The Airborne Hyperspectral Scanner (AHS) and the Hyperion satellite hyperspectral sensors were evaluated for their ability to predict topsoil organic carbon (C) in burned mountain areas of northwestern Spain slightly covered by heather vegetation. Predictive models that estimated total organic C (TOC) and oxidizable organic C (OC) content were calibrated using two datasets: a ground observation dataset with 39 topsoil samples collected in the field (for models built using AHS data), and a dataset with 200 TOC/OC observations predicted by AHS (for models built using Hyperion data). For both datasets, the prediction was performed by stepwise multiple linear regression (SMLR) using reflectances and spectral indices (SI) obtained from the images, and by the widely-used partial least squares regression (PLSR) method. SMLR provided a performance comparable to or even better than PLSR, while using a lower number of channels. SMLR models for the AHS were based on a maximum of eight indices, and showed a coefficient of determination in the leave-one-out cross-validation R² = 0.60–0.62, while models for the Hyperion sensor showed R² = 0.49–0.61, using a maximum of 20 indices. Although slightly worse models were obtained for the Hyperion sensor, which was attributed to its lower signal-to-noise ratio (SNR), the prediction of TOC/OC was consistent across both sensors. The relevant wavelengths for TOC/OC predictions were the red region of the spectrum (600–700 nm), and the short wave infrared region between ~2000–2250 nm. The use of SMLR and spectral indices based on reference channels at ~1000 nm was suitable to quantify topsoil C, and provided an alternative to the more complex PLSR method.

The study of the thermal behavior of permafrost and active layer on the South Shetland Islands, in the western side of the Antarctic Peninsula (Antarctica), has been our research topic since 1991, especially after 2006 when we established different active layer thickness and ground thermal monitoring sites of the CALM and GTN-P international networks of the International Permafrost Association. Along this period, the snow cover thickness did not change at those sites, but since 2010, we observed an elongation on the snow cover duration, with similar snow onset, but a delay on the snow offset. Due to the important effects of snow cover on the ground thermal behavior, we started in late 2015 a new research project (PERMASNOW) focused on the accurate monitoring of the snow cover (duration, density, snow water equivalent and distribution), from very different approaches, including new instrumentation, pictures analysis and remote sensing on optical and radar bands. Also, this interdisciplinary and international research team intends to compare the snow cover and ground thermal behavior with other monitoring sites in the Eastern Antarctic Peninsula where the snow cover is minimum and remains approximately constant.

Since 2006, our research team has been establishing in the islands of Livingston and Deception, (South Shetland archipelago, Antarctica) several monitoring stations of the active layer thickness within the international network Circumpolar Active Layer Monitoring (CALM), and the ground thermal regime for the Ground Terrestrial Network-Permafrost (GTN-P). Both networks were developed within the International Permafrost Association (IPA). In the GTN-P stations, in addition to the temperature of the air, soil, and terrain at different depths, the snow thickness is also monitored by snow poles. Since 2006, a delay in the disappearance of the snow layer has been observed, which could explain the variations we observed in the active layer thickness and permafrost temperatures. Therefore, in late 2015 our research group started the PERMASNOW project (2015-2019) to pay attention to the effect of snow cover on ground thermal This project had two different ways to study the snow cover. On the first hand, in early 2017 we deployed new instrumentation, including new time lapse cameras, snow poles with high number of sensors and a complete and complex set of instruments and sensors to configure a snow pack analyzer station providing 32 environmental and snow parameters. We used the data acquired along 2017 and 2018 years with the new instruments, together with the available from all our already existing sensors, to study in detail the snow cover. On the other hand, remote sensing data were used to try to map the snow cover, not only at our monitoring stations but the entire islands in order to map and study the snow cover distribution, as well as to start the way for future permafrost mapping in the entire islands. MODIS-derived surface temperatures and albedo products were used to detect the snow cover and to test the surface temperature. Since cloud presence limited the acquisition of valid observations of MODIS sensor, we also analyzed Terrasar X data to overcome this limitation. Remote sensing data validation required the acquirement of in situ ground-true data, consisting on data from our permanent instruments, as well as ad hoc measurements in the field (snow cover mapping, snow pits, albedo characterization, etc.). Although the project is finished, the data analysis is still ongoing. We present here the different research tasks we are developing as well as the most important results we already obtained about the snow cover. These results confirm how the snow cover duration has been changing in the last years, affecting the ground thermal behavior.

Air temperature (T_a) is a key variable in many environmental risk models and plays a very important role in climate change research. In previous studies we developed models for estimating the daily maximum (T_max), mean (T_mean), and minimum air temperature (T_min) in peninsular Spain over cloud-free land areas using Moderate Resolution Imaging Spectroradiometer (MODIS) data. Those models were obtained empirically through linear regressions between daily T_a and daytime Terra-MODIS land surface temperature (LST), and then optimized by including spatio-temporal variables. The best T_mean and T_max models were satisfactory (coefficient of determination (R²) of 0.91–0.93; and residual standard error (RSE) of 1.88–2.25 K), but not the T_min models (R² = 0.80–0.81 and RSE = 2.83–3.00 K). In this article T_min models are improved using night-time Aqua LST instead of daytime Terra LST, and then refined including total precipitable water (W) retrieved from daytime Terra-MODIS data and the spatio-temporal variables curvature (c), longitude (λ), Julian day of the year (JD) and elevation (h). The best T_min models are based on the National Aeronautics and Space Administration (NASA) standard product MYD11 LST; and on the direct broadcast version of this product, the International MODIS/AIRS Processing Package (IMAPP) LST product. Models based on Sobrino’s LST1 algorithm were also tested, with worse results. The improved T_min models yield R² = 0.91–0.92 and RSE = 1.75 K and model validations obtain similar R² and RSE values, root mean square error of the differences (RMSD) of 1.87–1.88 K and bias = 0.11 K. The main advantage of the T_min models based on the IMAPP LST product is that they can be generated in nearly real-time using the MODIS direct broadcast system at the University of Oviedo.

Frequent wildfires in the north-west region of Spain affect soil organic matter. Soil properties can be estimated both spatially and temporally using remote sensing. A wide range of satellite and airborne hyperspectral and multispectral sensors are currently available. The spectral resolution varies substantially among sensors, making it difficult to identify the most suitable sensors and spectral regions for a specific application. This study aims to identify the sensors and wavelengths with the greatest potential for topsoil organic C mapping. Total (TOC) and oxidisable organic carbon (OC) content were measured in 89 soil samples collected in a mountain region of north-western Spain. Reflectance spectra of the samples in the spectral region 400–2450 nm were resampled to the bands of five sensors: Compact Airborne Spectrographic Imager (CASI), Airborne Hyperspectral Scanner (AHS), Hyperion, Landsat 5 Thematic Mapper (TM) and Moderate Resolution Imaging Spectroradiometer (MODIS). Calibration models obtained using partial least-squares regression proved to be effective for hyperspectral sensors and also for the multispectral sensor MODIS (R2 = 0.75–0.89), which suggests that hyperspectral capability is not required to accurately predict topsoil organic C. Models based on Landsat performed well, but with an error ~30–45% greater than that obtained for the hyperspectral sensors and MODIS.