[Paper Review] Reviewing the role of the extinction coefficient in radar remote sensing

J. David Ballester‐Berman|arXiv (Cornell University)|Dec 4, 2020

Synthetic Aperture Radar (SAR) Applications and Techniques51 references4 citations

TL;DR

This paper investigates the role of the extinction coefficient in radar backscattering models for forest monitoring, using a physically based RVoG model to show that backscatter exhibits an increasing-decreasing trend with vegetation height due to wave extinction. The key finding is that saturation in backscatter is not absolute, and sensitivity beyond the peak may allow extended height estimation if extinction is accurately characterized.

ABSTRACT

This report revisits the role of the extinction coefficient in radar backscattering-based models for forest monitoring. A review of a number of works dealing with this issue has revealed a diversity of extinction values being unclear its dependence on the sensor frequency and the forest type. In addition, a backscattering model directly derived from the RVoG formulation is employed to analyse the saturation of backscattering level as a function of vegetation height and the presence of a decreasing trend of backscatter beyond the saturation point as suggested in previous works in the literature. According to this analysis it seems reasonable to think that further research specially focused on dedicated experimental measurements of the extinction coefficient should be carried out.

Motivation & Objective

• To assess the inconsistent use and interpretation of the extinction coefficient in radar backscattering models for forest monitoring.
• To investigate whether the observed backscatter saturation in forests is truly irreversible, or if a decreasing trend beyond the peak could enable extended height estimation.
• To evaluate the physical basis of the extinction coefficient in radar-vegetation interaction, especially in relation to sensor frequency and forest type.
• To advocate for dedicated experimental measurements of extinction to resolve ambiguity in current models.

Proposed method

• A review of literature on extinction coefficient values across different forest types and radar frequencies (P-, L-, C-band) was conducted.
• The RVoG (Random Volume over Ground) model was adapted to a backscattering-only formulation by removing interferometric diversity and assuming dominant volume and double-bounce scattering components.
• Theoretical backscattering response was simulated as a function of vegetation height using the RVoG model, with extinction coefficient as a key parameter.
• Parameter estimation was performed using regression on backscatter data, with sensitivity analysis on extinction and ground-to-volume ratio.
• The model was tested under both asymptotic and non-asymptotic conditions, comparing results with empirical data and prior studies.
• Theoretical predictions were compared with observed backscatter trends in tropical and temperate forests, particularly focusing on the post-saturation decline.

Experimental results

Research questions

• RQ1 Does the extinction coefficient vary systematically with forest type and radar frequency, or is it treated arbitrarily as a fitting parameter?
• RQ2 Can the observed backscatter saturation in forest SAR data be truly irreversible, or is there residual sensitivity beyond the peak?
• RQ3 To what extent does the RVoG-based backscattering model predict a decreasing trend in backscatter beyond the saturation point due to increasing wave extinction?
• RQ4 Can the extinction coefficient be reliably estimated from empirical data, and does this improve the accuracy of forest height and biomass retrieval?

Key findings

• The extinction coefficient is inconsistently reported across studies, with no clear dependence on frequency or forest type, suggesting a lack of standardized measurement or physical understanding.
• The RVoG-based backscattering model predicts a peak-and-decline behavior in backscatter as vegetation height increases, driven by increasing wave extinction.
• Backscatter reaches a maximum at a specific vegetation height that depends on the extinction coefficient and ground-to-volume ratio, indicating that saturation is not absolute.
• The sensitivity of backscatter to height beyond the saturation point is weak but theoretically present, suggesting potential for extended height estimation if extinction is accurately known.
• Theoretical results align with experimental observations from Mermoz et al. (2015) and Yu & Saatchi (2016), who reported a decreasing trend in L-band HV backscatter beyond saturation in tropical forests.
• Despite theoretical plausibility, practical implementation faces challenges due to data uncertainties, limiting the operational utility of post-saturation sensitivity.