Skip booklet index and go to page content

Technical Thematic Report No. 15. -Trends in residual soil nitrogen for agricultural land in Canada, 1981 2006

Residual Soil Nitrogen Indicator

The annual Residual Soil Nitrogen (RSN) values presented in this report were determined with a Canadian Agricultural Nitrogen Budget model (CANB v3.0) which estimates all of the nitrogen inputs (nitrogen fertilizer addition, manure addition, legume nitrogen fixation, and atmospheric deposition) and outputs (crop removal, ammonia volatilization, and denitrification) and then calculates the difference between the nitrogen inputs and nitrogen outputs (Yang et al., 2007; Yang et al., 2011). The RSN Indicator provides an estimate of the amount of “unused” nitrogen that remains in the soil at the end of the cropping season. The RSN values were calculated on a soil landscape of Canada (SLC) polygon basis and these values were then scaled up to an ecozone+ and national level to give an estimate of RSN per hectare (agricultural land area weighted average). The SLC polygons are mapping units which range in size from 10,000 ha to 1 million ha and contain soils with similar properties, landscape attributes, as well as climate (Lefebvre et al., 2005).

Census of Agriculture data (farm area, livestock types and numbers, and crop types and areas), fertilizer sales, crop yield and climatic data, and soil landscape information (such as, soil types and slopes) were the primary data sources used in the model (Bourque and Koroluk, 2003; Beaulieu, 2004; Agriculture and Agri-Food Canada and Statistics Canada, 2008). Published coefficients were also used to estimate both inputs and outputs from agricultural soils (ASABE, 2005). For example, manure production was based on the number and type of livestock raised in the region (for example, dairy, hog). However, it was also necessary to estimate manure nitrogen losses via ammonia volatilization and denitrification based on manure form (solid versus liquid), storage system (six liquid manure and seven solid manure storage systems), as well as application time and method (four application periods and three incorporation methods). Hence, the estimated manure nitrogen input is the amount of manure nitrogen that was applied to agricultural land after volatilization losses, denitrification losses, and storage losses were taken into account. The mineralization of organic nitrogen from manure and legume crop residues was estimated for the current year, as well as for the 2nd and 3rd years after application.

Risk classes based on the RSN level present in the soil at the end of the growing season (very low risk 0 to 9.9 kg N/ha; low risk 10 to 19.9 kg N/ha; moderate risk 20 to 29.9 kg N/ha; high risk 30 to 39.9  kg N/ha; and very high risk >40 kg N/ha) were assigned to farmland (Drury et al., 2010). The area of land in each risk class was mapped for the agricultural ecozones+ in Canada (Figure 1).

Figure 1. Residual soil nitrogen (RSN) risk classes for agricultural land in Canada, 2006.


Long Description for Figure 1

This map of Canada shows residual soil nitrogen (RSN) risk classes for agricultural land in 2006. Risk classes are based on the RSN level present in the soil at the end of the growing season (very low risk 0 to 9.9 kg N/ha; low risk 10 to 19.9 kg N/ha; moderate risk 20 to 29.9 kg N/ha; high risk 30 to 39.9 kg N/ha; and very high risk >40 kg N/ha). The map shows that farmland in interior British Columbia is mostly in the low risk category. The prairies are generally in the very low risk category with risk increasing north into the Boreal Plains ecozone+ and around the Lake Winnipeg region. Farmland around the Great Lakes, along the St. Lawrence and into the Atlantic Maritime ecozone+ are predominantly in the very high risk category, with an exception for the region northeast of Lake Heron to the Quebec border, which falls into the moderate risk category.


Top of Page

Model limitations

The resolution of the RSN model is the SLC polygon scale. Although the amount of fertilizer and manure nitrogen that is applied to soils can be estimated for each SLC polygon based on fertilizer sales and on the livestock type and numbers, we do not know the proportion of each nitrogen source that is applied to a particular crop in the polygon. Thus assumptions had to be made to reconcile crop nitrogen requirements with the amount of fertilizer nitrogen sold and manure produced in any given polygon (Huffman et al., 2008; Yang et al., 2011).

Most of the crop yield data was based on 60 Canadian agricultural regions, while other crop yields, such as for alfalfa and pasture, are based on provincial data. However, it would be preferable to have yield estimates based on each SLC polygon instead of the larger Canadian agricultural regions. Additional information concerning the assumptions used in the model for ammonia volatilization, denitrification gas partitioning, and others was reported by Yang et al. (2007).

Management practices such as conservation tillage, cover crops, and conversion from annual to perennial crops may have increased soil organic matter levels in some regions which could have tied up RSN as soil organic nitrogen. Conversely, soil organic nitrogen levels may decline in soils where land in perennial crops was converted to annual crops and/or when changes in management practices resulted in reduced crop yields and carbon inputs (for example, shift from mixed livestock farms to cash crop farms). We were unable to account for changes in soil organic matter levels across regions and over time in this model. Hence soil organic matter levels were assumed to be at steady state.

Top of Page