Contributors: T. Usedly (DWD)

Issued by: Deutscher Wetterdienst / Tim Usedly

Date: 01/08/2024

Ref: C3S2_D312a_Lot1.2.2.11_202408_PQAR_ECV_ERB_SLSTR_v1.2

Official reference number service contract: 2021/C3S2_312a_Lot1_DWD/SC1 

History of modifications

List of datasets covered by this document

Related documents

Acronyms

List of tables

List of figures

General definitions

Table 1: Summary of variables and definitions

Table 2: Definition of various technical terms used in the document

Table 3: Definition of processing levels

Table 4: Definition of statistical measures used in the document

B_{i,j}=F_{Data,i,j}-F_{Ref,i,j}

MB=\frac{1}{m*n}*\sum_{i=1}^m \sum_{j=1}^n w_j(B_{i,j})

MAB=\frac{1}{m*n}*\sum_{i=1}^m \sum_{j=1}^n w_j*|B_{i,j}-MB|

Scope of the document

This document provides a description of the product validation results for the Sea and Land Surface Temperature Radiometer (SLSTR) v4.0 based Interim Climate Data Record (ICDR) of the Essential Climate Variable (ECV) Earth Radiation Budget (ERB).

The dataset produced by RAL Space and Brockmann Consult (BC) under the Copernicus Climate Change Service (C3S) programme ranges from 01/2017 – 12/2023 and provides an Interim Climate Data Record (ICDR) to the brokered Thematic Climate Data Record (TCDR) from European Space Agency Cloud Climate Change Initiative (ESA’s Cloud_cci).

The TCDR is a brokered product based on processing of the (Advanced) Along-Track Scanning Radiometer ((A)TSR) onboard ERS-2 and Envisat by RAL Space for the ESA Cloud_cci programme and ranges from 06/1995 – 04/2012. Detailed validation methodology and results are presented in the Cloud_cci Product Validation and Intercomparison Report [D1].

The ICDR is derived with a five-year gap from SLSTR v4.0 onboard the Sentinel-3A and -3B satellites spanning from 01/2017 – 12/2023.

Executive summary

The Sea and Land Surface Temperature Radiometer onboard Sentinel-3A has provided data since January 2017. The launch of Sentinel-3B in October 2018 makes it possible to deliver not only individual data from both satellites but also a merged Sentinel-3A/3B product. The merged version (10/2018 - 12/2023) is validated against the following satellite-based datasets: CERES EBAF Ed4.2, CERES SYN1deg, CLARA-A3, HIRS, as well as ECMWF’s Reanalysis product ERA5. In addition to the merged SLSTR version, a second version on a different grid (equal area in addition to equal angle) is provided for the period from 07/2022 to 12/2023 and also validated against the same reference datasets as the equal angle version of SLSTR.

Validation to these SLSTR derived products is described in the following chapters of this document: Chapter 1 provides a summary of the product validation methodology while chapter 2 presents the validation results. A detailed validation methodology can be found in the Product Quality Assurance Document (PQAD) [D2]. Chapter 3 and 4 discuss possible application specific assessments and compliances with user requirements respectively.

Overall the SLSTR data meets the breakthrough/target GCOS requirement for the horizontal and temporal resolution. However, only OLR meets the threshold/target GCOS requirements in terms of accuracy, while RSF does not meet the threshold GCOS requirements (table 1); the values of the mean absolute bias that vary between 4-5 W/m² for OLR, and between 10-15 W/m² for RSF, depending on the reference, are much higher than the threshold requirement (1 W/m²).

1. Product validation methodology

Detailed information about the validation methodology can be found in the corresponding PQAD [D2], section 3. The validation process is separated into three parts: Data preparation (section 1.1), validation (section 1.2) and evaluation (1.3).

1.1 Data preparation

Table 1-1: provides a summary of the datasets used for the validation and their temporal availability, spatial- and temporal resolution.

Table 1-1: Summary of reference datasets used for validation:

All datasets are, if necessary, remapped to 1°x1° spatial resolution by bilinear interpolation and in case of HIRS monthly mean calculated based on daily means.

1.2 Validation

Following uncertainty metrics are calculated: Bias, Mean Bias and Mean Absolute Bias.

Bias is the difference of dataset and reference dataset for each month and grid box:

B_{i,j}=F_{Data,i,j}-F_{Ref,i,j}  (1)

With B as Bias and F as dataset/reference and i, j as indices. Prior to the bias calculation, the datasets are collocated and only grid point considered, where two (or more) datasets have valid values (not nan). Grid points with identical grid points set to nan for a different dataset are set to nan.

