Developing a methodology to improve soil C stock estimates: report

6. Conclusions

The work in this project included a number of different objectives, all aimed at improving national estimates of soil C and soil C change. The focus of the geostatistical analysis is to improve estimates of peat depths and total soil C stocks in Scottish peats using more targeted sampling approaches. The exploration of GPR and LIDAR assesses the potential of new methods to measure peat depth and monitor changes in soil C stocks in the peatlands of Scotland. The data derived from the NSIS_2 resampling is used to improve the accuracy of the ECOSSE model and its estimates of C change given a range of land use and climate change scenarios. This project provides a significant contribution to improved estimates of C stocks and C change in Scottish soils.

The geostatistical analysis of variation in peat depth suggests that more work is needed on the scales of variation in peat. A greater number of peatland examples should be examined over a wider geographical area to identify and separate regional components of spatial variation, which may be due to regional climatic differences, topographic factors or other short range factors which influence hydrology. Effort should be made to include a study of the polygon boundaries between peaty and mineral soils, using data on the presence and absence of a peaty layer.

The analysis of technologies available for improved determination of peat depth and C content, indicates GPR technology has been used much more in peat studies than LIDAR. Ground penetrating radar could be used on its own to measure peat depth whereas LIDAR could not, but LIDAR perhaps offers the better option for measuring changes in C stocks in peat. One clear advantage of GPR is that it can provide a continuous assessment of peat depth along a transect compared to the intermittent measurements achieved by probing. Most small GPR devices currently available can be operated by one or two personnel, but are not very suitable for use on uneven terrain, being wheeled devices that are cumbersome to use and may cause damage to the surface vegetation. Accessing sites on foot and measuring depths with depthing rods is likely to cause less damage to the site. There are a number of bogs with detailed depth measurements (eg. Glensaugh, Smith et al., 2007b) where testing a GPR device would provide an indication of the accuracy of the technique in measuring peat depth compared to traditional methods. This would indicate whether the results obtained would significantly impact the C stock estimates of the study area.

By contrast, research using LIDAR to measure the C content of peats is still at the experimental phase. LIDAR appears to have potential as a tool to monitor development of peat gullies in areas of peat erosion, but can only achieve height accuracies in the order of +/- 0.15m and with a horizontal accuracy of 1-2 metres. LIDAR has the capacity to generate very accurate representations of the ground surface, and when combined with depth to the peat bottom in a GIS, 3D models of the peat resource can be generated. A comparison of peat volumes calculated using depth data combined with the bog area to those obtained using LIDAR would indicate whether LIDAR adds sufficient value for it to be used more widely.

The study suggests that it is possible to make a prediction of peat depth in areas where no measurements are available using an analysis of the spatial statistics for peat depth. This relies on distance to the nearest known values and clearly in areas where little data exists, the variance about the prediction increases. The expert judgement used in the original estimation of peat depths across the country similarly relies upon the depth in adjacent peat polygons where these are known. For this reason, there is broad agreement between the two approaches, but there was some consistent deviation, particularly at the deeper end where the expert system gave deeper peats. This is attributed to the expert system using additional information in its judgement. We conclude that the expert system gives a fair prediction for most peats and that the similarity in mean values indicates little difference in overall C stocks computed using either approach.

To undertake a comprehensive targeted peat sampling programme will be time consuming and costly and should be considered against the benefits ensuing from collecting such data. However the data collected during the current NSIS_2 resampling programme will provide information on the thickness and depth of horizons, which combined with bulk density analysis and % C values, will allow C stocks for the soil profiles sampled to be calculated, enhancing the national soil datasets and the associated variability studies will also give an insight into the range of values around a point. The NSIS_2 work will also provide information on the thickness of organic horizons within organo-mineral soils and using hand held augers, the thickness of peat deposits at each site can be assessed to 2 metre depth, again enhancing previous information and C stocks. However to transport actual peat depthing rods to each site in conjunction with the required NSIS_2 equipment, is not thought a feasible proposition within current budgets and any additional sampling of peat below 1 metre should be undertaken as a separate exercise. Such an exercise would allow the known uneven distribution of data, both across the country (east to west) and between peat types (basin and blanket) to be built into the sampling programme.

The evaluation of the use of archived dry bulk density values for peat bogs to determine C stocks values retrospectively suggests that a pedotransfer function may be derived that enables dry bulk density to be estimated from peat moisture content. A restriction is that the peat samples should be from the saturated zone or at least have a wet bulk density in excess of 0.85. Since we are here interested in bulk density changes at depth, this restriction does not interfere too much with its application.

