The Fidelity and Uncertainty in Climate data records from Earth Observations consists of new climate datasets from Earth Observation Satellites which have received rigorous treatment of uncertainty informed by the discipline of metrology. The collection datasets include Fundamental Climate Data Records (FCDRs) containing harmonised radiances and [Thematic] Climate Data Records (CDRs) for ensemble sea and lake surface temperature, tropospheric humidity, aerosol optical depth and surface albedo that are widely applicable and metrologically rigorous. The archived FCDRs have a length relevant to climate (>20 years) and can support numerous CDRs. The selected CDRs illustrate new capabilities (e.g. equi-probable ensembles) as well as the benefiting from use of the new FCDRs, resulting in improved stability and traceable uncertainties. Specifically, the data collection contains: harmonised radiances (FCDRs) for the following sensors: Advanced Very High Resolution Radiometer (AVHRR), High Resolution Infrared Radiation Sounder (HIRS), Advanced Microwave Sounding Unit-B (AMSU-B), Microwave Humidy Sounder (MHS) and Meteosat Visible Infra-Red Imager (MVIR); and geophysical datasets with uncertainties.

The Fidelity and uncertainty in climate data records from Earth Observations (FIDUCEO) projcet Advanced Very-High-Resolution Radiometer (AVHRR ) Climate Data Record for Aerosol Optical Depth (AOD) dataset covers Europe and North Africa over land. It was inferred from AVHRR/3 instruments on board the NOAA-16 and NOAA-18 satellites. The dataset is provided on 3 processing levels: superpixels (L2B: 12x12 km2at nadir), gridded (1°x 1°) daily (L3 daily) and monthly (L3 monthly). The original lowest processing level (on selected dark field pixels) is not provided to users, but can be made available on request. The product contains the best AOD estimate but also a more detailed information on different aerosol types (most likely AOD value based on a multi-model ensemble climatology of the aerosol type and a 36 member ensemble of AOD values for a wide range of aerosol types spanning a realistic range in the atmosphere). A user can also process an application with all 36 ensemble members and then calculate the spread of the application results. Note that AOD values on the lowest processing level can be (slightly) negative reflecting radiometric calibration uncertaintiesand keeping un-cut AOD distributions. The products contain on all levels sophisticated and detailed estimates of total AOD uncertainties propagated from the input L1B products and the retrieval algorithm through all levels of the processing chain. These total uncertainties can be directly used for data assimilation or to constrain a confidence interval around the AOD solutions. AOD uncertainties are also kept separated into the (relevant) different parts with different correlation structures, so that a user can conduct averaging and uncertainty propagation as suitable for the intended applications. Uncertainties also include separate values for the dominant effects (reflectance inversion, albedo estimation, aerosol type, cloud masking); also estimates of a sampling uncertainty (due to missing pixels from the cloud masking or from failed inversions) are contained. More information including a report on the datset and scientific background is availible in the documentation section.

The FIDelity and Uncertainty in Climate data records from Earth Observations (FIDUCEO) project Fundamental Climate Data Record of recalibrated brightness temperatures for the High-resolution Infrared Radiation Sounder (HIRS) contains recalibrated brightness temperatures for HIRS for all editions of HIRS/2, HIRS/2I, HIRS/3, and HIRS/4, with metrologically traceable uncertainty estimates. This version is harmonised and anchored to infrared atmospheric sounding interferometer IASI via MetopA satellite. It contains 40 years worth of data covering the period period 1985-03-10 to 2016-12-31. Each file contains: Basic telemetry: longitude, latitude, time, satellite and solar angles; Brightness temperatures for channels1--19; Independent and structured uncertainty for channels 1--19; A lookup table to convert between radiances and brightness temperatures for channels 1—19; A channel error correlation matrix; Two bitfields indicating identified problems with the data. For any data field that varies across the channels (such as brightness temperatures and their uncertainties). Full documentation including product user guide, tutorials, the scientific basis and relevant publications are available in the documentation.

