Biomass production and carbon fixation by prairie shelterbelts:
a Green Plan Project

The PFRA Shelterbelt Centre at Indian Head, Saskatchewan has been promoting the establishment of shelterbelts and other tree plantings in the agricultural regions of the Prairie Provinces since 1903. Recent concerns over possible climate changes due to greenhouse gas emissions lead to the signing of the Rio de Janiero Protocol at the 1992 Earth Summit. Wtih this signing the countries of the world, including Canada, agreed to take steps to reduce net emissions of carbon into the atmosphere and stabilize them at 1990 levels by the year 2000. It would be valuable to have an accurate method of measuring the amount of biomass, and carbon fixed in shelterbelts. Therefore the objective of the Green Plan shelterbelt biomass project was to determine the amount of biomass and carbon in prairie shelterbelts and to explore the potential of future tree plantings to offset atmospheric carbon emissions. The differences in biomass production in trees grown in the different soil zones of Saskatchewan was also investigated.

A total of 12 tree and shrub species produced by the Shelterbelt Centre were sampled for their biomass production capabilities (Table 1). Caragana was investigated in the fall of 1994. The remaining species were sampled between October 1995 and May 1996. Sampling of deciduous species took place after leaf drop in the fall or prior to bud burst in the spring. The sampling of coniferous species took place in the spring of 1996.

The calculation of woody biomass has been of primary interest to foresters who need an accurate method of determining the merchantable portions of standing timber. More recently, whole tree biomass has been investigated for a number of applications. The biomass of trees is calculated using tree measurements with the results dependent on the size and shape of the trees within a stand. Trees grown in shelterbelts have unique characteristics. Differences in size and shape required the development of biomass calculations specific to plantings of this type.

DECIDUOUS TREES - Mature shelterbelts of green ash, Manitoba maple, hybrid poplar and Siberian elm were selected at nine locations in Saskatchewan and Manitoba, with three sites in each of the black, dark brown and brown soil zones (Figure 1).

For each species, 40 trees at each site were measured for height, number of stems, crown width and diameter at breast height (DBH). From the 40 measured trees, three representative individuals were cut down at ground level. These trees were weighed and their ages calculated. From this information the relationship between tree characteristics and biomass was determined.

Subsamples of the felled trees were weighed green and dried to find the moisture content. Subsamples consisted of two pieces of wood taken from the main stem and two pieces from the crown. Green tree weights were adjusted with the dry sample weights to determine the dry tree weight or above ground biomass (Table 2). Dry weights of all species differed by soil zone (Figure 2). This is likely due to several factors including age of the shelterbelt, condition, soil type and local climate. With the exception of poplar, there was an overall increase in the mean tree biomass from the brown to the dark brown to the black soil zone.

The trend for poplar was the same as the other species in the dark brown and black soil zones, but poplar had an unexpectedly large biomass in the brown soil zone, larger than in the dark brown soil zone. In this area, with its arid climate, the only living mature poplar shelterbelts are in ideal locations with available water. This water, combined with a long growing season, lots of sun and a lack of competition, results in very large trees.

Dry poplar wood is an average of 48.2 percent carbon, green ash 48.6 percent, Manitoba maple 48.0 percent and Siberian elm 49.4 percent carbon. An average 1 kilometre poplar shelterbelt planted with a 2.5 metre spacing between trees would have an above ground biomass of 174.8 tonnes and would contain 84.2 tonnes of carbon. A green ash shelterbelt of the same length at a 2.0 metre spacing would have an average of 44.7 tonnes of carbon above ground. Manitoba maple would contain up to 41.8 tonnes of carbon above ground, while Siberian elm would hold 51.9 tonnes of carbon above ground. The total amount of carbon contained in a shelterbelt of these species can be determined using an assumed root to top ratio of 0.4:1 (Table 3).

*assumes a root to top ratio of 0.4:1

Linear regression analysis was applied to find the best relationship between above ground biomass and shelterbelt measurements. This was done to develop a method of predicting shelterbelt biomass. With the exception of Siberian elm, linear regressions of biomass on stem cross-sectional area at breast height alone produced larger R² values, indicating a better fit, than regressions of biomass on stem cross-sectional area multiplied by tree height. Linear regressions were run to determine the relationship between total stem cross-sectional area and above ground biomass (Figure 3). Regression R² values between 0.506 and 0.883 were obtained. The total stem cross-sectional area was calculated by summing the cross-sectional area at breast height for all stems in an individual tree.

