Spring enhancement and summer reduction in carbon uptake during the 2018 drought in northwestern Europe.

We analysed gross primary productivity (GPP), total ecosystem respiration (TER) and the resulting net ecosystem exchange (NEE) of carbon dioxide (CO2) by the terrestrial biosphere during the summer of 2018 through observed changes across the Integrated Carbon Observation System (ICOS) network, through biosphere and inverse modelling, and through remote sensing. Highly correlated yet independently-derived reductions in productivity from sun-induced fluorescence, vegetative near-infrared reflectance, and GPP simulated by the Simple Biosphere model version 4 (SiB4) suggest a 130-340 TgC GPP reduction in July-August-September (JAS) of 2018. This occurs over an area of 1.6 × 106 km2 with anomalously low precipitation in northwestern and central Europe. In this drought-affected area, reduced GPP, TER, NEE and soil moisture at ICOS ecosystem sites are reproduced satisfactorily by the SiB4 model. We found that, in contrast to the preceding 5 years, low soil moisture is the main stress factor across the affected area. SiB4's NEE reduction by 57 TgC for JAS coincides with anomalously high atmospheric CO2 observations in 2018, and this is closely matched by the NEE anomaly derived by CarbonTracker Europe (52 to 83 TgC). Increased NEE during the spring (May-June) of 2018 (SiB4 -52 TgC; CTE -46 to -55 TgC) largely offset this loss, as ecosystems took advantage of favourable growth conditions. This article is part of the theme issue 'Impacts of the 2018 severe drought and heatwave in Europe: from site to continental scale'.

Inter-comparison of 2 microm Heterodyne Differential Absorption Lidar, Laser Diode Spectrometer, LICOR NDIR analyzer and flasks measurements of near-ground atmospheric CO2 mixing ratio.

Remote sensing and in situ instruments are presented and compared in the same location for accurate CO(2) mixing ratio measurements in the atmosphere: (1) a 2.064 microm Heterodyne DIfferential Absorption Lidar (HDIAL), (2) a field deployable infrared Laser Diode Spectrometer (LDS) using new commercial diode laser technology at 2.68 microm, (3) LICOR NDIR analyzer and (4) flasks. LDS, LICOR and flasks measurements were made in the same location, LICOR and flasks being taken as reference. Horizontal HDIAL measurements of CO(2) absorption using aerosol backscatter signal are reported. Using new spectroscopic data in the 2 microm band and meteorological sensor measurements, a mean CO(2) mixing ratio is inferred by the HDIAL in a 1 km long path above the 15m height location of the CO(2) in situ sensors. We compare HDIAL and LDS measurements with the LICOR data for 30 min of time averaging. The mean standard deviation of the HDIAL and the LDS CO(2) mixing ratio results are 3.3 ppm and 0.89 ppm, respectively. The bias of the HDIAL and the LDS measurements are -0.54 ppm and -0.99 ppm, respectively.