Estimation of forage biomass and vegetation cover in grasslands using UAV imagery

Abstract

Grasslands are among the most widespread ecosystems on Earth and among the most degraded. Their characterization and monitoring are generally based on field measurements, which are incomplete spatially and temporally. The recent advent of unmanned aerial vehicles (UAV) provides data at unprecedented spatial and temporal resolutions. This study aims to test and compare three approaches based on multispectral imagery acquired by UAV to estimate forage biomass or vegetation cover in grasslands. The study site is composed of 30 pasture plots (25 × 50 m), 5 bare soil plots (25 x 50), and 6 control plots (5 × 5 m) on a 14-ha field maintained at various biomass levels by grazing rotations and clipping over a complete growing season. A total of 14 flights were performed. A first approach based on structure from motion was used to generate a volumetric-based biomass estimation model (R2 of 0.93 and 0.94 for fresh biomass [FM] and dry biomass [DM], respectively). This approach is not very sensitive to low vegetation levels but is accurate for FM estimation greater than 0.5 kg/m2 (0.1 kg DM/m2). The Green Normalized Difference Vegetation Index (GNDVI) was selected to develop two additional approaches. One is based on a regression biomass prediction model (R2 of 0.80 and 0.66 for FM and DM, respectively) and leads to an accurate estimation at levels of FM lower than 3 kg/m2 (0.6 kg DM/m2). The other approach is based on a classification of vegetation cover from clustering of GNDVI values in four classes. This approach is more qualitative than the other ones but more robust and generalizable. These three approaches are relatively simple to use and applicable in an operational context. They are also complementary and can be adapted to specific applications in grassland characterization.

The Application of an Unmanned Aerial System and Machine Learning Techniques for Red Clover-Grass Mixture Yield Estimation under Variety Performance Trials

“…Other previous studies have indicated that NDVI is more commonly used for pasture biomass measurements [88,89], as well as in larger-scale grassland followed by seasonal monitoring [90]. However, the findings of this study indicate that NDVI may not be the most suitable VI, which was supported by a previous study [86]. This highlights the importance of considering the saturation, sensitivity, stages of crop development, canopy structure, and the type of environment when testing various vegetation indexes [15].…”

“…Different VI values at the level of leaf area indices were likely caused by the diverse canopy structures of clover (horizontal) and grass leaves (vertically orientated) [71]. A recent study confirmed that GNDVI is suitable as a biomass predictor for perennial forage crops, where R 2 = 0.80 for freshly-cut, and 0.66 for dry yields [86], as well as in the grain yield estimation in maize [87]. These results resemble the RFR and ANN modelling of this study, where the GNDVI, GDVI, and MSR had the highest average contributions.…”

Smart Sensor Monitoring in Energy Crop Biomass Production

“…All three models allow fine spatial resolution. In this case, the use of RGB imagery gave better agreement with the data collected by conventional methods of sampling and characterization of biomass compared to volumetric measurement because it allows to distinguish between living and dead biomass (ie photosynthetically active and dried plants) [11]. Such models are suitable for monitoring the spatial distribution of biomass but are not sufficiently usable in relation to the time scale.…”

“…Figure 5 -Comparison of the representation of the same experimental field of sorghum by different models: a -RGB, b -volumetric model, c -GNDVI for biomass,d -GNDVI for vegetation cover[11] …”

Short-Term Recovery of the Aboveground Carbon Stock in Iberian Shrublands at the Extremes of an Environmental Gradient and as a Function of Burn Severity

“…In this sense, LiDAR sensors on-board unmanned aerial vehicles (UAV-LiDAR) allows for acquiring higher point density than traditional airborne LiDAR [128,129]. Simultaneous acquisition of UAV multispectral imagery (UAV-MS) at very high spatial resolution also allows for improving the classification of ground/non-ground returns of UAV-LiDAR data in areas with low vegetation and rough topography [95,130] and discriminating dead standing and lying biomass [131,132]. Therefore, UAV consumer-grade technology may enhance ACD estimation in shrubland ecosystems and should be considered in future research.…”