Organic Resource Management in Kenya

Atmospheric change and global warming are pressing concerns that affect the future of humanity. While a majority of greenhouse gas emissions result from burning fossil fuels and from inefficient industrial processes, about 1/3 of carbon dioxide imbalances occur from land use change, particularly deforestation and soil degradation (Noble and Scholes, 2001). Accordingly, one of the most important short-term sinks for atmospheric carbon dioxide (CO₂) is through increasing carbon stocks in the plants and soils of managed and degraded lands, including smallhold farming systems of the tropics. Land managers in East Africa seldom recognize the importance of carbon stocks on their farms (Woomer et al., 1997), or that environmental interests are likely to develop large scale projects in the tropics that are designed to protect and increase carbon stocks at the landscape and national scales (Trexler, 1993). Land managers who document carbon gains on their land could become compensated through these projects.

It is unrealistic to expect individuals or communities to protect and foster that which they do not understand, and this is certainly the case for carbon stocks in smallhold farming systems. Carbon (C) exists as an inseparable component of vegetation, litter and soil organic matter, and is primarily lost as an invisible gas (CO₂), factors which complicate the understanding of carbon stocks and dynamics to non-scientists. When asked what is the likely crop yield of maize in a maturing field or meters of poles in a woodlot, a land manager can often provide an educated guess, but this is not the case for system C stocks within those same land uses. Carbon seems too intangible for approximation. Yet carbon is predictable from certain perspectives. For example, it is a near constant proportion within vegetation (45% to 49%).

An important empirical relationship exists between the tree diameter at breast height (DBH) of trees and tree aboveground biomass. Allometric equations based upon power functions, which intercept the origin, are recommended above quadratic approaches because of their greater accuracy for assigning biomass to smaller trees. For general purposes, we recommend the equations from FAO (1997) in Dry Zones (<1500 mm yr^-1):

where Y is the aboveground tree biomass in kg, exp = 2.71828… and D is the measured DBH in cm. Other equations are available for drier (<900 mm yr^-1) and wet zone (>4000 mm yr^-1) from FAO (1997). Allometric equations may be further refined by including factors for tree height and wood density (Ketterings, 2001). Measurement of tree diameter is easily made using either a diameter tape or callipers (Figure 1) but the mathematics required to convert from diameter to biomass is probably too complex for most land managers in Africa. Tree diameter (D) is readily calculated from tree circumference (C) by division by pi (pi = 3.14159…) where C = D x pi.

Figure 1. Measuring tree diameter with a dial caliper (left) and a diameter tape (right).

A simple table was prepared that allows for the estimation of tree carbon based upon established biophysical relationships. Table 1a provides the total C (in tons) contained in aboveground woody biomass of different sized trees based upon a widely employed allometric relationship between tree diameter at breast height (DBH) and total tree biomass. Table 1b also includes C in roots (+0.35) and the turnover of leaf drop ( 0.15 woody biomass) and fine roots (0.15 woody biomass) assuming modest (0.12) annual C sequestration in soil. Table 1 provides estimates of total C resulting from different sizes and numbers of growing trees. For example, a row of 10 trees that have grown to 30 cm diameter contain 2.42 t of carbon.

The tree biomass carbon relationship is independent of land area, so that tree numbers (rows) may be obtained from different size categories (columns) and the carbon stocks estimated for any known land area, such as different sized smallholdings. Carbon stocks may not be readily interpolated between columns because of the exponential nature of the allometric function. In other words, tree biomass C for a tree 27.5 cm in diameter does not occur midway between the 25 and 30 cm diameters but rather is skewed toward the higher diameter. Extrapolation may be made, however, by extending the values obtained within the rows. For example, the value for 35 trees from a diameter size category is equal to that of 30 trees + 5 trees of that same size category. Some practice in the field will show that Table 1 provides a useful and fairly simple tool to estimate tree carbon.

