Abstract
Abstract. Grassland represents the largest single agricultural vegetation in Germany and provides a multitude of ecosystem services. Timely and accurate information about herbage yield and quality is essential for an efficient use of resources, e.g. to be able to match the actual available feed with a demand of animals or with other industrial uses. Grasslands frequently exhibit small-scale botanical and structural heterogeneity with pronounced spatio-temporal dynamics. These features present particular challenges for sensor applications, which, apart from limitations posed by high costs and low temporal and spatial resolutions of many available remote sensing (RS) systems, may be the reason for so far little commercial applications of RS in practical grassland farming. This paper considers recent developments in the use of spectral and point-cloud data for herbage yield and quality assessment of grasslands. Former research showed that single sensor systems mounted on unmanned aerial vehicles produce similar prediction errors in crude protein or acid detergent fibre concentrations as proximal sensing tools (e.g. field spectroscopy). However, further improvements are needed. Beside improvements of single sensor types, the development of systems with complementary sensors is seen as a promising research area. It will help to overcome the limitations of single sensors and provide better information about herbage yield and quality. From an agronomic point of view, thematic maps of farm fields are suggested as the central outcome of RS and data analysis. These maps are representing the relevant grassland features and therefore can be used as low-cost, appropriate and timely information to support farmers’ decision-making.
Highlights
In Europe, grasslands represent approximately 30 to 35 % of the total agricultural area (Huyghe et al, 2014)
This paper reports recent findings on the potentials of point cloud and spectral data, either obtained by proximal measurements or with sensors based on unmanned aerial vehicles (UAV)
This paper considers recent developments in using spectral and point-cloud data for herbage yield and quality assessment of grasslands
Summary
In Europe, grasslands represent approximately 30 to 35 % of the total agricultural area (Huyghe et al, 2014). To achieve representative data for large areas, high number of field measurements is needed It is labour-intensive and time consuming, on sites that are remote or difficult to access. Information collected by RS systems has different features, such as the spectral, radiometric, spatial, and temporal resolution, and sensors used for biomass and quality estimation vary substantially in these characteristics. Most studies involving herbage yield and quality estimation from RS data have used a single sensor or single-date image, which may not be sufficient for applications in heterogeneous areas or grasslands with high botanical and structural diversity. Since RS data are available from a range of sensors, each with its own characteristics, a combination of sensors may be beneficial in terms of providing better information on the observed stand
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