Simulations have been run at all 62 sites included in the first phase of the NSIS_2 analysis. The simulated values show a high degree of association with the measurements in both total C and change in C content of the soil. The uncertainty in the simulations is 20% of the average C content over all land use types, uncertainty increasing in the order natural/semi-natural < forestry < arable < grassland. Over all sites where land use change has occurred, the average deviation between the simulated and the measured values of percentage change in soil C is less than the experimental error (11% simulation error, 535% measurement error). Simulated values are within the 95% confidence interval of the 1:1 line between simulated and measured values, so the simulated values are within experimental error with respect to correlation. Only a small bias in the simulations compared to the measured values is observed, suggesting that a small underestimate of the change in soil C should be expected in the national simulation (-4%). A large proportion of the uncertainty is associated with uncertainties in the input data: these include uncertainties in

1. timing of land use change,

2. actual management of arable land, grassland and forestry, and

3. land use change before the start of the simulation.

There is potential to greatly decrease the uncertainty in the national simulations by developing algorithms to estimate of the likely rate of C accumulation or loss at the start of the simulation. Improved estimates of uncertainty at the national scale could also be achieved by sub-sampling all land use units within >20 x 1km ² grid squares across the country and repeating the simulations of changes in C content done here. This would require changes to the long term soil sampling strategies.

The national simulations using ECOSSE have been improved during this project by the extensive developments in the model and by using improved data to drive the simulations. Data required for the national simulations have been collated from a number of different sources and prepared into files that can be used to drive the ECOSSE model. ECOSSE has been adapted to use the new data, to improve the initialisation of the model and to incorporate improved descriptions of processes. Comparison of estimates of changes in soil C stocks by CEH and ECOSSE for 1990-1999 and 2000-2009 show a highly significant correlation (P < 0.001), and mean values estimated by the two approaches that are not significantly different in either decade. This provides confidence in the results because these two very different approaches provide broadly similar results. The estimates of changes in soil C made by ECOSSE on land use change semi-natural to grassland show a larger loss of soil C than estimated by the CEH method. Similarly, ECOSSE estimates a larger gain in soil C on land use change grassland to semi-natural. Correspondingly, a larger loss of soil C is estimated by ECOSSE than by the CEH method for the land use change forestry to semi-natural and a smaller gain in soil C for the land use change semi-natural to forestry. These differences occur due to processes that are included in ECOSSE but not in the CEH approach associated with soil disturbance and reduced plant inputs when vegetation is immature. Carbon losses due to soil disturbance and an initial reduction in the plant input result in a loss of soil C or a smaller gain than estimated by CEH for land use changes between semi-natural land and grassland or forestry. A literature review provides evidence for the need to include soil disturbance and reduced plant inputs in simulations. In 2000-2009, the ECOSSE estimate of the annual change in soil C stocks for Scotland is -810 kt year ^-1 (differing from CEH estimates by -68 kt year ^-1).

The results suggest that increasing the area of land use change from arable to grass and grass to semi-natural is likely to sequester more C, and decreasing the area of grass to arable and forestry to semi-natural is likely to reduce losses of soil C. However, note that these simulations have assumed no difference in the drainage status of the soil when converted to semi-natural; if the soil becomes less drained, a different pattern of changes in soil C would be expected.

Across Scotland, the simulated change in soil C from organic soils (defined here as soils with a C content of over 6%; Smith et al., 2007b) between 1950 and 2009 is -62512 kt, compared to -35351 kt from mineral soils; losses from organic soils are 64% of the total soil C losses. This emphasizes the importance of organic soils in any national estimates of greenhouse gas emissions, and the need to avoid C emitting land use changes on these soils. These results also suggest that mitigation options to reduce losses of soil C might recommend different policies for land use change on mineral and organic soils.

Climate change alone is predicted to result in a decline in the soil C stocks of only -93 to -125 kt between 1990 and 2060 (-1.3 to -1.8 kt year ^-1). The largest changes in soil C are seen under unmanaged semi-natural land, -38 to -60 kt between 1990 and 2060 (-0.5 to -0.9 kt year ^-1). If plant inputs are not maintained, as may be the case especially in semi-natural land uses, the losses of soil C may be greater than predicted. Projected changes in land use result in losses of soil C stocks that are nearly 50 times greater than the losses due to climate change. This illustrates the potential for C losses due to climate change to be mitigated by changing land use.

Four mitigation options have been identified with high potential for achieving zero net losses of C from Scottish soils:

1. Decrease in the rate of conversion of grassland to arable to 28% of the current rate;

2. Stop conversion of semi-natural land to arable or grassland and increase the conversion of grassland to semi-natural by 125% of the current rate;

3. Stop conversion of semi-natural land to arable or grassland and increase the conversion of arable to grassland by 63% of the current rate; and

4. Stop conversion of semi-natural land to arable or grassland and decrease the conversion of grassland to arable to 77% of the current rate.

Possible ways of incentivising these changes may come about through improving implementation of existing good agricultural and environmental condition ( GAEC) standards for soil protection, through strengthening rural development policy, or though redesign of CAP to encourage the maintenance of existing C stocks (Frelih-Larsen et al., 2008).