Mapping soil carbon stocks across Scotland using a neural network model

Aitkenhead, Matt; Coull, Malcolm

doi:10.1016/j.geoderma.2015.08.034

Cited by 64 publications

(33 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Second, ANN provides excellent modeling capabilities for complex, noisy environmental datasets where the relationship between input and output parameters is not well understood. Finally, NNs are frequently considered "black box" models because extracting the information they develop is easier on the modeled system than with other approaches [71]. Thus, the approach is convenient to use.…”

Section: Artificial Neural Network (Ann) Modelmentioning

confidence: 99%

Prediction of Soil Organic Carbon based on Landsat 8 Monthly NDVI Data for the Jianghan Plain in Hubei Province, China

et al. 2019

View full text Add to dashboard Cite

High-precision maps of soil organic carbon (SOC) are beneficial for managing soil fertility and understanding the global carbon cycle. Digital soil mapping plays an important role in efficiently obtaining the spatial distribution of SOC, which contributes to precision agriculture. However, traditional soil-forming factors (i.e., terrain or climatic factors) have weak variability in low-relief areas, such as plains, and cannot reflect the spatial variation of soil attributes. Meanwhile, vegetation cover hinders the acquisition of the direct information of farmland soil. Thus, useful environmental variables should be utilized for SOC prediction and the digital mapping of such areas. SOC has an important effect on crop growth status, and remote sensing data can record the apparent spectral characteristics of crops. The normalized difference vegetation index (NDVI) is an important index reflecting crop growth and biomass. This study used NDVI time series data rather than traditional soil-forming factors to map SOC. Honghu City, located in the middle of the Jianghan Plain, was selected as the study region, and the NDVI time series data extracted from Landsat 8 were used as the auxiliary variables. SOC maps were estimated through stepwise linear regression (SLR), partial least squares regression (PLSR), support vector machine (SVM), and artificial neural network (ANN). Ordinary kriging (OK) was used as the reference model, while root mean square error of prediction (RMSE P ) and coefficient of determination of prediction (R 2 P ) were used to evaluate the model performance. Results showed that SOC had a significant positive correlation in July and August (0.17, 0.29) and a significant negative correlation in January, April, and December (−0.23, −0.27, and −0.23) with NDVI time series data. The best model for SOC prediction was generated by ANN, with the lowest RMSE P of 3.718 and highest R 2 P of 0.391, followed by SVM (RMSE P = 3.753, R 2 P = 0.361) and PLSR (RMSE P = 4.087, R 2 P = 0.283). The SLR model was the worst model, with the lowest R 2 P of 0.281 and highest RMSE P of 3.930. ANN and SVM were better than OK (RMSE P = 3.727, R 2 P = 0.372), whereas PLSR and SLR were worse than OK. Moreover, the prediction results using single-data NDVI or short time series NDVI showed low accuracy. The effect of the terrain factor on SOC prediction represented unsatisfactory results. All these results indicated that the NDVI time series data can be used for SOC mapping in plain areas and that the ANN model can maximally extract additional associated information between NDVI time series data and SOC. This study presented an effective method to overcome the selection of auxiliary variables for digital soil mapping in plain areas when the soil was covered with vegetation. This finding indicated that the time series characteristics of NDVI were conducive for predicting SOC in plains.

show abstract

Section: Artificial Neural Network (Ann) Modelmentioning

confidence: 99%

Prediction of Soil Organic Carbon based on Landsat 8 Monthly NDVI Data for the Jianghan Plain in Hubei Province, China

et al. 2019

View full text Add to dashboard Cite

show abstract

“…In addition, fjord sediments hold a greater quantity of C than all the living vegetation in Scotland (Forestry Commission, 2015;Henrys et al, 2016;Vanguelova et al, 2013). While Scotland's soils (Aitkenhead and Coull, 2016), particularly the peatlands (Chapman et al, 2009), contain a greater quantity of OC than the fjords, it must be remembered that the fjord sediments also hold IC and the areal extent of these stores differs greatly. When normalised by area (Fig.…”

