforestry systems in Kenya (Kuyah et al., 2012a). To estimate biomass of tree species-specific allometric equation from woody biomass inventory for Ethiopia was used for almost all tree species but for Citrus sinensis general allometric equation (Kuyah et al., 2012a) was used.
AGB = 0:0905 * DBH2:4718; R2 = 0:98; n = 72 --------------------- (1)
This equation was developed in areas having similar environmental conditions (climate and soils) with the study area. CO2 was calculated using this formula; the amount of carbon is multiplied by the ratio of the molecular weight of carbon dioxide to the atomic weight of carbon (44/12) (Tadesse, 2015).
Below ground biomass (BGB) (stump plus coarse roots, > 2 cm diameter and fine root) was estimated by using the allometric equation (Kuyah et al., 2012a);
BGB = 0.490AGB0:923; R2 = 0.95; n = 72 -------------------------- (2)
The soil samples were analyzed to determine SOC, soil organic matter, pH, soil texture, and bulk density in Debre Dirhan Agricultural Research Center soil laboratory. We determined the soil bulk density by dividing the weight of oven-dried soil sample by the volume of core.
SOC is determined through laboratory analysis of the soil carbon concentration, volume of the soil sample, and the bulk density of soil samples collected in the study area (David, 2013). The soil samples were dried in oven-dry by 70 oc for SOC content (%) and 105 oc for bulk density until getting constant weight and analyzed (Negash and Starr, 2015). Before that, the soil samples were treated with HCL acid to remove inorganic carbon. Then the soil organic carbon was determined following the wet digestion method (Walkley, 1934), A correction factor of 1.33 was applied to account for the incomplete oxidation of organic carbon that is known to occur with the Walkley-Black method (Rosell and Gasparoni, 2000).
Soil organic carbon (SOC) was determined by using the equation (Subedi and Pandey, 2010; Pearson et al., 2005);
SOC (Mg/ha) = BD (g/cm3) × depth (cm) × carbon%×10− 1. ------------------------------- (3)
Where: BD = bulk density and 10− 1 is a unit factor (10_9 mg Mg_1 × 108 cm2 ha_1).
BD (gm/cm3) = (oven-dry weight of the soil) / (volume of the core) -------------------- (4)