Search

The monitoring of forest ecosystems is significantly affected by the lack of consistent historical data of low-severity (forest partially disturbed) or gradual disturbance (e.g. eastern spruce budworm epidemic). The goal of this paper is to explore the use of a subset of Landsat time series and deep learning models to identify both the type and the year of disturbances, including low-severity and gradual disturbances, in the boreal forest of eastern Canada at the pixel level. Remote sensing data such as the spectral information from Landsat time series are the best available option for large scale observations of disturbances that go back decades. Traditional modeling approaches, like LandTrendr, require substantial handcrafted pre-processing to remove noise and to extract temporal features from the image sequences before using them as input to a classical machine-learning model. Deep-learning models can autonomously discern which features are relevant within the coarse temporal and spectral information from the Landsat annual dense time series. We evaluated the performance of TempCNN and Transformer model in detecting and classifying the type and the year of the forest disturbance using Landsat time series subsequences. Our findings resulted in the generation of four disturbance maps outlining the forest history from 1986 to 2021 within the eastern Canadian boreal forest. Our experimental outcomes demonstrate several significant benefits of employing deep learning models. Firstly, using noisy Landsat time series they achieve comparable accuracy for classifying fire and total harvesting than existing publicly available disturbance maps. Secondly, the use of shorter time series subsequence with deep learning models enables to map adequately different overlapping disturbances occurring in the complete time series. Finally, they increase the number of distinguishable disturbance classes by adding partial harvesting, gradual disturbances, and forest recovery from older events, making them useful approaches for obtaining the first remote sensing-based map for areas affected by the eastern spruce budworm. • TempCNN, Transformer tested on Landsat time series for disturbance classification. • The type and the year of disturbance events were classified with good accuracy. • Deep learning effective for identifying partial and progressive disturbances. • Landsat subsequences extract overlapping disturbances. [ABSTRACT FROM AUTHOR]

This paper investigates the potential for the Soil Moisture Active Passive (SMAP) L-band radiometer to estimate the percentage of frozen soil inside a pixel during fall periods. To evaluate the spatial and temporal variability of the autumn freeze in northeastern Canada boreal forest, a network of compact and self-recording temperature sensors (iButton) along transects averaging 10 km was deployed on the soil surface at two different sites. Results show important spatial variability in soil freezing timing with maximums, for the two sites, spanning from 7.5 to 9.5 weeks to reach a frozen state. It is known that L-band radiometry such as Soil Moisture Active Passive (SMAP) can monitor freeze/thaw (F/T) seasonality at low spatial resolution (~40 km). However, the possible strong surface variability within a 40 km × 40 km pixel can lead to spatial delay in fall freezing. Simulations of brightness temperature (TB) weighted by spatially distributed iButton F/T records and using the ω-τ vegetation model show good agreement with SMAP satellite observations during the fall transition period with an average RMSE of 3.5 K. A new intra-pixel frozen soil percentage retrieval algorithm using only SMAP observations and air temperature is proposed. The percentage of frozen soil in falls 2015 and 2016 from the new algorithm and the in situ measurements are in good agreement with a coefficient of determination (R2) ranging between 0.63 and 0.88. • Substantial spatial variability of soil freezing timing in boreal forest • L-band SMAP radiometer sensitivity to spatial variability of soil freezing • A new SMAP algorithm to estimate the intra-pixel fraction of frozen soil [ABSTRACT FROM AUTHOR]