Vegetation greening is observed over the Arctic, and its feedback to Arctic amplification has attracted increasing attention. Previous studies have pr…Vegetation greening is observed over the Arctic, and its feedback to Arctic amplification has attracted increasing attention. Previous studies have primarily focused on the temperature effect of a single environmental variable (e.g., albedo), while the separate contributions of land surface albedo, evapotranspiration (ET) and water vapor remain underexamined. In this study, we develop knowledge-based data-driven models (i.e., path analysis and machine learning) to estimate the temperature effect of vegetation greening and quantify the separate contributions of albedo, ET and water vapor in July and August from 1982 to 2015. The results show a wide range of temperature sensitivity to the NDVI (Normalized Difference Vegetation Index), and vegetation greening has led to Arctic warming of 0.76 °C, 0.68 °C, 0.83 °C in July and August and the average of the two months, respectively. Path analysis suggested that vegetation greening affects Arctic air temperature mainly by regulating albedo and water vapor. In July, changes in water vapor contributed the most to the temperature effect of vegetation greening with a contribution of 0.25 ± 0.08 °C, while in August, changes in albedo and water vapor had similar effects with a contribution of 0.21 ± 0.08 °C. In contrast, changes in ET have generated a negligible cooling effect due to small changes in ET. Further analysis shows similar positive contributions of albedo and water vapor in barren, graminoid tundra, prostrate-shrub tundra and erect-shrub, with contributions ranging from 0.18 ± 0.05°C to 0.30 ± 0.11°C, while changes in water vapor dominate vegetation’s temperature effect in wetlands, with contributions ranging from 0.26 ± 0.11°C to 0.32 ± 0.16°C. This study emphasizes the importance of considering multiple driving factors to assess the temperature effect of vegetation greening in a consistent framework and highlights the critical role of water vapor change in addition to the widely examined albedo in explaining Arctic warming.more
Scatterometer observations over land are sensitive to the water content in soil and vegetation, but have been rarely used to study seasonal changes in…Scatterometer observations over land are sensitive to the water content in soil and vegetation, but have been rarely used to study seasonal changes in the plant water status and seasonal development of deciduous trees. Here we use Advanced Scatterometer (ASCAT) observations to investigate the sensitivity of C-band backscatter to spring phenology of temperate deciduous broadleaf forests in Austria. ASCAT's multi-angle looking capability enables the observation of backscatter over a large range of incidence angles. The vegetation status affects the slope of the backscatter-incidence angle relationship. We discovered a maximum in the slope around the month April, hereafter referred to as spring peak, predominantly in regions covered by deciduous broadleaf forest. We hypothesized that the spring peak indicates the average timing of leaf emergence in the deciduous trees in the sensor footprint. The hypothesis was tested by comparing the timing of the spring peak to leaf unfolding observations from the PEP725 phenology database, to the increase of leaf area index (LAI) during spring, and to temperature. Our results demonstrate a good agreement between the ASCAT spring peaks, phenology observations and temperature conditions. The steepest increase in LAI however lags behind the ASCAT peak by several days to a few weeks, suggesting that the spring peak in fact marks the timing of maximum woody water content, which occurs right before leaf emergence. Based on these observations, we conclude that the ASCAT signal has a high sensitivity to spring reactivation and in particular water uptake of bare deciduous broadleaf trees. Our findings might provide the basis for novel developments to estimate eco-physiological changes of forests during spring at large scales.more
Antarctic sea ice is mostly seasonal. While changes in sea ice seasonality have been observed in recent decades, the lack of process understanding rem…Antarctic sea ice is mostly seasonal. While changes in sea ice seasonality have been observed in recent decades, the lack of process understanding remains a key challenge to interpret these changes. To address this knowledge gap, we investigate the processes driving the ice season onset, known as sea ice advance, using remote sensing and in situ observations. Here, we find that seawater freezing predominantly drives advance in the inner seasonal ice zone. By contrast, in an outer band a few degrees wide, advance is due to the import of drifting ice into warmer waters. We show that advance dates are strongly related to the heat stored in the summer ocean mixed layer. This heat is controlled by the timing of sea ice retreat, explaining the tight link between retreat and advance dates. Such a thermodynamic linkage strongly constrains the climatology and interannual variations, albeit with less influence on the latter.more
In this work, the effect that two basic air quality indexes, aerosols and tropospheric NO2, exert on surface solar radiation (SSR) is studied, along w…In this work, the effect that two basic air quality indexes, aerosols and tropospheric NO2, exert on surface solar radiation (SSR) is studied, along with the effect of liquid and ice clouds over 16 locations in Greece, in the heart of the Eastern Mediterranean. State-of-the-art satellite-based observations and climatological data for the 15-year period 2005–2019, and a radiative transfer system based on a modified version of the Santa Barbara DISORT Atmospheric Radiative Transfer (SBDART) model are used. Our SSR simulations are in good agreement with ground observations and two satellite products. It is shown that liquid clouds dominate, with an annual radiative effect (RE) of −36 W/m2, with ice clouds (−19 W/m2) and aerosols (−13 W/m2) following. The radiative effect of tropospheric NO2 is smaller by two orders of magnitude (−0.074 W/m2). Under clear skies, REaer is about 3–4 times larger than for liquid and ice cloud-covered skies, while RENO2 doubles. The radiative effect of all the parameters exhibits a distinct seasonal cycle. An increase in SSR is observed for the period 2005–2019 (positive trends ranging from 0.01 to 0.52 W/m2/year), which is mostly related to a decrease in the aerosol optical depth and the liquid cloud fraction.more
Cyprus plans to drastically increase the share of renewable energy sources from 13.9% in 2020 to 22.9% in 2030. Solar energy can play a key role in th…Cyprus plans to drastically increase the share of renewable energy sources from 13.9% in 2020 to 22.9% in 2030. Solar energy can play a key role in the effort to fulfil this goal. The potential for production of solar energy over the island is much higher than most of European territory because of the low latitude of the island and the nearly cloudless summers. In this study, high quality and fine resolution satellite retrievals of aerosols and dust, from the newly developed MIDAS climatology, and information for clouds from CM SAF are used in order to quantify the effects of aerosols, dust, and clouds on the levels of surface solar radiation for 2004–2017 and the corresponding financial loss for different types of installations for the production of solar energy. Surface solar radiation climatology has also been developed based on the above information. Ground-based measurements were also incorporated to study the contribution of different species to the aerosol mixture and the effects of day-to-day variability of aerosols on SSR. Aerosols attenuate 5–10% of the annual global horizontal irradiation and 15–35% of the annual direct normal irradiation, while clouds attenuate 25–30% and 35–50% respectively. Dust is responsible for 30–50% of the overall attenuation by aerosols and is the main regulator of the variability of total aerosol. All-sky annual global horizontal irradiation increased significantly in the period of study by 2%, which was mainly attributed to changes in cloudiness.more