An automated method to quantify crop height and calibrate satellite-derived biomass using hypertemporal lidar

作者：Eitel, Jan U. H.^*; Magney, Troy S.; Vierling, Lee A.; Greaves, Heather E.; Zheng, Guang

来源：Remote Sensing of Environment, 2016, 187: 414-422.

DOI：10.1016/j.rse.2016.10.044

• Autonomously operating terrestrial laser scanner(ATLS)
• Leaf area index (LAI)
• Hypertemporal lidar
• Crop biomass
• Biomass scaling
• Error accounting
• Laser allometry

Crop biomass information is of great importance for a variety of applications, ranging from supporting farm management decisions to modeling the crop-environment system. Dimensionless spectral vegetation index values derived from satellite imagery are commonly used to derive crop biomass. However, the highly empirical nature of spectrally derived biomass estimates requires frequent and costly calibration with manually collected ground data. Recently, low cost, autonomously operating terrestrial laser scanners (ATLSs) have become available for near-surface applications. In contrast to the dimensionless nature of spectral index values, autonomous light detection and ranging (lidar) technology measures physical vegetation structure by recording the x, y, z coordinates of canopy components at very high spatial (<10 cm), and temporal (<2 days) resolution. The objective of this study was to assess the suitability of an ATLS to i) monitor crop growth dynamics and ii) calibrate satellite imagery for estimating crop biomass. Wheat (Triticum aestivum spp.) growth was monitored by acquiring hypertemporal (every 28 h for a full growing season) ATLS data at three different field sites across a range of experimentally manipulated crop growth conditions. The ATLS-derived crop height explained nearly three-quarters of the variability in destructively sampled wheat biomass (r(2) = 0.74, RMSE = 514.20 kg ha(-1)), showing a slightly stronger correlation to crop biomass than did leaf area index (LAI) measurements collected in the field using a LAI-2000 Plant Canopy Analyzer (r(2) = 0.71, RMSE = 546 kg ha(-1)). Satellite-based crop biomass estimates calibrated with ATLS data captured the variability in wheat biomass throughout a farm field with a biomass error of 730.96 and 727.60 kg ha(-1) (RMSE) during the jointing (Development stage: Zadoks 37) and heading (Development stage: Zadoks 50) growth stages, respectively. These findings suggest that the hypertemporal lidar information provided via ATLS technology could constitute a major step forward in operational monitoring and mapping of crop biomass.

• 出版日期2016-12-15
• 单位南京大学