The MVIRI Aerosol Optial depth demonstration dataset contains the aerosol optical thickness (AOT) as retrieved from the visible channel of the Meteosat Visible and Infrared Imager (MVIRI) operated on board Meteosat First Generation (MFG) spacecrafts. The channel is centred around 0.7 µm but the spectral coverage of this channel is very broad. The dataset is produced for 2 of the 7 Meteosat satellites, Meteosat -5 and Meteosat-7, that were operated during the period between 1991 and 2007. While Meteosat-7 was, during the considered period, positioned above 0° longitude, Meteosat-5 was moved from 0° to 63° longitude in support of the INDOEX Experiment in 1998, with continued service in the course of the Indian Ocean Data Coverage (IODC) mission. The aerosol optical thickness (AOT) was retrieved from the MVIRI fundamental climate data record (FCDR) using the Combined Inversion of Surface and AeRosol (CISAR) Algorithm. Both datasets were produced as part of the FIDUCEO (Fidelity and uncertainty in climate data records from Earth Observations) EU Horizon 2020 project. The primary objective of this data record is to assess and demonstrate how the recalibrated and uncertainty-quantified MVIRI FCDR can support improved retrieval of geophysical parameters. Of particular interest is the impact of in-flight reconstructed and spectrally degrading spectral response functions. More information is available in the MVIRI Report and Release Note in the documentation

The FIDUCEO Microwave Fundamental Climate data record, v4.1, contains microwave brightness temperatures and uncertainties for series of satellite instruments (all mission years of SSMT2 on F11, F12, F14, F15; AMSU-B on NOAA15, NOAA16 and NOAA17; and MHS missions (NOAA18, NOAA19, MetopA,-B)). The presented FCDR is a long data record of increased consistency among the instruments compared to the operational data record and is a long enough data record to generate climate data records (CDRs) for climate research. The improvements are based on the strict application of the measurement equation as well as dedicated corrections and improvements within the calibration process. The data record contains quantified uncertainty components, respecting the correlation behaviour of underlying effects. Full documentation including product user guide, tutorials, the scientific basis and relevant publications are available in the documentation

The MVIRI Albedo and Uncertainties demonstration dataset contains the broadband surface albedo as retrieved from the visible channel of the Meteosat Visible and Infrared Imager (MVIRI) operated on board Meteosat First Generation (MFG) spacecrafts. The channel is centered around 0.7 µm but the spectral coverage of this channel is very broad. The dataset is produced for 2 of the 7 Meteosat satellites, Meteosat -5 and Meteosat-7, that were operated during the period between 1991 and 2007. While Meteosat-7 was, during the considered period, positioned above 0° longitude, Meteosat-5 was moved from 0° to 63° longitude in support of the INDOEX Experiment in 1998, with continued service in the course of the Indian Ocean Data Coverage (IODC) mission. The albedo data was retrieved from the MVIRI fundamental climate data record; both were produced as part of the FIDUCEO (Fidelity and uncertainty in climate data records from Earth Observations) EU Horizon 2020 project.

The The FIDelity and Uncertainty in Climate data records from Earth Observations (FIDUCEO) project Upper Tropospheric Humidity (UTH) Climate Data Record version 1.2 dataset is derived from satellite brightness temperatures and uncertainties from the FIDUCEO Microwave Fundamental Climate Data Record (FCDR). The instantaneous observations from the FIDUCEO Microwave FCDR are used to derive a spatio-temporal averaged data record, which contains monthly mean UTH and brightness temperature mapped to a regular latitude/longitude grid covering the tropical region (-30° to 30° N), with a spatial resolution of 1° x 1°. It covers all mission years of the Special Senson Microwave for Temperature (SSMT2) instrument on the F11, F12, F14, F15 satellites, the Advanced Mircowave Sounding Unit (AMSU-B ) instrument on the NOAA 15-17 satellites and the Microwave Humidity Sounder (MHS) instruments on the NOAA18, NOAA19, MetopA, Metop-B satellites. Full documentation including product user guide, tutorials and relevant publications are available in the documentation

