Quality assessment of ICESat-2 ATL03 denoising data based on spatial distribution and photon weight

Abstract. The Ice, Cloud, and land Elevation satellite-2 (ICESat-2) adopts an advanced single photon detection system, which is susceptible to the influence of solar background noise. There are currently multiple methods for denoising ICESat-2 photon data, but due to the availability and limitations of reference data, it is difficult to evaluate the denoising results of large-scale photon data. Therefore, in this paper, we propose a quality evaluation method that does not rely on auxiliary data. By considering the spatial distribution and photon weight, it was determined whether the signal and noise photons were extracted correctly, and the precision (P), recall (R), and F-measure (F) of the denoising results were evaluated. The results showed that for the strong beam and weak beam data obtained during the day and night, the average difference between the evaluation results of this method and the true label were 0.0177, 0.0794 and 0.0018, 0.0042, respectively. In general, the quality evaluation results of this study are very close to those of the real labels, indicating that the method in this paper can effectively evaluate the quality of the denoising results.