Poplar is one of the most widespread fast-growing forest species. In Northern Italy, plantations are characterized by large interannual fluctuations, requiring frequent monitoring to inform on wood supply and to manage the stands. The use of radar satellite data is proving useful for forest monitoring, being weather independent and sensitive to the changes in forest canopy structure, but it has been scarcely tested in the case of poplar. Here, L-band ALOS2 (Advanced Land Observing Satellite-2) dual-pol data were tested to detect clear-cut plantations in consecutive years. ALOS2 quad-pol data were used to discriminate among different age classes, a much complex task than detecting poplar plantations extent. Results from different machine learning algorithms indicate that with dual-pol data, poplar forest can be discriminated from clear-cut areas with 80% overall accuracy, similar to what is usually obtained with optical data. With quad-pol data, four age classes were classified with moderate overall accuracy (73%) based on polarimetric decompositions, three 3 age classes with higher accuracy (87%) based on HV band. Sources of error are represented by poplar areas of intermediate age when stems, branches and leaves were not developed enough to detect by scattering mechanisms. This study demonstrates the feasibility of monitoring poplar plantations with satellite radar, which represents a growing source of information thanks to already-planned future satellite missions. © 2022 by the authors.

The interest in tree phenology monitoring is increasing because this trait is a robust indicator of the impacts of climate change on natural and managed ecosystems. Different approaches to monitor phenology at different spatial scales, from in situ monitoring to remote sensing, are used to investigate spring and/or autumn phenological changes. In Mediterranean area, most of phenological changes occur during cloudy periods (spring and autumn), leading to a loss of information also for very high temporal resolution satellites. Instead, cloud-uninfluenced sensors, such as radar sensors, can allow to bypass this problem and produce a temporally continuous coverage. In this paper, we analyzed the spring phenology of two European beech (Fagus sylvatica L.) populations, located at different latitudes in Mediterranean area. Weekly in situ monitoring of leaf-out has been correlated with data collected by Synthetic Aperture Radar. Spring phenological phases were monitored in situ following a modified BBCH-code with a 5-scores scale (from 1 - buds closed and covered by scales, to 5 - leaf completely unfolded). The score 3 (young leaves starting to emerge from the bud) was considered the bud break. Different site conditions based on aspect (northern and southern) and altitudinal gradient (high and low altitude) have been considered. The aim was to test and implement a new methodology able to decrease the frequency of the field sampling, using remote data, to extend more detailed information on geographical scale, and to reconstruct past phenology. Results showed a statistically significant different length of the vegetative spring period, spanning from dormant buds, up to leaves completely unfolded, between sites. Through Synthetic Aperture Radar estimation, this study demonstrates that leaf-out can be monitored with an extreme accuracy. The phenophase score 4 and 5 estimation showed the best performance (RMSE < of 4 days), phenophases score 2 and 3 showed promising performances (4 days < RMSE <5 days), while phenophases score 1 seems to be not easily detectable, although it can be extrapolated with an RMSE <6 days. This radar approach fixes the cloud problem typical of multispectral approach and very frequent in phenophase change periods in Mediterranean climate. This study promotes the proposed remote sensing approach as a very useful tool to monitor growing season starting in remote areas, helping to reduce in situ observations and allowing past phenology reconstruction. © 2020 Elsevier Inc.