Preprints
https://doi.org/10.5194/amt-2021-253
https://doi.org/10.5194/amt-2021-253

  24 Sep 2021

24 Sep 2021

Review status: this preprint is currently under review for the journal AMT.

Dealing with Spatial Heterogeneity in Pointwise to Gridded Data Comparisons

Amir H. Souri1, Kelly Chance1, Kang Sun2,3, Xiong Liu1, and Matthew S. Johnson4 Amir H. Souri et al.
  • 1Atomic and Molecular Physics (AMP) Division, Harvard–Smithsonian Center for Astrophysics, Cambridge, MA, USA
  • 2Department of Civil, Structural and Environmental Engineering, University at Buffalo, Buffalo, NY, USA
  • 3Research and Education in Energy, Environment and Water Institute, University at Buffalo, Buffalo, NY, USA
  • 4Earth Science Division, NASA Ames Research Center, Moffett Field, CA, USA

Abstract. Atmospheric modelers and the trace gas retrieval community typically presuppose that pointwise measurements, which roughly represent the element of space, should compare well with satellite (model) pixels (grids). This assumption implies that the field of interest must possess a high degree of spatial homogeneity within the pixels (grids), which may not hold true for species with short atmospheric lifetimes or in the proximity of plumes. Results of this assumption often lead to a perception of a nonphysical discrepancy between data, resulting from different spatial scales, potentially making the comparisons prone to overinterpretation. Semivariogram is a mathematical expression of spatial variability in discrete data. Modeling the semivariogram behavior permits carrying out spatial optimal linear prediction of a random process field using kriging. Kriging can extract the spatial information (variance) pertaining to a specific scale, which in turn translating pointwise data to a grid space with quantified uncertainty such that a grid-to-grid comparison can be made. Here, using both theoretical and real-world experiments, we demonstrate that this classical geostatistical approach can be well adapted to solving problems in evaluating model-predicted or satellite-derived atmospheric trace gases. This study demonstrates that satellite validation procedures must take kriging variance and satellite spatial response functions into account. We present the comparison of Ozone Monitoring Instrument (OMI) tropospheric NO2 columns against 11 Pandora Spectrometer Instrument (PSI) systems during the DISCOVER-AQ campaign over Houston. The least-squares fit to the paired data shows a low slope (OMI=0.76×PSI+1.18×1015 molecules cm−2, r2=0.67) which is indicative of varying biases in OMI. This perceived slope, induced by the problem of spatial scale, disappears in the comparison of the convolved kriged PSI and OMI (0.96×PSI+0.66×1015 molecules cm−2, r2=0.72) illustrating that OMI possibly has a constant systematic bias over the area. To avoid gross errors in comparisons made between gridded data versus pointwise measurements, we argue that the concept of semivariogram (or spatial auto-correlation) should be taken into consideration, particularly if the field exhibits a strong degree of spatial heterogeneity at the scale of satellite and/or model footprints.

Amir H. Souri et al.

Status: open (until 29 Oct 2021)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • RC1: 'Review', Anonymous Referee #3, 13 Oct 2021 reply
  • RC2: 'Comment on amt-2021-253', Anonymous Referee #2, 13 Oct 2021 reply

Amir H. Souri et al.

Amir H. Souri et al.

Viewed

Total article views: 509 (including HTML, PDF, and XML)
HTML PDF XML Total Supplement BibTeX EndNote
418 86 5 509 19 5 2
  • HTML: 418
  • PDF: 86
  • XML: 5
  • Total: 509
  • Supplement: 19
  • BibTeX: 5
  • EndNote: 2
Views and downloads (calculated since 24 Sep 2021)
Cumulative views and downloads (calculated since 24 Sep 2021)

Viewed (geographical distribution)

Total article views: 465 (including HTML, PDF, and XML) Thereof 465 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 
Latest update: 16 Oct 2021
Download
Short summary
The central tenants of satellite and model validation are pointwise measurements. Point is an element of space, whereas satellite (model) pixels represent an averaged area. These two datasets are inherently different in nature. We leveraged some geostatistical tools to transform discrete points to gridded data with quantified uncertainty, comparable to satellite footprint (and response functions). This in part alleviated some complications with respect to point-pixel comparisons.