Section: Comparison To Other Mid-latitude Carbon Stocks: Significancementioning

confidence: 99%

Scotland's forgotten carbon: a national assessment of mid-latitude fjord sedimentary carbon stocks

et al. 2017

View full text Add to dashboard Cite

Abstract. Fjords are recognised as hotspots for the burial and long-term storage of carbon (C) and potentially provide a significant climate regulation service over multiple timescales. Understanding the magnitude of marine sedimentary C stores and the processes which govern their development is fundamental to understanding the role of the coastal ocean in the global C cycle. In this study, we use the mid-latitude fjords of Scotland as a natural laboratory to further develop methods to quantify these marine sedimentary C stores on both the individual fjord and national scale. Targeted geophysical and geochemical analysis has allowed the quantification of sedimentary C stocks for a number of mid-latitude fjords and, coupled with upscaling techniques based on fjord classification, has generated the first full national sedimentary C inventory for a fjordic system. The sediments within these mid-latitude fjords hold 640.7 ± 46 Mt of C split between 295.6±52 and 345.1 ± 39 Mt of organic and inorganic C, respectively. When compared, these marine mid-latitude sedimentary C stores are of similar magnitude to their terrestrial equivalents, with the exception of the Scottish peatlands, which hold significantly more C. However, when areanormalised comparisons are made, these mid-latitude fjords are significantly more effective as C stores than their terrestrial counterparts, including Scottish peatlands. The C held within Scotland's coastal marine sediments has been largely overlooked as a significant component of the nation's natural capital; such coastal C stores are likely to be key to understanding and constraining improved global C budgets. Highlights-Scottish fjords are a more effective store of C than the terrestrial environment.-A total of 640.7 ± 46 Mt of C is stored in the sediment of Scotland's 111 fjords.-An estimated 31 139-40 615 t yr −1 of C is buried in the sediment of Scotland's fjords.-Fjord sediments are potentially the most effective store of C globally.

show abstract

“…The main sources of uncertainty are peat depth and bulk density, for which there are not enough values measured from field survey to provide a robust statistical overview of Scotland's peats, and the extent of peatland, which varies according to the definition used to map it. The most recent estimates are closer to each other: 813 Mt C for the first meter of depth according to Aitkenhead and Coull (2016) and 1620 Mt C for the first two meters of depth according to Chapman et al (2009).…”

Section: Carbon Stockmentioning

confidence: 68%

“…In this work, we used carbon stock data obtained with a Neural Network approach (NN), which considers soil, topographic and climatic parameters, and combines them with loss on ignition, bulk density and peat depth data. These data (Aitkenhead, in prep) are the update of Aitkenhead and Coull (2016), and they consider the whole peat depth (instead of just the first meter). The resolution of this dataset is 100 × 100 m.…”

Section: Carbon Stockmentioning

confidence: 99%

Potential carbon loss from Scottish peatlands under climate change

et al. 2019

View full text Add to dashboard Cite

The Scottish Government is committed to reduce carbon emissions by 80% by 2050 (compared to a 1990-1995 baseline). Peatlands have been recognised as a key environment for the carbon balance as they sequester and store great quantities of carbon, but they also have the potential to release it. In Scotland, peatlands cover more than 20% of the surface (more than 90% of which is blanket bog) and store more than 2500 Mt of carbon. Blanket bogs are very climate reliant, and as a consequence of climate change, many areas in Scotland may not be able to support peatlands in the near future. In this study, two bioclimatic envelope models (Linsday Modified model and Blanket Bog Tree model) have been used to obtain a first estimate of how the distribution of blanket bogs in Scotland could vary according to climate change in the 2050s and in the 2080s. The potential losses of carbon arising from climate change have then been calculated. Results showed that in 2050, more than half of the carbon currently stored in Scottish blanket bogs will be at risk of loss. This is 4.4-6.6 times the amount of carbon emitted in 2016 from all the sectors in Scotland and, if emissions from peatland occur and are taken into account, it will greatly hamper efforts to meet emission reduction targets set out in the Climate Change (Scotland) Act of 2009.

show abstract

Mapping soil carbon stocks across Scotland using a neural network model

Cited by 64 publications

References 45 publications

Prediction of Soil Organic Carbon based on Landsat 8 Monthly NDVI Data for the Jianghan Plain in Hubei Province, China

Prediction of Soil Organic Carbon based on Landsat 8 Monthly NDVI Data for the Jianghan Plain in Hubei Province, China

Scotland's forgotten carbon: a national assessment of mid-latitude fjord sedimentary carbon stocks

Potential carbon loss from Scottish peatlands under climate change

Contact Info

Product

Resources

About