[Optimized Spectral Indices Based Estimation of Forage Grass Biomass]

Area of Science:

• Agricultural Science
• Remote Sensing
• Ecology

Background:

• Aboveground biomass is a key indicator of forage production and grassland management.
• Traditional biomass measurement methods are labor-intensive and time-consuming.
• Hyperspectral remote sensing offers a non-destructive approach for real-time biomass monitoring.

Purpose of the Study:

• To evaluate the effectiveness of existing and optimized spectral indices for estimating forage grass biomass.
• To compare the performance of spectral indices across different grassland types and treatments.
• To develop improved methods for accurate and efficient biomass estimation.

Main Methods:

• Field experiments were conducted in desert steppe (natural pasture) and artificial forage fields.
• Canopy reflectance and aboveground biomass were measured under varying grazing densities and nitrogen rates.
• Published and optimized spectral indices were analyzed for their predictive ability.

Main Results:

• Canopy reflectance varied significantly based on forage species, canopy structure, and biomass.
• The performance of spectral indices differed across species and treatments, with some losing sensitivity under specific conditions.
• Optimized spectral indices derived from combined datasets showed significantly improved biomass prediction (R² = 0.72).

Conclusions:

• Optimizing spectral indices by combining data from natural and artificial grasslands enhances biomass estimation accuracy.
• Optimized indices demonstrated superior predictive ability and lower noise compared to published indices.
• Waveband optimization presents a promising algorithm for improving forage grass biomass prediction using hyperspectral remote sensing.