Data of the Advanced Very High Resolution Radiometer (AVHRR) onboard of NOAA- and since 2006 on EUMETSAT MetOp-satellites is the only source to provide long time series based on an almost unchanged sensor of the last 40 years with daily availability and 1.1km spatial resolution. This unique data set is of high scientific value to study, independent of ground based measurements, climate change induced shifts of Essential Climate Variables (ECV) and fulfills the requirements of the World Meteorological Organization for a sufficient long period (> 30 years) to analyze changes of the environment.
In Europe a few regional AVHRR archives exist, which are partly not accessible or the data holdings are not well maintained to be used by research teams. In addition, ESA has thousands of data sets in their archive received by European stations. All of these local AVHRR data are stored in different formats and mostly without any meta-information. In the frame of ESA’s Heritage Space Programme we developed software and procedures to read and reformat the various data sets combining the data holdings of University of Bern (Switzerland), European Space Agency and Dundee Satellite Receiving Station (UK). This results in a consolidated data set for the period 1981 – 2020 with at least daily resolution. In a final step, metafiles and quicklooks are generated to be included in a data container (EO-SIP) of more than 250.000 AVHRR data in level 1b format (https://doi.org/10.5270/AVH-f1i8784). The metafiles include some quality indicators to guarantee the usability of the data set. This may help the user to select the needed files for their own investigations. Finally, the homogenized data set was transferred to ESA to keep the unique AVHRR images alive for the next decades and make it accessible for all interested parties via ESA web interfaces (https://earth.esa.int/eogateway/catalog/avhrr-level-1b-local-area-coverage-imagery). The whole process for consolidation of the AVHRR data was in-line with the recommendations of the CEOS Working Group on Information Systems and Services (WGISS) to fulfill the needs for data management and stewardship maturity.
The presentation will give some details about the consolidation process to build one of the largest AVHRR archive in Europe, which are a valuable source to retrieve ECVs. The second part will show a few examples of ECV time series to highlight the value of this exceptional data set. Due to the open and free accessibility of the AVHRR LAC data via ESA web dissemination services the data set was and will be also used for projects in the frame of ESA Climate Change Initiative (CCI), e.g. ESA CCI+ snow project.

The analysis of NDVI (Normalized Difference Vegetation Index) time-series enables researchers to derive changes in vegetation condition over time. Especially the availability of multi-decadal remote sensing data with high temporal resolution, such as the AVHRR (Advanced Very High Resolution Radiometer) products, offers a unique possibility to investigate seasonal and interannual variations of vegetation and to identify climate relevant trends.
Within the TIMELINE (TIMe Series Processing of Medium Resolution Earth Observation Data assessing Long-Term Dynamics In our Natural Environment) project the German Remote Sensing Data Center (DFD) of the German Aerospace Center (DLR) aims at generating a homogeneous multi-decadal time series from AVHRR/1, AVHRR/2 and AVHRR/3. Starting in the early 80s, this time series allows to monitor global change related processes in Europe and North Africa at 1 km spatial resolution. Within TIMELINE, different higher-level (L3) land and atmosphere products are developed. The product suite comprises NDVI, land surface temperature (LST), sea surface temperature (SST), snow cover, fire hotspots and burnt area, as well as different cloud properties (e.g. cloud top temperature).
The L3 NDVI product suite from TIMELINE consists of three products: daily, 10-day and monthly NDVI composites. The NDVI is calculated using data from AVHRR channel 1 (RED) and channel 2 (near infrared, NIR). The three NDVI composite products are created by applying a weighted compositing approach utilizing surface reflectance, viewing geometry and quality measures for the respective time period. The TIMELINE L3 NDVI products are gridded and projected to LAEA-ETRS89. The data are provided as netCDF files including the NDVI value for the composite period and additional layers (variance of NDVI values in compositing period, Julian day and corresponding time of acquisition for the selected NDVI value, a quality layer, and geographic coordinates). A quicklook is also provided for each product.
We will present the TIMELINE L3 NDVI product suite including daily, 10-day and monthly NDVI composites, explain the data generation, quality assessment and validation of the products, and discuss possibilities and constraints to utilize this unique data set for vegetation trend analyses.

The Fundamental Data Record for ATMOSpheric Composition (FDR4ATMOS) project is part of the ESA Long Term Data Preservation (LTDP) programme. A Fundamental Data Record (FDR) is a long-term record of selected Earth observation Level 1 parameters (radiance, irradiance, reflectance), possibly multi-instrument, which provides improvements of performance with respect to the individual mission data sets. The aim of task B of the project is to be a pathfinder for future harmonisation of spectrally highly resolved data from other instruments, starting with 2 well known instruments GOME-1 and SCIAMACHY and the spectral ranges for the retrieval of SO2, O3 (UV), NO2 (VIS) and cloud properties (NIR). The FDR will contain harmonised irradiances and reflectances with associated uncertainties. The GOME-1 and SCIAMACHY instruments together span 17 years of spectrally highly resolved data essential for air quality, climate, ozone trend and UV radiation applications. We plan to generate harmonised data sets that allows to directly use it in long-term trend analysis, independently of the instrument. Since this was never done for highly resolved spectrometers, new methods have to be developed that e.g. take into account the different observation geometries for the Earth measurements: GOME-1 and SCIAMACHY had different orbits with a local descending node crossing time of 10:30 and 10:00 respectively resulting in different solar zenith angles. The spatial resolution also differs with GOME-1 having a resolution 40 × 320 km^2 and SCIAMACHY a resolution of 32 × 233 km^2 to 26 × 30 km^2, depending on the spectral range and orbit phase. Since the retrieval of atmospheric trace gas content and other parameters depends on the relative structures of the spectrum, any harmonisation must keep these structures and at the same time take care of broader band differences between the instruments.