Table 1a. Estimates of total tree biomass for different tree numbers and diameters at breast height (DBH) based upon aboveground biomass (AGB) where AGB C = 0.47 x exp^{(-1.997
+2.32 (ln DBH)} and root biomass = 0.35 AGB.

Table 1b. Estimates of combined tree biomass and soil carbon gains for different tree numbers and diameters at breast height (DBH) based upon aboveground biomass (AGB) assuming that AGB C = 0.47 x exp^{(-1.997 +2.32 (ln DBH)}, root biomass = 0.35 AGB, leaf drop = 0.15 AGB, fine root turnover = 0.15 AGB and soil sequestration = 0.12 t SOC t^-1 leaf and fine root inputs

Carbon stocks may also be estimated for crops based upon their yield, harvest index and root-to-shoot ratio. Harvest index is the proportion of aboveground biomass that is removed as crop yield. For example, if a one ton crop of maize grain has a harvest index of 0.35, then the total crop aboveground biomass is 1.00/0.35 or 2.86 t, and the stover is 1.86 t (or 2.86 aboveground – 1.00 t grain). Furthermore, if one assumes that grain, shoots and roots all contain 47% C and that root biomass is 0.35 of shoot biomass, then the total crop carbon at peak biomass before harvest is 1.81 t C (2.86 x 1.35 x 0.47). This relationship may be summarized as:

Peak biomass C = crop C content x (crop yield / harvest index) x (1 + root: shoot ratio)

This approach was used to develop a table of crop carbon contents for different yields and harvest indices (Table 2). For example, a 2750 kg crop (= 2.75 t) with a harvest index of 0.25 contains 7.0 t C in its grain, shoots and roots, regardless of the land area upon which it was produced. This value refers to the peak biomass carbon, and it should be time-averaged throughout the year based upon the length of the growing season. Time-averaging requires that the peak season biomass and the number of wet months (the growing season) be known, and is calculated as:

Table 2. Total crop carbon (tons of grain, shoots and roots) at peak biomass before harvest for different harvest indices and crop yields assuming 47% C content in biomass and roots are 35% aboveground biomass.

Figure 2. Peak biomass carbon, mean biomass carbon and time-averaged biomass carbon for a crop containing 6.0 t C where mean biomass C = 0.5 x (peak biomass C) and time-averaged biomass C = mean biomass C/(12 / wet months)

For example, if the growing season is 6 months during the year, the mean carbon content for the wet season is 3.5 t C (7.0/2). If the fields sit barren during the following dry six months, then the time averaged standing carbon stock is 1.75 t C throughout the year (see Figure 2). As the length of the growing season increases, so does the time-averaged biomass C. These equations also hold for intercrops or bimodal rainfall patterns if one combines the two annual crop yields in Table 2 and sums the total wet months. For example, if 3 t maize with a harvest index of 0.35 is grown in a five-month growing season in the first rains, and 1.5 t beans with a harvest index of 0.25 is produced during the three-month “short season”, then:

Large amounts of carbon reside in the soil, but this C may not be as easily estimated as that in trees or crops. The measurement of soil organic carbon requires a laboratory where either wet digestion or dry combustion is performed. The results are expressed as grams of carbon per kilogram of soil (= parts per thousand) or as percent C (= parts per hundred). In general, soils range from about 5 to 25 g kg^-1, or 0.5 to 2.5% C. But this value, the carbon content, does not describe how much carbon resides in a particular field unless we know how much the soil weighs because some soils are heavier (more dense) than others. To convert from volume of soil to the weight of soil, we must also know the soil bulk density, the mass of soil per unit volume, and the depth of soil that is of interest.

