Wide-area mapping of small-scale features in agricultural landscapes using airborne remote sensing

Abstract

Natural and semi-natural habitats in agricultural landscapes are likely to come under increasing pressure with the global population set to exceed 9 billion by 2050. These non-cropped habitats are primarily made up of trees, hedgerows and grassy margins and their amount, quality and spatial configuration can have strong implications for the delivery and sustainability of various ecosystem services. In this study high spatial resolution (0.5 m) colour infrared aerial photography (CIR) was used in object based image analysis for the classification of non-cropped habitat in a 10,029 ha area of southeast England. Three classification scenarios were devised using 4 and 9 class scenarios. The machine learning algorithm Random Forest (RF) was used to reduce the number of variables used for each classification scenario by 25.5 % ± 2.7%. Proportion of votes from the 4 class hierarchy was made available to the 9 class scenarios and where the highest ranked variables in all cases. This approach allowed for misclassified parent objects to be correctly classified at a lower level. A single object hierarchy with 4 class proportion of votes produced the best result (kappa 0.909). Validation of the optimum training sample size in RF showed no significant difference between mean internal out-of-bag error and external validation. As an example of the utility of this data, we assessed habitat suitability for a declining farmland bird, the yellowhammer (Emberiza citronella), which requires hedgerows associated with grassy margins. We found that ∼22% of hedgerows were within 200 m of margins with an area >183.31 m2. The results from this analysis can form a key information source at the environmental and policy level in landscape optimisation for food production and ecosystem service sustainability.

Quantifying and modelling decay in forecast proficiency indicates the limits of transferability in land‐cover classification

“…Ten sites were selected in the UK (Figure a) based on access to ground validation data from previous studies (O'Connell et al., ; Relu ). Sites varied in size from 1,197 to 15,136 ha and were located in intensively‐managed agricultural landscapes which were primarily dominated by arable crops and permanent grassland (Figure c).…”

“…We focus on the transferability of H/LC models in ten agricultural sites located in central and eastern England, UK. Providing high spatial resolution maps within UK agricultural landscape is of great importance with respect to habitat fragmentation, diversity and sustainability (Benton, Vickery, & Wilson, ; O'Connell, Bradter, & Benton, ). The coverage and spatial distribution of land‐cover classes in intensively managed agricultural landscapes can have direct and indirect consequences on ecosystem service delivery and sustainability (Benton, ).…”

“…We focus on the transferability of H/LC models in ten agricultural sites located in central and eastern England, UK. Providing high spatial resolution maps within UK agricultural landscape is of great importance with respect to habitat fragmentation, diversity and sustainability (Benton, Vickery, & Wilson, 2003;O'Connell, Bradter, & Benton, 2015).…”

Effects of urban sprawl on arthropod communities in peri-urban farmed landscape in Shenbei New District, Shenyang, Liaoning Province, China

“…These non-cropped habitats are primarily comprised of trees, hedgerows, and grassy margins and their amount, quality, and spatial configuration can have strong implications for the delivery and sustainability of various ecosystem services 13–15 . The presence of a complex landscape with a high proportion of non-cropped habitat is more beneficial than uniform landscapes for maintaining biodiversity, and the focus of attention and research is currently on the effects on biodiversity from the proportion 16 and diversification 17 of non-cropped habitats.…”

Spatio-temporal Analysis of Agricultural Landscape Images: A Superpixel-based Approach

“…A large number of attributes were extracted for each image-object, including spectral, spatial, and textural features. Other works on the same line have been made by Galletti andMyint (2014), andO'Connell et al (2015).…”