Mean Bias (MB) describes the overall bias with respect to a reference dataset. It is defined as the bias of two gridded data records and a subsequently calculation of the global spatial average. This results in one value per month which can be averaged over the whole time period.

MB=\frac{1}{m*n}*\sum_{i=1}^m \sum_{j=1}^n w_j(B_{i,j})  (2)

with MB as Mean Bias, i and j (m and n, respectively) as indices, w as cosine weighting factor and B as Bias.

Mean Absolute Bias (MAB) is a bias corrected uncertainty metric and calculated by subtracting the previously calculated MB from every grid box bias. Subsequently the same steps as for the calculation of the mean bias are applied.

MAB=\frac{1}{m*n}*\sum_{i=1}^m \sum_{j=1}^n w_j*|B_{i,j}-MB|  (3)

1.3 Evaluation

The previously calculated Mean Absolute Bias is used as evaluation against the requirements defined by the Global Climate Observing System (GCOS) in The 2022 GCOS ECVs Requirements (GCOS 245) [D3]. They are summarized in table 1-2.

Table 1-2: Summary of requirements for OLR and RSF based on GCOS [D3]

 2. Validation results

Chapter 2.1 – 2.2 show the validation results for the two variables Outgoing Longwave Radiation and Reflected Shortwave Flux with a climatology of collocated, deseasonalized, centered and weighted global averages. After the collocation the seasonality of each dataset is removed from the climatology as well as the average of each dataset subtracted.

2.1 Outgoing Longwave Radiation

Figure 2-1: Climatology of collocated, latitude-weighted global monthly means of Outgoing Longwave Radiation for SLSTR and reference datasets

The climatology of the global average shows identical annual cycles for SLSTR as well as all reference datasets with higher values towards the summer months and lower values in winter. ERA5 has the highest average (244.54 W/m²), followed by CERES EBAF (242.70 W/m²), CERES SYN (240.99 W/m²), CLARA-A3 (240.43 W/m²) and HIRS (240.14 W/m²). The SLSTR data record is closer to the three lowest references with an anomaly in summer 2020 leading to an average of 241.96 W/m².

Figure 2-2: Climatology of collocated, deseasonalized, centered and latitude-weighted global monthly means of Outgoing Longwave Radiation for SLSTR and reference datasets

The same data centered around the individual mean and without seasonality indicates that there is an overall good stability between the reference datasets (Figure 2-2). However, the SLSTR based data has unusual negative anomalies for the 2019 and 2021 and positive anomalies for 2020. Figures 2-3 to 2-7 show the anomalies for each year (2019 – 2023) and reference dataset.

Negative anomalies compared to ERA5 (Figure 2-5) are seen for every year on ocean and land areas between 60°N and 60°S. Except for mountainous regions (Rocky Mountains, Andes, Himalaya) and South East Asia where anomalies are positive. Outside of 60° the anomaly is positive on land areas and negative over ocean area. Similar pattern is noticed compared to CERES EBAF (Figure 2-4) with negative anomalies between 60°N and 60°S and positive anomalies at higher latitudes (especially for 2022 and 2023). Mountainous areas contain positive anomalies while there is also a clear structure of positive anomalies following the Inter Tropical Convergence Zone (ITCZ) around the equator.

CLARA-A3 (Figure 2-3), HIRS (Figure 2-6) and CERES SYN (Figure 2-7) have an overall positive bias (except for land area at the northern hemisphere) which is increasing with time. It is also noticeable that the positive bias is even more significant for mountains and the ITCZ, where SLSTR overestimates the Outgoing Longwave Radiation.

Figure 2-3: Yearly mean bias for Outgoing Longwave Radiation of SLSTR and CLARA-A3 dataset for the period 2019 – 2023

Figure 2-4: Yearly mean bias for Outgoing Longwave Radiation of SLSTR and CERES EBAF Ed4.2 dataset for the period 2019 – 2023

Figure 2-5: Yearly mean bias for Outgoing Longwave Radiation of SLSTR and ERA5 dataset for the period 2019 – 2023

Figure 2-6: Yearly mean bias for Outgoing Longwave Radiation of SLSTR and HIRS dataset for the period 2019 – 2023

Figure 2-7: Yearly mean bias for Outgoing Longwave Radiation of SLSTR and CERES SYN1deg dataset for the period 2019 – 2023

Figure 2-8: Global mean absolute bias for SLSTR OLR on the equal angle grid from 10/2018 – 12/2023 compared to reference datasets. Reference datasets are collocated with SLSTR

Figures 2-8 and 2-9 show the global mean absolute bias for SLSTR compared to the reference datasets. The absolute biases show a reasonable stability within the short period of five years and range between 4.60 W/m² and 5.00 W/m². Absolute biases for the 1.5 years of the equal area grid version are slightly higher and range from 4.8 W/m² to 5.40 W/m². There is no bias between the two grid versions provided for the SLSTR data.