The data will also contain uncertainties that are based on metrological principles. For that purpose, we reviewed the error sources and error correlations of the original Level 1 data. These uncertainties will flow into the error propagation, leading to final uncertainties of the FDR.

The resulting algorithms will be implemented into a processing system using the DLR developed GCAPS framework (Generic Calibration and Processing System). The purpose is to keep the methods and the implementation as generic as possible to be able to easily transfer the methodology to other wavelength ranges and to other instruments (e.g. GOME-2 and Sentinel-5p) for a future extension of the time series.

Starting with the solar irradiance as the simpler problem (compared to the Earth radiance measurements) we investigated different methods for the harmonisation and will present the results in the paper. We will also describe the overall validation concept and the status of the harmonisation of the reflectances.

In the framework of the European Long Term Data Preservation Program (LTDP+) which aims to generate innovative Earth system data records named Fundamental Data Records (basically level 1 altimeter and radiometer data) and Thematic Data Records (basically level 2+ geophysical products), ESA/ESRIN has launched a reprocessing activity of ERS-1, ERS-2 and ENVISAT altimeter and radiometer datasets. A large consortium of thematic experts has been formed to take in charge these activities which are 1) to define products including the long, harmonized record of uncertainty-quantified observations, 2) to define the most appropriate level 1 and level 2 processing, 3) to reprocess the whole times series according to the predefined processing and, 4) to validate the different products and provide them to large communities of users focused on the observation of the atmosphere, ocean, coastal, hydrology, sea ice, ice sheet regions.

The level 1 processing of microwave data have been thoroughly reviewed for the three missions. Errors in the processing have been corrected when sufficient information could be found, and radiometric corrections were updated to the latest standards. Uncertainties along the radiometric transfer model are identified and most of them will be evaluated. The aim is to provide the uncertainty for each measurement provided in the Fundamental Data Record. Following previous reprocessing efforts, intercalibration of the three microwave radiometers is achieved with diagnosis such as vicarious calibration over ocean or Amazon Forest. In addition to these harmonized brightness temperatures, a bias-corrected version of the brightness temperatures will be also part of the FDR.

The synthetic aperture radar (SAR) Doppler centroid shift has been demonstrated to contain geophysical information about sea surface wind, waves and current at an accuracy of 5 Hz, and pixel spacing of 3.5-9×8 square km. This corresponds to a horizontal surface velocity of about 20 cm/s at 35 degrees incidence angle. With a goal to retrieve geophysical Doppler shift images from the full archive of global Envisat ASAR wide swath mode SAR acquisitions (2002-2012), we have implemented a new Doppler centroid estimation algorithm based on single look complex data. The algorithm is similar to the method described in Bamler (1991), but extended to correct the azimuth spectra for energy aliased from neighboring areas (side-band effects). The ASAR Doppler algorithm follows the basics of the corresponding Sentinel-1 level 2 algorithm (see, e.g., Engen and Johnsen [2015]). Currently, we have processed Doppler centroid shift grids at about 2 km resolution from all WS acquisitions between 1st January 2008 and 8th April 2012 at the G-POD facility at ESA ESRIN.

In order to estimate a geophysical Doppler shift related to sea surface wind, waves, and current, a geometric Doppler shift arising from the relative velocity, Vrel, between the radar antenna and the rotating Earth at the beam centre must be calculated and subtracted from the Doppler centroid shift. The relative velocity varies with the antenna view angle and the surface topography of the Earth. The effects of orbit and attitude variations (affecting the view angle) and the surface topography can be corrected by employing information available in Envisat restituted attitude files (AUX FRA AX) and precise orbital state vectors from the Doppler Orbitography and Radio-positioning Integrated by Satellite (DORIS) products, a precise surface elevation model, and by using Doppler centroid shift estimates over land. However, malfunctioning antenna elements and/or gain changes of individual antenna elements also cause mispointing. In previous studies, this has been corrected by using land reference, or a corrected ocean signal, in each individual acquisition. This is, however, related to high uncertainty and potential large accuracy variations between the scenes. In the presented work, we have used global land coverage data to create look-up-tables (LUT) for known periods of stable antenna patterns. As such, we can properly calibrate all the ocean scenes to retrieve a more accurate geophysical Doppler shift (2.5 Hz accuracy at 8 km grid) than previously obtained.

Here, we present the calibration routines developed to enable the ASAR WS reprocessing. The key steps in the geophysical Doppler shift estimation are explained, addressing both geometric (satellite orbit and attitude) and electronic (antenna mis-pointing) contributions and corrections. Geophysical Doppler shift retrievals and the inverted surface velocity are then demonstrated, also in comparison to other direct and indirect estimates of the upper ocean current and near-surface wind.