Climate change, fire return intervals and the growing risk of permanent forest loss in boreal Eurasia.

Climate change has driven an increase in the frequency and severity of fires in Eurasian boreal forests. A growing number of field studies have linked the change in fire regime to post-fire recruitment failure and permanent forest loss. In this study we used four burned area and two forest loss datasets to calculate the landscape-scale fire return interval (FRI) and associated risk of permanent forest loss. We then used machine learning to predict how the FRI will change under a high emissions scenario (SSP3-7.0) by the end of the century. We found that there are currently 133,000 km2 forest at high, or extreme, risk of fire-induced forest loss, with a further 3 M km2 at risk by the end of the century. This has the potential to degrade or destroy some of the largest remaining intact forests in the world, negatively impact the health and economic wellbeing of people living in the region, as well as accelerate global climate change.

Detection of oil pollution impacts on vegetation using multifrequency SAR, multispectral images with fuzzy forest and random forest methods.

Oil pollution harms terrestrial ecosystems. There is an urgent requirement to improve on existing methods for detecting, mapping and establishing the precise extent of oil-impacted and oil-free vegetation. This is needed to quantify existing spill extents, formulate effective remediation strategies and to enable effective pipeline monitoring strategies to identify leakages at an early stage. An effective oil spill detection algorithm based on optical image spectral responses can benefit immensely from the inclusion of multi-frequency Synthetic Aperture Radar (SAR) data, especially when the effect of multi-collinearity is sufficiently reduced. This study compared the Fuzzy Forest (FF) and Random Forest (RF) methods in detecting and mapping oil-impacted vegetation from a post spill multispectral optical sentinel 2 image and multifrequency C and X Band Sentinel - 1, COSMO Skymed and TanDEM-X SAR images. FF and RF classifiers were employed to discriminate oil-spill impacted and oil-free vegetation in a study area in Nigeria. Fuzzy Forest uses specific functions for the selection and use of uncorrelated variables in the classification process to yield an improved result. This method proved an efficient variable selection technique addressing the effects of high dimensionality and multi-collinearity, as the optimization and use of different SAR and optical image variables generated more accurate results than the RF algorithm in densely vegetated areas. An Overall Accuracy (OA) of 75% was obtained for the dense (Tree Cover Area) vegetation, while cropland and grassland areas had 59.4% and 65% OA respectively. However, RF performed better in Cropland areas with OA = 75% when SAR-optical image variables were used for classification, while both methods performed equally well in Grassland areas with OA = 65%. Similarly, significant backscatter differences (P < 0.005) were observed in the C-Band backscatter sample mean of polluted and oil-free TCA, while strong linear associations existed between LAI and backscatter in grassland and TCA. This study demonstrates that SAR based monitoring of petroleum hydrocarbon impacts on vegetation is feasible and has high potential for establishing oil-impacted areas and oil pipeline monitoring.

Impact of fertiliser, water table, and warming on celery yield and CO2 and CH4 emissions from fenland agricultural peat.

Peatlands are globally important areas for carbon preservation; although covering only 3% of global land area, they store 30% of total soil carbon. Lowland peat soils can also be very productive for agriculture, but their cultivation requires drainage as most crops are intolerant of root-zone anoxia. This leads to the creation of oxic conditions in which organic matter becomes vulnerable to mineralisation. Given the demand for high quality agricultural land, 40% of the UK's peatlands have been drained for agricultural use. In this study we present the outcomes of a controlled environment experiment conducted on agricultural fen peat to examine possible trade-offs between celery growth (an economically important crop on the agricultural peatlands of eastern England) and emissions of greenhouse gases (carbon dioxide (CO2) and methane (CH4)) at different temperatures (ambient and ambient +5 °C), water table levels (-30 cm, and -50 cm below the surface), and fertiliser use. Raising the water table from -50 cm to -30 cm depressed yields of celery, and at the same time decreased the entire ecosystem CO2 loss by 31%. A 5 °C temperature increase enhanced ecosystem emissions of CO2 by 25% and increased celery dry shoot weight by 23% while not affecting the shoot fresh weight. Fertiliser addition increased both celery yields and soil respiration by 22%. Methane emissions were generally very low and not significantly different from zero. Our results suggest that increasing the water table can lower emissions of greenhouse gases and reduce the rate of peat wastage, but reduces the productivity of celery. If possible, the water table should be raised to -30 cm before and after cultivation, and only decreased during the growing season, as this would reduce the overall greenhouse gas emissions and peat loss, potentially not affecting the production of vegetable crops.

Mapping terrestrial oil spill impact using machine learning random forest and Landsat 8 OLI imagery: a case site within the Niger Delta region of Nigeria.

Terrestrial oil pollution is one of the major causes of ecological damage within the Niger Delta region of Nigeria and has caused a considerable loss of mangroves and arable croplands since the discovery of crude oil in 1956. The exact extent of landcover loss due to oil pollution remains uncertain due to the variability in factors such as volume and size of the oil spills, the age of oil, and its effects on the different vegetation types. Here, the feasibility of identifying oil-impacted land in the Niger Delta region of Nigeria with a machine learning random forest classifier using Landsat 8 (OLI spectral bands) and Vegetation Health Indices is explored. Oil spill incident data for the years 2015 and 2016 were obtained from published records of the National Oil Spill Detection and Response Agency and Shell Petroleum Development Corporation. Various health indices and spectral wavelengths from visible, near-infrared, and shortwave infrared bands were fused and classified using the machine learning random forest classifier to distinguish between oil-free and oil spill-impacted landcover. This provided the basis for the identification of the best variables for discriminating oil polluted from unpolluted land. Results showed that better results for discriminating oil-free and oil polluted landcovers were obtained when individual landcover types were classified separately as opposed to when the full study area image including all landcover types was classified at once. Similarly, the results also showed that biomass density plays a significant role in the characterization and classification of oil contaminated and oil-free pixels as tree cover areas showed higher classification accuracy compared to cropland and grassland.

Next generation of elevated [CO2] experiments with crops: a critical investment for feeding the future world.

A rising global population and demand for protein-rich diets are increasing pressure to maximize agricultural productivity. Rising atmospheric [CO(2)] is altering global temperature and precipitation patterns, which challenges agricultural productivity. While rising [CO(2)] provides a unique opportunity to increase the productivity of C(3) crops, average yield stimulation observed to date is well below potential gains. Thus, there is room for improving productivity. However, only a fraction of available germplasm of crops has been tested for CO(2) responsiveness. Yield is a complex phenotypic trait determined by the interactions of a genotype with the environment. Selection of promising genotypes and characterization of response mechanisms will only be effective if crop improvement and systems biology approaches are closely linked to production environments, that is, on the farm within major growing regions. Free air CO(2) enrichment (FACE) experiments can provide the platform upon which to conduct genetic screening and elucidate the inheritance and mechanisms that underlie genotypic differences in productivity under elevated [CO(2)]. We propose a new generation of large-scale, low-cost per unit area FACE experiments to identify the most CO(2)-responsive genotypes and provide starting lines for future breeding programmes. This is necessary if we are to realize the potential for yield gains in the future.