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

Remote sensing of hedgerows, windbreaks, and winter cover crops in California's Central Coast reveals low adoption but hotspots of use

Non-crop vegetation, such as hedgerows and cover crops, are important on-farm diversification practices that support biodiversity and ecosystem services; however, information about their rates and patterns of adoption are scarce. We used satellite and aerial imagery coupled with machine learning classification to map the use of hedgerows/windbreaks and winter cover crops in California's Central Coast, a globally important agricultural area of intensive fresh produce production. We expected that adoption of both practices would be relatively low and unevenly distributed across the landscape, with higher levels of adoption found in marginal farmland and in less intensively cultivated areas where the pressure to remove non-crop vegetation may be lower. Our remote sensing classification revealed that only ~6% of farmland had winter cover crops in 2021 and 0.26% of farmland had hedgerows or windbreaks in 2018. Thirty-seven percent of ranch parcels had cover crops on at least 5% of the ranch while 22% of ranches had at least one hedgerow/windbreak. Nearly 16% of farmland had other annual winter crops, some of which could provide services similar to cover crops; however, 60% of farmland had bare soil over the winter study period, with the remainder of farmland classified as perennial crops or strawberries. Hotspot analysis showed significant areas of adoption of both practices in the hillier regions of all counties. Finally, qualitative interviews revealed that adoption patterns were likely driven by interrelated effects of topography, land values, and farming models, with organic, diversified farms implementing these practices in less ideal, lower-value farmland. This study demonstrates how remote sensing coupled with qualitative research can be used to map and interpret patterns of important diversification practices, with implications for tracking policy interventions and targeting resources to assist farmers motivated to expand adoption.

Large-Scale Debris Cover Glacier Mapping Using Multisource Object-Based Image Analysis Approach

Large-scale debris cover glacier mapping can be efficiently conducted from high spatial resolution (HSR) remote sensing imagery using object-based image analysis (OBIA), which works on a group of pixels. This paper presents the spectral and spatial capabilities of OBIA to classify multiple glacier cover classes using a multisource approach by integrating multispectral, thermal, and slope information into one workflow. The novel contributions of this study are effective mapping of small yet important geomorphological features, classification of shadow regions without manual corrections, discrimination of snow/ice, ice-mixed debris, and supraglacial debris without using shortwave infrared bands, and an adaptation of an area-weighted error matrix specifically built for assessing OBIA’s accuracy. The large-scale glacier cover map is produced with a high overall accuracy of ≈94% (area-weighted error matrix). The proposed OBIA approach also proved to be effective in mapping minor geomorphological features such as small glacial lakes, exposed ice faces, debris cones, rills, and crevasses with individual class accuracies in the range of 96.9–100%. We confirm the portability of our proposed approach by comparing the results with reference glacier inventories and applying it to different sensor data and study areas.

Forest Road Detection Using LiDAR Data and Hybrid Classification

Knowledge about forest road networks is essential for sustainable forest management and fire management. The aim of this study was to assess the accuracy of a new hierarchical-hybrid classification tool (HyClass) for mapping paved and unpaved forest roads with LiDAR data. Bare-earth and low-lying vegetation were also identified. For this purpose, a rural landscape (area 70 ha) in northwestern Spain was selected for study, and a road network map was extracted from the cadastral maps as the ground truth data. The HyClass tool is based on a decision tree which integrates segmentation processes at local scale with decision rules. The proposed approach yielded an overall accuracy (OA) of 96.5%, with a confidence interval (CI) of 94.0–97.6%, representing an improvement over pixel-based classification (OA = 87.0%, CI = 83.7–89.8%) using Random Forest (RF). In addition, with the HyClass tool, the classification precision varied significantly after reducing the original point density from 8.7 to 1 point/m2. The proposed method can provide accurate road mapping to support forest management as an alternative to pixel-based RF classification when the LiDAR point density is higher than 1 point/m2.

Classifying grass-dominated habitats from remotely sensed data: The influence of spectral resolution, acquisition time and the vegetation classification system on accuracy and thematic resolution

Detailed maps of vegetation facilitate spatial conservation planning. Such information can be difficult to map from remotely sensed data with the detail (thematic resolution) required for ecological applications. For grass-dominated habitats in the South-East of the UK, it was evaluated which of the following choices improved classification accuracies at various thematic resolutions: 1) Hyperspectral data versus data with a reduced spectral resolution of eight and 13 bands, which were simulated from the hyperspectral data. 2) A vegetation classification system using a detailed description of vegetation (sub)-communities (the British National Vegetation Classification, NVC) versus clustering based on the dominant plant species (Dom-Species). 3) The month of imagery acquisition. Hyperspectral data produced the highest accuracies for vegetation away from edges using the NVC (84-87%). Simulated 13-band data performed also well (83-86% accuracy). Simulated 8-band data performed poorer at finer thematic resolutions (77-78% accuracy), but produced accuracies similar to those from simulated 13-band or hyperspectral data for coarser thematic resolutions (82-86%). Grouping vegetation by NVC (84-87% accuracy for hyperspectral data) usually achieved higher accuracies compared to Dom-Species (81-84% for hyperspectral data). Highest discrimination rates were achieved around the time vegetation was fully developed. The results suggest that using a detailed description of vegetation (sub)-communities instead of one based on the dominating species can result in more accurate mapping. The NVC may reflect differences in site conditions in addition to differences in the composition of dominant species, which may benefit vegetation classification. The results also suggest that using hyperspectral data or the 13-band multispectral data can help to achieve the fine thematic resolutions that are often required in ecological applications. Accurate vegetation maps with a high thematic resolution can benefit a range of applications, such as species and habitat conservation.

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

Peri-urban farmland provides a diversity of ecological services. However, it is experiencing increasing pressures from urban sprawl. While the effects of land use associated with farming on arthropod assemblages has received increasing attention, most of this research has been conducted by comparing conventional and organic cropping systems. The present study identifies the effects of urban sprawl and the role of non-cropped habitat in defining arthropod diversity in peri-urban farmed landscapes. Multi-scale arthropod data from 30 sampling plots were used with linear-mixed models to elucidate the effects of distance from urban areas (0–13 km; 13–25 km and >25 km, zones I, II, and III, respectively) on arthropods. Results showed that urban sprawl, disturbed farm landscapes, and disturbance in non-cropped habitats had negative effects on arthropods, the latter requiring arthropods to re-establish annually from surrounding landscapes via dispersal. While arthropod species richness showed no obvious changes, arthropod abundance was lowest in zone II. Generally, patch density (PD), Shannon diversity index (SHDI), and aggregate index (AI) of non-cropped habitat were major drivers of changes in arthropod populations. This study contributes to identifying the effects of urban sprawl on arthropod diversity and documenting the multiple functions of farm landscapes in peri-urban regions.