CONIFEROUS TREES - White spruce, Scots pine and Colorado spruce have been important shelterbelt species for prairie farmers with over 135 000, 155 000 and 375 000 trees respectively shipped from the Shelterbelt Centre in 1994 alone. Most of these trees are planted around farm yards and are highly prized by the owners. Sampling of these species took place on the black soils of the Agriculture and Agri-food Canada's Experimental Farm and the PFRA Shelterbelt Centre, both at Indian Head, Saskatchewan.

The measurement and sampling methods for these species was the same as that used for the deciduous tree species described earlier. Subsamples of these trees were taken, weighed, dried and weighed again to determine the dry weight or biomass of the trees. For these species tree weights and biomass calculations included the needles.

Above ground biomass was calculated for each species (Table 4). The above ground biomass of white spruce trees in shelterbelts ranged from 68.0 to 529.4 kilograms per tree, Scots pine from 91.9 to 274.6 kilograms per tree and Colorado spruce from 134.4 to 449.7 kilograms per tree. The variability among dry weights is due to several factors including shelterbelt age, condition and spacing between trees.

Of the shelterbelts sampled, none had the recommended spacing of 3.5 metres between trees. This resulted in tree crown areas smaller than if spaced correctly. In shelterbelts, conifer tree crowns have a rectangular form determined by the distance between a tree and its neighbours (the tree spacing) and the spread of the crown on the remaining two sides. This results in a crown with less overall biomass than if the tree had no near neighbours.

Dry wood is approximately 50% carbon, with this value the amount of carbon in each tree can be determined, and from that the amount of carbon sequestered in a shelterbelt (Table 5). A one kilometre shelterbelt of white spruce at a 2.25 metre spacing would have a calculated above ground biomass of 128.2 tonnes and could contain 61.5 tonnes of carbon. A Scots pine shelterbelt at a 1.8 metre spacing would have 48.4 tonnes of carbon stored above ground. A Colorado spruce shelterbelt with 1.8 metre spacing would contain 63.2 tonnes of carbon.

The total amount of carbon sequestered in a shelterbelt containing these species can be determined by assuming a root to top ratio of 0.3:1. A mature white spruce shelterbelt one kilometre in length would then contain 80.0 tonnes of carbon. Similar Scots pine and Colorado spruce shelterbelts of the same length would contain 62.9 and 82.2 tonnes of carbon respectively.

Linear regression analysis was applied to find the best relationship between above ground biomass and tree measurements. Regression R² values ranged from 0.764 to 0.990. The linear regressions of biomass on stem cross-sectional area at breast height and crown area produced the best fit for Scots pine and Colorado spruce, R² values of 0.897 and 0.901 respectively. For white spruce, the best fit was found with multiple regression analysis of biomass on cross-sectional area at breast height multiplied by height and crown area (R² = 0.990). However, R² values ranging from 0.864 for Colorado spruce to 0.921 for white spruce from regression analysis of biomass on cross-sectional area at breast height indicated a good fit and represented the easiest sampling method and was therefore adopted (Figure 4).

SHRUB SPECIES - These species are some of the most important species produced at the Shelterbelt Centre. Caragana is the most widely distributed shelterbelt species on the prairies, with over 230 million produced. Choke cherry, villosa lilac, buffaloberry and sea-buckthorn are also popular shrub species with prairie farmers with 354 000, 650 000, 276 000 and 220 000 seedlings respectively shipped in 1994.

A total of twenty-three caragana shelterbelts were sampled at Wynyard, Conquest and Aneroid, Saskatchewan in the black, dark brown and brown soil zones, respectively (Table 8). At each shelterbelt, three 10-metre sections were randomly selected for sampling. Choke cherry, villosa lilac, buffaloberry and sea-buckthorn sampling occurred at the PFRA Shelterbelt Centre at Indian Head, Saskatchewan. With the exception of sea-buckthorn, three mature shelterbelts of each species were sampled. For sea-buckthorn, only two shelterbelts were sampled. At each sampled shelterbelt, measurements of average height, number of stems, number of shrubs, shelterbelt width and diameter at breast height (DBH) of all stems was recorded. The shelterbelt sample was then cut down at ground level and weighed. Subsamples of representative plant material (stem, branches and twigs) were dried to a constant weight to determine the dry weight or above ground biomass of the sampled shelterbelts (Table 6).

