Detecting peatland drains with Object Based Image Analysis and Geoeye-1 imagery

Background

Peatlands play an important role in the global carbon cycle. They provide important ecosystem services including carbon sequestration and storage. Drainage disturbs peatland ecosystem services. Mapping drains is difficult and expensive and their spatial extent is, in many cases, unknown. An object based image analysis (OBIA) was performed on a very high resolution satellite image (Geoeye-1) to extract information about drain location and extent on a blanket peatland in Ireland. Two accuracy assessment methods: Error matrix and the completeness, correctness and quality (CCQ) were used to assess the extracted data across the peatland and at several sub sites. The cost of the OBIA method was compared with manual digitisation and field survey. The drain maps were also used to assess the costs relating to blocking drains vs. a business-as-usual scenario and estimating the impact of each on carbon fluxes at the study site.

Results

The OBIA method performed well at almost all sites. Almost 500 km of drains were detected within the peatland. In the error matrix method, overall accuracy (OA) of detecting the drains was 94% and the kappa statistic was 0.66. The OA for all sub-areas, except one, was 95–97%. The CCQ was 85%, 85% and 71% respectively. The OBIA method was the most cost effective way to map peatland drains and was at least 55% cheaper than either field survey or manual digitisation, respectively. The extracted drain maps were used constrain the study area CO₂ flux which was 19% smaller than the prescribed Peatland Code value for drained peatlands.

Conclusions

The OBIA method used in this study showed that it is possible to accurately extract maps of fine scale peatland drains over large areas in a cost effective manner. The development of methods to map the spatial extent of drains is important as they play a critical role in peatland carbon dynamics. The objective of this study was to extract data on the spatial extent of drains on a blanket bog in the west of Ireland. The results show that information on drain extent and location can be extracted from high resolution imagery and mapped with a high degree of accuracy. Under Article 3.4 of the Kyoto Protocol Annex 1 parties can account for greenhouse gas emission by sources and removals by sinks resulting from “wetlands drainage and rewetting”. The ability to map the spatial extent, density and location of peatlands drains means that Annex 1 parties can develop strategies for drain blocking to aid reduction of CO₂ emissions, DOC runoff and water discoloration. This paper highlights some uncertainty around using one-size-fits-all emission factors for GHG in drained peatlands and re-wetting scenarios. However, the OBIA method is robust and accurate and could be used to assess the extent of drains in peatlands across the globe aiding the refinement of peatland carbon dynamics .

     

3-D geochemical interpolation guided by geophysical inversion models

3-D geochemical subsurface models, as constructed by spatial interpolation of drill-core assays, are valuable assets across multiple stages of the mineral industry's workflow.However, the accuracy of such models is limited by the spatial sparsity of the underlying drill-core, which samples only a small fraction of the subsurface.This limitation can be alleviated by integrating collocated 3-D models into the interpolation process, such as the 3-D rock property models produced by modern geophysical inversion procedures, provided that they are sufficiently resolved and correlated with the interpolation target.While standard machine learning algorithms are capable of predicting the target property given these data, incorporating spatial autocorrelation and anisotropy in these models is often not possible.We propose a Gaussian process regression model for 3-D geochemical interpolation, where custom kernels are introduced to integrate collocated 3-D rock property models while addressing the trade-off between the spatial proximity of drill-cores and the similarities in their collocated rock properties, as well as the relative degree to which each supporting 3-D model contributes to interpolation.The proposed model was evaluated for 3-D modelling of Mg content in the Kevitsa Ni-Cu-PGE deposit based on drill-core analyses and four 3-D geophysical inversion models.Incorporating the inversion models improved the regression model's likelihood(relative to a purely spatial Gaussian process regression model) when evaluated at held-out test holes, but only for moderate spatial scales(100 m).     

Seasonal vegetation and management influence overland flow velocity and roughness in upland grasslands

There is considerable interest in how headwater management may influence downstream flood peaks in temperate humid regions. However, there is a dearth of data on flow velocities across headwater hillslopes and limited understanding of whether surface flow velocity is influenced by seasonal changes in roughness through vegetation cycles or management. A portable hillslope flume was used to investigate overland flow velocities for four common headwater grassland habitats in northern England: Low-density Grazing, Hay Meadow, Rank Grassland and Juncus effusus Rush pasture. Overland flow velocity was measured in replicate plots for each habitat, in response to three applied flow rates, with the experiments repeated during five different periods of the annual grassland cycle. Mean annual overland flow velocity was significantly lower for the Rank Grassland habitat (0.026 m/s) followed by Low-density Grazing and Rushes (0.032 and 0.029 m/s), then Hay Meadows (0.041 m/s), which had the greatest mean annual velocity (examples from 12 L/min flow rate). Applying our mean overland flow velocities to a theoretical 100 m hillslope suggests overland flow is delayed by >1 hr on Rank Grassland when compared to Hay Meadows in an 18 mm storm. Thus grassland management is important for slowing overland flow and delaying peak flows across upland headwaters. Surface roughness was also strongly controlled by annual cycles of vegetation growth, decay, grazing and cutting. Winter overland flow velocities were significantly higher than in summer, varying between 0.004 m/s (Rushes, November) and 0.034 m/s (Rushes, June); and velocities significantly increased after cutting varying between 0.006 m/s (Hay meadows, July) and 0.054 m/s (Hay meadows, September). These results show that seasonal vegetation change should be incorporated into flood modelling, as cycles of surface roughness in grasslands strongly modify overland flow, potentially having a large impact on downstream flood peak and timing. Our data also showed that Darcy-Weisbach roughness approximations greatly over-estimated measured flow velocities.     

