Browsing by Author "Waine, Toby William"

Now showing 1 - 3 of 3
  • Thumbnail Image
    ItemOpen Access
    Continuous isoprene measurements in a UK temperate forest for a whole growing season: effects of drought stress during the 2018 heatwave
    (American Geophysical Union (AGU), 2020-07-08) Ferracci, Valerio; Bolas, Conor G.; Freshwater, Ray A.; Staniaszek, Zosia; King, Thomas; Jaars, Kerneels; Otu‐Larbi, Frederick; Beale, John; Malhi, Yadvinder; Waine, Toby William; Jones, Roderic L.; Ashworth, Kirsti; Harris, Neil
    Isoprene concentrations were measured at four heights below, within and above the forest canopy in Wytham Woods (UK) throughout the summer of 2018 using custom-built gas chromatographs (the iDirac). These observations were complemented with selected ancillary variables, including air temperature, photosynthetically active radiation (PAR), occasional leaf gas exchange measurements and satellite retrievals of normalized difference vegetation and water indices (NDVI and NDWI). The campaign overlapped with a long and uninterrupted heatwave accompanied by moderate drought. Peak isoprene concentrations during the heatwave-drought were up to a factor of 4 higher than those before or after. Higher temperatures during the heatwave could not account for all the observed isoprene; the enhanced abundances correlated with drought stress. Leaf-level emissions confirmed this and also included compounds associated with ecosystem stress. This work highlights that a more in-depth understanding of the effects of drought stress is required to better characterize isoprene emissions.
  • No Thumbnail Available
    ItemOpen Access
    Non-Invasive soil property measurement for precision farming
    (Cranfield University, 1999-10-01) Waine, Toby William; Blackmore, S.
    This work investigates the application of new sensors to enable agronomists and farm managers to make decisions for variable treatment strategies at key crop growth stages. This is needed to improve the efficiency of crop production in the context of precision farming. Two non-invasive sensors were selected for investigation. These were: 1) The MGD-1 ion mobility gas detector made by Environics OY, Finland. 2) The EM38 electromagnetic induction (EMI) sensor made by Geonics Inc., Canada. The gas detector was used to determine residual nitrogen and to measure carbon dioxide gas as a surrogate indicator of soil quality. In the latter, increased microbial carbon dioxide production was expected on soils with high organic matter content. Overall, the results of gas detection were disappointing. The main problems inherent in the system were; lack of control of the gas sampling, insufficient machine resolution and cross contamination. This led to the decision to discontinue the gas detection research. Instead, the application of electromagnetic induction (EMI) to measure soil variation was investigated. There were two principle advances in the research. Firstly the application of EMI to the rapid assessment of soil textural class. Secondly the mapping of available water content in the soil profile. These were achieved through the development of a new calibration procedure based on EMI survey of the sites at field capacity, working with field experiments from five sites over two years. Maps of total available water holding capacity were produced. These were correlated with yield maps from wet and dry seasons and used to explain some of the seasonal influences on the spatial variation in yield. A product development strategy for a new EMI sensor was considered which produced a recommendation to design a new EMI sensor specifically for available water content and soil texture mapping, that could be mounted on a tractor. For the first time, this procedure enables routine monitoring of the spatial variation in available water content. This enables the effects of seasonal and spatial variation to be included in crop models, targeted irrigation and to aid decisions for the variable application of inputs.
  • Thumbnail Image
    ItemOpen Access
    Replacing human interpretation of agricultural land in Afghanistan with a deep convolutional neural network
    (Taylor and Francis, 2021-01-18) Hamer, Alex M.; Simms, Daniel M.; Waine, Toby William
    Afghanistan’s annual opium survey relies upon time-consuming human interpretation of satellite images to map the area of potential poppy cultivation for statistical sample design. Deep Convolutional Neural Networks (CNNs) have shown ground-breaking performance for image classification tasks by encoding local contextual information, in some cases outperforming trained analysts. In this study, we investigate the development of a CNN to automate the classification of agriculture from medium-resolution satellite imagery as an alternative to manual interpretation. The residual network (ResNet50) CNN architecture was trained and validated for delineating the agricultural area using labelled multi-seasonal Disaster Monitoring Constellation (DMC) satellite imagery (32 m) of Helmand and Kandahar provinces. The effect of input image chip size, training sampling strategy, elevation data, and multi-seasonal imagery were investigated. The best-performing single-year classification used an input chip size of 33 × 33 pixels, a targeted sampling strategy and transfer learning, resulting in high overall accuracy (94%). The inclusion of elevation data marginally lowered performance (93%). Multi-seasonal classification achieved an overall accuracy of 89% using the previous two years’ data. Only 25% of the target year’s training samples were necessary to update the model to achieve >94% overall accuracy. A data-driven approach to automate agricultural mask production using CNNs is proposed to reduce the burden of human interpretation. The ability to continually update CNN models with new data has the potential to significantly improve automatic classification of vegetation across years