TEMPO Evaluation Dashboard Prototype

Evaluation of TEMPO observed nitrogen dioxide (NO2) and formaldehdye (HCHO) for the total atmosphere (totNO2, totHCHO) and NO2 in the troposphere (tropNO2). The analysis below highlights statistical performance of TEMPO compared to reference datasets. For more details, read the Methods section. Or, get started by perusing the Results section maps, scatter plots, box plots, or error plots. Use the "Plot Selector" select specific datasets, sites, and data stratification.

Results

Plot Selector


Figure 1: Points represent each location (Pandora or Non-attainment Area) for TEMPO (left) and reference dataset (right) or statistcs for that site.
Click on site in map (or choose site) to subset Data Range and Error Statistics for just that site.
Figure 2: Points represent the median for each group bin and the error bars (grey) extend out to the interquartile ranges (25% to 75%) in both x and y dimensions. The best fit line from the data subset with Deming regression is shown with the Pearson correlation in parentheses.
Figure 3: Boxes represent the interquartile ranges (25% to 75%) with the line in the middle showing median and the whiskers extending to the min and max values.
Figure 4: Each bar represents a statistic for a group bin.

Methods

Datasets

The best reference is Pandora, but comparing to TropOMI is a useful benchmark. The L2 variables include tropospheric column nitrogen dioxide (tropNO2), total column nitrogen dioxide (totNO2), and total column formaldehyde (totHCHO). For Pandora the totNO2 product is a single measure of the total columns. For TEMPO and TropOMI, the totNO2 product is the sum of tropNO2 and stratospheric NO2 that are separately retrieved.

Data Filtering

All data uses the version 3 (v3) TEMPO data products until 2025-09-07 when the algorithm was updated to v4. For all analysis, all TEMPO data has been filtered for effective cloud fraction less than 0.15 based on SAO recommendation at the GeoXO ACX May 2025 meeting. All data has also been filtered for high quality and having solar zenith angle less than 70 degrees. The notable exceptions are when evaluating by effective cloud (no cloud filtering) or as a function of solar zenith angle (no angle filtering).

Cosampling

Pandora intersections are any TEMPO pixel that overlaps a 0.03-degree radius. Only intersections within 15 minutes of a TEMPO measurement are used.

TropOMI intersections are any TropOMI pixel that overlaps a TEMPO pixel with a region.

TropOMI regions have two types: Pandora sites or Non-attainment Area (NAA). TropOMI at Pandora sites are sampled within a 0.1-degree box around the site. NAA use lat/lon boxes that contain the area, but may be substantially larger than the area. Compared to Pandora sites, the NAA are larger and include more pixels. The NAA have more area in the West, so the time-series including all sites are skewed by the western performance. By comparison, the Pandora are more in the East. As a result, the NAA region analysis of TropOMI should not be compared to the Pandora analysis at Pandora sites.

Statistical Analyses

All statistics are applied to samples either from all sites pooled or by each site. Within those groupings, statistical analyses are also applied withing groupings: The statistics applied to each grouping include: Statistics like MB, MAB, and RMSE are also shown as percents by normalizing to the reference dataset mean.

Citation

This dashboard is developed using the same techniques discussed in the TEMPO Validation Report (doi: 10.5067/doc/tempo/L2TraceGasValidationAssessment/V1.0).