This Fundamental Climate Data Record (FDCR) ensemble product contains both recalibrated AVHRR/3 MetOp-A Radiance/Brightness Temperature data with associated metrologically traceable uncertainties in the FIDUCEO FCDR format. It also contains files containing an Ensemble dataset consisting of perturbations to the associated FIDUCEO FCDR radiances and brightness temperatures. By applying the 10 perturbations to the baseline FCDR radiances and brightness temperatures a user is able to generate 10 sets of new measurements whose variance capture the associated underlying uncertainty distributions contained in the Easy FCDR itself. The FIDelity and Uncertainty in Climate data records from Earth Observations (FIDUCEO) project AVHRR FCDR improves on existing AVHRR level-1B: in the infrared the calibration has been improved with a measurement function approach such that the data is of better quality (noise has been reduced, outliers have been filtered) the metrologically traceable uncertainties have been derived together with their associated effects, cross-channel correlations and long-term correlation structures have now been calculated from the processed data and are being understood and used to improve data quality and consistency. For the Ensemble product the sensors have been calibrated against the Advanced Along-Track Scanning Radiometer (AATSR) sensor with additional corrections to calibration parameters which make the data better able to derive sea surface temperature estimates that are consistent with theInternational Comprehensive Ocean-Atmosphere Data Set (ICOADS) drifting buoy network. Because the Ensemble has been tuned for Sea Surface Temperature retrieval it should only be used over ocean scenes.

This Fundamental Climate Data Record (FDCR) of recalibrated brightness temperatures for the Advanced Very-High-Resolution Radiometer (AVHRR) AVHRR/1, AVHRR/2 andAVHRR/3 with metrologically-traceable uncertainty estimates. Error covariance information is also provided.In this data set , in addition to relative reflectance for channels 1, 2 and 3A( when available) together with estimated independent, common and structured uncertainties are also provided. The FIDelity and Uncertainty in Climate data records from Earth Observations (FIDUCEO) project AVHRR FCDR improves on existing AVHRR level-1B data (such as that processed by NOAA or EUMETSAT): the calibration has been improved with a measurement function approach such that the data is of better quality (noise has been reduced, outliers have been filtered) the metrologically traceable uncertainties have been derived together with their associated effects cross-channel correlations and long-term correlation structures have now been calculated from the processed data and are being understood and used to improve data quality and consistency all the sensors are calibrated to a common reference (AATSR series).The products have been harmonised across the satellite series using Simultaneous (Nadir) Overpasses (SNOs). Full documentation including product user guide, tutorials, the scientific basis and relevant publications are available in the documentation.

The FIDelity and Uncertainty in Climate data records from Earth Observations (FIDUCEO) project Sea and Lake Surface Temperature Climate Data Record core retrieved quantity is the skin (radiometric) temperature of the Earth’s water surfaces (sea and large lakes). This is provided as a best estimate, plus an ensemble of 10 perturbations capturing known uncertainties. The CDR contains grid-cell instantaneous averagesof retrieved surface temperature over ice-free oceans and 300 large lakes. The FIDUCEO Surface Temperature CDR differs from the ESA Sea Surface Temperature Climate Change Initiative CDRs ; which were generated using in the using the same cloud detection and SST retrieval methodology in the following points: - The calibration of the brightness temperatures used is revised for the FIDUCEO ST CDR. The first step in this has been multi-sensor harmonisation to obtain baseline calibration coefficients (Giering et al., 2019). For specific ST application, these coefficients were adjusted such that SSTs had lower bias, using a method of cross-referencing to matched drifting buoys (Merchant et al., 2019) - Perturbations to the obtained ST and quality level determination are provided for an ensemble of 10 members, for the purpose of propagating uncertainty in ST in complex (large scale, non-linear) applications. - The FIDUCEO ST CDR includes retrievals over the world’s 300 largest lakes, unlike the SST-only product. (Lakes, including much smaller lakes,are addressed in other CDRs requiring significantly different methodsto cope with the difficulties of small target water bodies.) Full documentation including product user guide, tutorials, the scientific basis and relevant publications are available in the documentation.