Carbon

SaskBarley supports the Saskatchewan Soil Conservation Association (SSCA), a non-profit, producer-based organization with a mission to “promote conservation production systems that improves the land and environment for future generations.” The Association’s motto is “Farming for your Future Environment.”

A major focus of the SSCA is facilitating the ongoing Prairie Soil Carbon Balance Project (PSCB), which monitored soil organic carbon change across Saskatchewan farms from 1996 to 2018. This project is a very important tool for Saskatchewan farmers in assessing our beneficial impact on carbon levels.

Read the PSCB 2018 report

Executive summary of the PSCB

There is great interest in the status of soil organic carbon (SOC) as an indicator of soil health and as a measure of removal of the greenhouse gas, carbon dioxide, from the atmosphere. Each ton of SOC in the soil represents a past removal of 3.67 tons of carbon dioxide from the atmosphere. There is a potential value of SOC as a greenhouse gas offset. Many large corporations want to reduce their carbon footprint and document their sustainability by sourcing from agricultural products from areas with increasing SOC.

The Prairie Soil Carbon Balance Project (PSCB) was initiated by the Saskatchewan Soil Conservation Association to establish a system to monitor SOC on commercial farm fields across Saskatchewan that were converted from conventional management to direct seeding and continuous cropping in 1997. Under this project a network of 136 commercial farm fields in Saskatchewan was established with initial measurement of SOC in fall 1996 with the plan to track SOC change with repeated samplings. A small benchmark, 16 x 7 ft, is in each field, locatable with a buried marker. Within each benchmark at each sampling, a composite sample of six soil cores was taken to 16 inches below the surface in 4-inch increments. For each sampling, a new set of six cores were taken offset from other samplings by 20 to 40 inches from the previous coring locations. The mass of SOC is estimated from analysed SOC concentration and the soil density. The network was resampled in 1999 (136 fields), 2005 (121 fields), 2011 (80 fields), and 2018 (90 fields). The number of fields differed between samplings due to changes with co-operators over the project duration.
An unexpected finding was there was massive spatial variability of SOC within the benchmark. Therefore, when the new six cores were taken at each sampling, by chance, the six cores could sample soils with high relative SOC or soils with low relative SOC. This creates a random difference between samplings due to spatial variability. With another sampling, the physical offset for the new six cores creates introduces another difference between measurements. It could increase the measured SOC or decrease the measured SOC depending on the SOC of the soil that happens to be included in the composite sample of six cores. Because of the spatial variability, it is impossible to reliably detect SOC change due to adoption of direct seeding from a single benchmark within a field.

An analogy to help understand the challenges of detecting the effect of adoption of direct seeding is the challenge of detecting “loaded” dice (i.e. dice weighted so some faces will show more likely than equal chance). Looking at the 4 different rolls of the dice, the numbers shown are true (not misread) with the total of each roll will vary between 2 and 12. For four successive rolls, the totals are variable and there is no way to relate the value of one roll with another. The fact that the dice are loaded cannot be detected from chance variation using just four rolls. The same holds for SOC from multiple samplings on one field. Looking at the results from four samplings on one field, each SOC value is true (no mistakes), but it is not possible to relate the SOC in one sampling with those for other samplings. It is not possible to detect any signal through the background SOC noise. However, if one carefully keeps track of the results for about 40 rolls of the dice, then it becomes apparent that the overall average value of the rolls is not consistent with what would be expected by chance alone. At that point, the signal from the loaded dice on the roll values is detected from the background noise of variable roll values. Similarly, looking at the average SOC results from at least 40 fields, then the SOC signal can be detected from the background noise of SOC variability. This SOC signal is the average effect of change in management of the sites.

The first major finding of the PSCB was that it is not possible to use one benchmark to reliably estimate SOC change due to the spatial variability of SOC. Nevertheless, the average differences for many fields in the PSCB network are reliable measures that can be detected from differences due to chance alone. To about a 1-ft depth, the change in SOC from 1996 was 0.76 ton/acre in 1999, 0.30 ton/acre in 2005, 1.35 ton/acre in 2011, and 1.07 ton/acre in 2018. The change to 2018 amounts to about 5% of initial SOC in 1996. Although these changes are modest, they conclusively show that SOC in increasing on direct seeded commercial farm fields in Saskatchewan. The smaller change from 1996 for 2005 sampling was related to the depressing effect on SOC of widespread droughts over the 2001-2003 period.
Another important finding was that SOC was increasing at deeper depths than expected based on research on small plots. For shallow soils with unaltered parent material (“C horizon”) within 11 inches of the surface, there was significant SOC change for upper 8 inches only. For deeper soils with unaltered parent material deeper than 11 inches below the soil surface, there was significant SOC change to 16 inches. Because measurement stopped at 16 inches, we do not know how much deeper significant SOC gain may have occurred.

A third finding was that the soils with the least SOC initially within the PSCB tended to gain the most, while the soils with the most SOC initially often lost some SOC over the project duration.

The measured changes from the PSCB generally agree with the estimates of SOC change that are contained in Canada’s national inventory of greenhouse gas emissions and removals.

We are indebted to the many farm cooperators who have made the PSCB project possible. We need the cooperators’ further assistance to collect data on the management history of the fields so that we can investigate the effect of management on SOC behaviour.

The PSCB project has provided some new and unique information about the behaviour of SOC on commercial farm fields throughout Saskatchewan. We now have confirmation that the fields are increasing in soil carbon and that has market value.