Figure 2-9: Global mean absolute bias for SLSTR OLR on the equal area grid from 07/2022 – 12/2023 compared to reference datasets. Reference datasets are collocated with SLSTR

2.2 Reflected Shortwave Flux

Unlike the OLR data from SLSTR, the Reflected Shortwave Flux shows generally higher values compared to the four reference datasets (see Figure 2-10): SLSTR (105.24 W/m²), CERES EBAF (100.35 W/m²), ERA5 (99.11 W/m²), CERES SYN (98.62 W/m²) and CLARA-A3 (97.64 W/m²). Figure 2-11 shows the collocated, deseasonalized and centered datasets where the reference datasets are similar to each other while SLSTR having negative anomalies for 2019 and positive anomalies for 2020/2021. Positive anomalies occur for all years and references (compare Figures 2-12 to 2-15) and are clearly visible for mountainous regions as well as high latitudes. Only ocean areas between 30°N and 30°S have a partly negative anomaly.

Figure 2-10: Climatology of collocated, latitude-weighted global monthly means of Reflected Shortwave Flux for SLSTR and reference datasets

Figure 2-11: Climatology of collocated, deseasonalized, centered and latitude-weighted global monthly means of Reflected Shortwave Flux for SLSTR and reference datasets

Figure 2-12: Yearly mean bias for Reflected Shortwave Flux of SLSTR and CLARA-A3 dataset for the period 2019 – 2023

Figure 2-13: Yearly mean bias for Reflected Shortwave Flux of SLSTR and CERES EBAF Ed4.2 dataset for the period 2019 – 2023

Figure 2-14: Yearly mean bias for Reflected Shortwave Flux of SLSTR and ERA5 dataset for the period 2019 – 2023

Figure 2-15: Yearly mean bias for Reflected Shortwave Flux of SLSTR and CERES SYN1deg dataset for the period 2019 – 2023

Figure 2-16: Global mean absolute bias for SLSTR RSF compared to reference datasets. Reference datasets are collocated with SLSTR

Figures 2-16 and 2-17 presents the monthly absolute biases of each reference datasets compared to the SLSTR product on the equal angle/area grid. Biases for CLARA-A3 (10.03 W/m²), CERES EBAF (10.08 W/m²) and CERES SYN (10.40 W/m²) are close to each other while ERA5 has the highest bias with 13.90 W/m². Biases for the 1.5 year period for the equal angle grid are slightly higher (range from 10.84 W/m² to 14.40 W/m²), with no bias between the two different grid versions.

Figure 2-17: Global mean absolute bias for SLSTR RSF on the equal area grid from 07/2022 – 12/2023 compared to reference datasets. Reference datasets are collocated with SLSTR

3. Application(s) specific assessments

This section is not applicable. There are no additional application specific assessments known since the dataset has just been published.

4.Compliance with user requirements

The GCOS requirements [D3] for the ECV Earth Radiation Budget are used to evaluate the compliance for different users needs. Tables 4-1 and 4-2 show the requirements as well as the results.

GCOS defines three requirements depending on user’s needs:

• Goal (G): The strictest requirement, indicating no further improvements necessary

• Breakthrough (B): Intermediate level between threshold and goal. Breakthrough indicates that it is recommended for certain climate monitoring activities

• Threshold (T): Minimum requirement

The SLSTR ICDR meets the breakthrough/target requirement (closely) for the horizontal/temporal resolution, respectively. However, the accuracy for OLR (depending on the reference between 4-5W/m²) and RSF (10-15 W/m²) do not meet the threshold requirement (1/W/m²).

It is worth mentioning, that the GCOS requirements, defined by the World Meteorological Organisation (WMO), are not focused on satellite-based data records but also on climate models. Satellite-based data records, especially historical observing systems, are often not able to achieve the requirements.

Table 4-1: Results of evaluation against GCOS requirements for SLSTR OLR

Table 4-2: Results of evaluation against GCOS requirements for SLSTR RSF

Related articles

Related articles appear here based on the labels you select. Click to edit the macro and add or change labels.