Soil C (t ha^-1) = C content (kg kg^-1) x bulk density (kg liter^-1) x (10 x soil depth (m^-2))x 10000 m² ha^-1 x (0.001 t kg^-1)

Soil C (t ha^-1) = C (kg kg^-1) x bulk density (kg l^-1) x soil depth (cm) x 100

In general, soil bulk density ranges between 1.1 to 1.6 kg of soil per liter (= 1000 cubic centimeters) depending on the soil texture. Usually, the plow layer is considered to be 0 to 20 cm depth, and the root zone is from 0 to 50 cm depth. The amount of soil C in one hectare (tons C per ha where 1.0 ha equals 10,000 square meters) to a depth of 20 cm (= 200 per square meter), with a bulk density of 1.3 kg per litre (kg l^-1) and a carbon content of 15 g C per kg soil (= 0.015 kg C per kg soil) is calculated as:

Soil C (t ha^-1) = 0.015 kg kg^-1x 1.3 kg l^-1 x 200 l m^-2x 10000 m²ha^-1 x 1 t (1000 kg)^-1 = 39 t C per ha

Again, this equation is rather complex for most non-scientists but tables may be constructed that simplify the mathematics. Table 3 provides the total soil organic carbon per ha in the top 20 cm and 50 cm horizons for soils of different textures (columns) and C contents (rows). In this case, it is not possible to generate an estimate independently of land area because the soil C stocks are a direct function of land area and soil depth; therefore, land managers who employ these tables are then expected to adjust their estimate based upon the land area under consideration.

An important feature of this table is its potential for interpolation, as all relationships are linear. For example, the C stock value of loamy clay or sandy clay is midway between the tabular values presented within the respective columns. Furthermore, the relationships within this table also may be applied to soil C fluxes as well as stocks. For example, if 10 g C per kg soil is lost due to soil erosion or intensive tillage, that carbon loss (or gain) may be estimated directly from the table.

Lengthy mathematical discussion of these tables may distract from their overall purpose, to allow for rapid and accurate estimation of woody biomass and soil carbon stocks based upon minimum information. Carbon stocks which could not be “visualized” by land managers, development specialists or extensionists may now be quantified using these tables (Figure 3). Take for example the carbon gain resulting from an improved tree fallow producing 1000 trees per ha of 15 cm diameter and that increased total soil organic carbon in the loam by 0.8% C.

The woody biomass gain per ha is 0.033 t x 1000 = 33 t ha^-1. The soil C gain (for 8 g C per kg soil) is 35.2 t C ha^-1, yielding a system C estimate of 68.2 t C ha^-1. This value is best adjusted over time. For example, if the fallow interval is five years, then the C sequestration rate for woody biomass and total system C is 6.6 and 13.6 t ha^-1yr^-1, respectively.

Table 3. Soil organic carbon (SOC) (t ha^-1) in different textured soils resulting from changes in the SOC content (g kg^-1 soil) at soil depths (0-20 and 0-50 cm).

Figure 3. A stepwise approach to estimating a farm carbon baseline that considers trees, crops and soil that may be adjusted for different land areas.

It must be emphasized that these tables are intended to assist a wider cross-section of the land management and environmental communities to become involved in the estimation of carbon stocks, and in some ways are over-simplified. Table 1 is based upon a preliminary assumption of a single widely applicable allometric equation predicting aboveground tree biomass and this table could be better refined for more applicable DBH size categories and different tree species and vegetation zones. Table 2 presents yield increments of 500 kg and assumes that one is aware of the harvest index for a given crop. Table 3 assumes that the range of interest for soil C stocks is 25 g C kg soil^-1 and that five textures adequately cover the range of soil texture conditions.

Figure 3 illustrates how to use the information in this chapter to estimate the carbon in a field or on your farm. To do this, one must know the number of trees and their diameters, crop yields and have an estimate of the harvest index, the soil organic C content and the soil bulk density. Get a paper and pencil (or a good calculator) and then refer to Table 4 to compile a farm or field carbon baseline, the carbon gains from tree planting and the value of that carbon. Table 4 is completed using the following procedure:

Step 1. Establish baseline: tree biomass C. This section of the form is intended for completion before the initiation of a carbon offset project, or may be completed by comparing a cropland adjacent to tree planting, assuming that past land use and soil are representative. Enter the DBH and number of trees that fall into up to three different size categories and refer to Table 1a to identify the tree biomass C for each category by matching the tree diameter (columns) and number of trees (rows). Additional categories may be included on a separate sheet if necessary. Sum the categories to obtain the Total Tree Biomass C and enter this value into the far right column of Table 4.

