L‐Band InSAR Snow Water Equivalent Retrieval Uncertainty Increases With Forest Cover Fraction

There is a pressing need for global monitoring of snow water equivalent (SWE) at high spatiotemporal resolution, and L‐band (1–2 GHz) interferometric synthetic aperture radar (InSAR) holds promise. However, the technique has not seen extensive evaluation in forests. We evaluated this technique across varying forest canopy conditions using eight InSAR pairs collected at the Fraser Experimental Forest, Colorado, USA by NASA UAVSAR during the 10‐week NASA SnowEx 2021 Campaign. Compared with in situ measurements, we found root mean squared errors (RMSEs) of 14–17 mm for SWE changes in forest cover fractions (FCF) 0.50. Statistical distributions between normalized lidar snow depths and normalized UAVSAR SWE were similar at FCF < 0.5, but diverged at FCF > 0.50. Thus, the upcoming NISAR L‐band satellite has strong potential for global snowpack monitoring, including below sparse to moderate forest cover. Plain Language Summary Monitoring the amount of water stored in seasonal snowpacks is essential for water resource management, but it remains challenging, particularly in mountain and forest environments. Satellite radar techniques may provide a viable path forward for snowpack monitoring, particularly at longer radar wavelengths (>20 cm) such as the radar used for the upcoming NASA‐ISRO SAR satellite mission. At these longer wavelengths, the radar signal can penetrate forest canopy, but the canopy interferes with the signal and may reduce the accuracy of the radar snowpack measurement. We examined the influence of forest cover on airborne radar measurements of the snowpack in the Fraser Experimental Forest, Colorado, USA and observed errors that increased with greater forest cover. Notably, the radar measurements were accurate for sparse to moderately dense forest covers. The radar snowpack measurements reproduced elevational trends observed in lidar‐measured snow depths, but we identified snowpack measurement errors that correlated with oblique radar viewing geometries. Considering these limitations, we conclude that the NASA‐ISRO SAR satellite mission represents a promising path toward global snowpack monitoring. Key Points We evaluated L‐band interferometry for snowpack monitoring in a montane forest in Colorado Retrievals of changes in snow water equivalent from L‐band interferometry were accurate and unbiased in forest cover fractions