The impact of ditch blocking on the hydrological functioning of blanket peatlands

Ditch blocking in blanket peatlands is common as part of peatland restoration. The effects of ditch blocking on flow regimes and nearby water tables were examined in a field trial. After an initial 6‐month monitoring period, eight ditches had peat dams installed 10 m apart along their entire length (dammed), four of these ditches were also partially infilled through bank reprofiling (reprofiled). Four ditches were left open with no dams or reprofiling (open). These 12 ditches and the surrounding peat were monitored for 4 more years. An initial five‐fold reduction in discharge occurred in the dammed and the reprofiled ditches with the displaced water being diverted to overland flow and pathways away from the ditches. However, there was a gradual change over time in ditch flow regime in subsequent years, with the overall volume of water leaving the dammed and the reprofiled ditches increasing per unit of rainfall to around twice that which occurred in the first year after blocking. Hence, monitoring for greater than one year is important for understanding hydrological impacts of peatland restoration. Overland flow and flow in the upper ~4 cm of peat was common and occurred in the inter‐ditch areas for over half of the time after ditch blocking. There was strong evidence that topographic boundaries of small ditch catchments, despite being defined using a high‐resolution Light Detection And Ranging‐based terrain model, were not always equivalent to actual catchment areas. Hence, caution is needed when upscaling area‐based fluxes, such as aquatic carbon fluxes, from smaller scale studies including those using ditches and small streams. The effect of ditch blocking on local water tables was spatially highly variable but small overall (time‐weighted mean effect <2 cm). Practitioners seeking to raise water tables through peatland restoration should first be informed either by prior measurement of water tables or by spatial modelling to show whether the peatland already has shallow water tables or whether there are locations that could potentially undergo large water‐table recoveries.     

Water‐level dynamics in natural and artificial pools in blanket peatlands

Perennial pools are common natural features of peatlands, and their hydrological functioning and turnover may be important for carbon fluxes, aquatic ecology, and downstream water quality. Peatland restoration methods such as ditch blocking result in many new pools. However, little is known about the hydrological function of either pool type. We monitored six natural and six artificial pools on a Scottish blanket peatland. Pool water levels were more variable in all seasons in artificial pools having greater water level increases and faster recession responses to storms than natural pools. Pools overflowed by a median of 9 and 54 times pool volume per year for natural and artificial pools, respectively, but this varied widely because some large pools had small upslope catchments and vice versa. Mean peat water‐table depths were similar between natural and artificial pool sites but much more variable over time at the artificial pool site, possibly due to a lower bulk specific yield across this site. Pool levels and pool‐level fluctuations were not the same as those of local water tables in the adjacent peat. Pool‐level time series were much smoother, with more damped rainfall or recession responses than those for peat water tables. There were strong hydraulic gradients between the peat and pools, with absolute water tables often being 20–30 cm higher or lower than water levels in pools only 1–4 m away. However, as peat hydraulic conductivity was very low (median of 1.5 × 10^?5 and 1.4 × 10^?6 cm s^?1 at 30 and 50 cm depths at the natural pool site), there was little deep subsurface flow interaction. We conclude that (a) for peat restoration projects, a larger total pool surface area is likely to result in smaller flood peaks downstream, at least during summer months, because peatland bulk specific yield will be greater; and (b) surface and near‐surface connectivity during storm events and topographic context, rather than pool size alone, must be taken into account in future peatland pool and stream chemistry studies.     

Well Conditioning in a Fluvial Reservoir Model

This paper describes a method for conditioning an object model of a fluvial reservoir on facies observations. The channels are assumed parametrized at sections normal to their main channel direction. Projections of the observations on these sections generates a map suitable for drawing conditioning values. This map contains the information from every facies observation between two adjacent sections, enabling handling of any well path. Coupling between well observations is also discussed. The methodology is implemented and demonstrated in examples with complex wells.