2. Step 1. Crop Biomass C. Enter the yield and harvest index for up to two crops grown either sequentially within the same year or as intercrops. And identify the peak C for each by matching the harvest indices (columns) and crop yield (rows) from Table 2. Sum these values to obtain the Total peak crop C and enter this value into the far right column of Table 4. Time-average this value by including the total number of wet months.

3. Step 1. Soil C. This section requires that the soil be analyzed for C and the results expressed as g C per kg soil (= 0.1 x C%). Based on either soil texture or bulk density, identify the appropriate column and match this with the appropriate C content in Table 3 to obtain the value for total soil C (t C per ha) and enter it into the far right column of Table 4.

4. Step 1. Total system baseline C. Calculate this value as the sum of total tree, time-averaged crop and soil C and enter it into the far right column. This value is the baseline C.

5. Step 2. Project C gains: Tree and Soil C. This part of the form is intended to be completed at regular intervals (e.g. once a year) after the planted trees are established and growing. Enter the tree numbers and diameters and identify their C contents, this time using Table 1b, which also considers C gain in the soil beneath the trees.

6. Step 2. Intercrop C. Include the time-averaged C content contained in intercrops (Table 2), understorey or cover crops and adjust the value by wet months. Many cover crops lack “yield” so the biomass C must be obtained through destructive sampling.

7. Step 2. Project C gains. Sum the tree and crop C values. This is the unadjusted Total C gain.

8. Step 3. Net C gain. Calculate this value by subtracting the baseline value, but do not include the baseline soil C (baseline tree and crop C, but not soil C) and enter in the far right column. Calculate the value of this C by multiplying it by the C price, usually $10 per t C.

Time-averaged crop C (TACC) = (0.5 x (TPCC)) / (12 – wet months) = __________ t C

Total soil C (TSC) = Soil C (t ha^-1) / land area (ha) = _______________ t C

Intercrop 1 ______________ yield ______ harvest index _____ peak C _____________

Intercrop 2 ______________ yield ______ harvest index _____ peak C + ___________

Time-averaged crop C gain (TACG) = (0.5 x (TPCC)) / (12 – wet months) = _________ t C

Food and Agriculture Organization of the United Nations (FAO). 1997. Estimating Biomass and Biomass Change of Tropical Forests: A Primer. FAO Forestry Paper 134. FAO, Rome. 55 pp.

Ketterings, Q.M., Coe, R., van Noordwijk, M., Ambangau, Y. and Palm, C.A. 2001. Reducing uncertainty in the use of allometric biomass equations for predicting aboveground tree biomass in mixed secondary forests. Forest Ecology and Management 146:199-209.

Noble, I., Scholes, R.J. 2001. Sinks and the Kyoto Protocol. Climate Policy 1:5-25

Trexler M.C. 1993. Manipulating biotic carbon sources and sinks for climate change mitigation: Can science keep up with practice? Water, Air and Soil Pollution 70:579-593

Woomer P.L., Palm C.A., Qureshi J.N., Kotto-Same J. 1997. Carbon sequestration and organic resource management in African smallholder agriculture. In: Lal, R., Kimble, J.M., Follett, R.F. and Stewart, B.A. (Eds.) Management of Carbon Sequestration in Soils. Advances in Soil Science (series), CRC Press, Boca Raton, US. pp. 58-78

Back | Next

Chapter 7

Estimating Carbon Stocks in Smallhold Agricultural Systems