The above ground biomass of 10 metre sections of caragana averaged 516 kilograms and ranged from 412 kilograms to 602 kilograms. Shelterbelts from the dark brown soil zone had the greatest biomass. An explanation for this is that the climate and conditions in this area most closely resemble the conditions of its native range resulting in healthier, sounder shelterbelts. The above ground biomass of the five metre long sections of choke cherry ranged from 122.8 to 252.7 kilograms, villosa lilac from 114.1 to 216.9 kilograms, buffaloberry from 103.4 to 226.8 kilograms and sea-buckthorn from 106.4 to 106.6 kilograms.

* Carbon content assumed to be 50% of dry weight

**Assumes a root to top ratio of 0.5:1

Caragana wood samples were sent for elemental analysis (Appendix Table 1). Dry caragana wood is an average of 50.1% carbon. With this value, and an estimated carbon percentage of 50% for the other species, the amount of carbon stored in shelterbelts of these species was determined (Table 7). An average caragana shelterbelt one kilometre in length would contain 38.9 tonnes of carbon above and below ground. Similar choke cherry, villosa lilac, buffaloberry and sea-buckthorn shelterbelts would contain 30.2, 25.1, 23.4 and 16.1 tonnes of carbon respectively. The total amount of carbon held in these shelterbelts can be calculated using an assumed root to top ratio of 0.5:1.

Regression analysis was performed on above ground biomass of the shrub species as it depended on the shelterbelt measurements. For caragana, the best correlation was between above ground biomass and height and stem cross-sectional area. Linear regressions of biomass on shelterbelt volume produced R² values of 0.722 and 0.618 for choke cherry and villosa lilac. Shelterbelt volume was calculated by multiplying the shelterbelt sample's length by width by height. Biomass and shelterbelt volume analysis for buffaloberry produced an R² value of 0.865, but an R² value of 0.927 obtained from regression analysis of biomass on cross-sectional area at breast height times height indicated a better fit.

Shelterbelt volume was chosen as the indicator of biomass for all shrub species. It is a relatively simple sampling method and the most accurate (Figure 5). The growth of these species in shelterbelts help explain the degree of fit between biomass and shelterbelt volume. Caragana soon fills in the area in which it was planted. Choke cherry grown in shelterbelts suckers readily and soon fills in, becoming a dense "rectangular" hedge with a volume determined by lenght of the shelterbelt, its height and width. Although villosa lilac does not sucker, it is a wide shrub which produces a dense growth. Buffaloberry and sea-buckthorn both sucker freely and form dense irregular hedges.

SUMMARY - Traditionally, the benefits of shelterbelts have been measured in terms of their ability to reduce erosion, control blowing snow, protect livestock and buildings and increase crop yields. Other benefits including farm diversification, wildlife habitat and farm yard beautification are also recognized. With the ability to quickly and easily calculate the biomass of shelterbelts we can now determine the value of mature shelterbelts in terms of their carbon holding potential.

This study has produced reliable models for determining the amount of above ground biomass, and the amount of fixed carbon, in a non-destructive manner (Table 8). The sampling models were selected on the criteria of accuracy and sampling simplicity. For deciduous and coniferous tree species, the above ground biomass can be calculated for any shelterbelt by measuring average diameter at breast height and determining the stem cross-sectional area of all stems. For the shrub species, shelterbelt volume is needed for above ground biomass calculation.

ABG = Above ground biomass (kg/tree).

X1 = Total stem cross-sectional area at breast height (cm²/tree).

X2 = Shelterbelt volume (m³) (Shelterbelt length(m) x width(m) x height(m)).

+ Only two sample points therefore little confidence in this value.

++ Indicates the accuracy of the equation, the closer it is to 1.0 the more accurate.

With this information, we can determine the environmental benefits of past prairie tree plantings and predict the benefits of future plantings in terms of their potential to sequester atmospheric carbon. We have identified prairie tree plantings as efficient methods of reducing existing atmospheric carbon. As governments and industry work towards reducing atmospheric carbon emissions, the need for effective means of reducing existing atmospheric carbon will increase. The natural ability of trees to act as carbon sinks and the Shelterbelt Centre's existing infrastructure and expertise in producing large numbers of trees for prairie plantings could result in an increased demand for trees in the future.