The effects of di _ erent footprint sizes and cloud algorithms on the top-of-atmosphere radiative flux calculation from CERES instrument on Suomi-NPP

This study assumes that CERES-NPP and CERES-Aqua are identical instruments with compatible performance. But one may ask if there were improvements in CERES-NPP instrumentation/electronics or calibration. Possible degradation of CERES-Aqua instrument can be mentioned, if any. Before discussing uncertainties from footprint sizes and cloud algorithms, comprehensive uncertainty analysis of two CERES instruments is necessary to quantify errors from spatial sampling and cloud algorithm differences. I understand that this work is the first step towards identification of the two different CERES observations, but further uncertainty analysis and detailed descriptions are required. The methods to identify the effects of different footprints and cloud algorithms need to be described in detail.


Introduction
The Clouds and Earth's Radiant Energy System (CERES) project has been providing data products critical to advancing our understanding of the effects of clouds and aerosols on radiative energy within the Earth-atmosphere system.CERES data are used by the science community to study the Earth's energy balance (e.g., Trenberth et al. 2009;Kato et al. 2011;Loeb et al. 2012;Stephens et al. 2012), aerosol direct radiative effects (e.g., Satheesh and Ramanathan 2000;Zhang et al. 2005;Loeb and Manalo-Smith 2005;Su et al. 2013), aerosol-cloud interactions (Loeb and Schuster 2008;Quaas et al. 2008;Su et al. 2010b), and to evaluate global general circulation models (e.g., Pincus et al. 2008;Su et al. 2010a;Wang and Su 2013;Wild et al. 2013).
Six CERES instruments have flown on four different spacecrafts thus far.CERES pre-Flight Model (FM) on Tropical Rainfall Measuring Mission (TRMM) was launched on November 27, 1997 into a 350-km circular precessing orbit with a 35 • inclination angle and flied together with the Visible and Infrared Scanner (VIRS).CERES instruments (FM1 and FM2) on Terra were launch on December 18, 1999 into a 705-km sun-synchronous orbit with a 10:30 a.m.equatorial crossing time.CERES instruments (FM3 and FM4) on Aqua spacecraft were launched on May 2, 2002 into a 705-km sun-synchronous orbit with a 1:30 p.m. equatorial crossing time.CERES on Terra and Aqua flies alongside Moderate-Resolution Imaging Spectroradiometer (MODIS).CERES FM5 instrument was launched onboard Suomi-NPP (hereafter referred to as NPP) on October 28, 2011 into a 824-km sunsynchronous orbit with a 1:30 p.m. equatorial crossing time and flies alongside the Visible Infrared Imaging Radiometer Suite (VIIRS).As the orbit altitudes differ among these spacecrafts, the spatial resolutions of CERES instruments also vary from each other.TRMM has the lowest orbit altitude and offers the highest spatial resolution of CERES measurements, about 10 km at nadir; the spatial resolution of CERES on Terra and Aqua is about 20 km at nadir; and is about 24 km at nadir for NPP as it has the highest orbit altitude.
The CERES instrument consists of a three-channel broadband scanning radiometer (Wielicki et al. 1996).The scanning radiometer measures radiances in shortwave (SW, 0.3-5 µm), window (WN, 8-12 µm), and total (0.3-200 µm) channels.The longwave (LW) component is derived as the difference between total and SW channels.These measured radiances at a given sun-Earth-satellite geometry are converted to outgoing reflected solar and emitted thermal TOA radiative fluxes.To do so, the angular distribution of the radiance field must be characterized for different scene types.Here scene type is a combination of variables (e.g., surface type, cloud fraction, cloud optical depth, cloud phase, aerosol optical depth, precipitable water, lapse rate, etc) that are used to group the data to develop distinct angular distribution models (ADMs).To facilitate the construction of ADMs, there are pairs of identical CERES instruments on both Terra and Aqua spacecrafts.At the beginning of these missions one of the instruments on each spacecraft was always placed in a rotating azimuth plane (RAP) scan mode.In this mode, the instrument scans in elevation as it rotates in azimuth, thus acquiring radiance measurements from a wide range of viewing combinations.CERES instruments fly alongside high-resolution imagers, which provide accurate scene type information within CERES footprints.Cloud and aerosol retrievals based upon high-resolution imager measurements are averaged over CERES footprints by accounting for the CERES point spread function (PSF, Smith 1994) and are used for scene type classification.
TRMM ADMs were developed using 9 months of CERES observations and the scene identification information retrieved from VIRS observations (Loeb et al. 2003).Terra/Aqua ADMs were developed using multi-year CERES measurements in RAP mode and in crosstrack mode using the scene identification information from MODIS (Loeb et al. 2005;Su et al. 2015a).The high-resolution MODIS imager provides cloud conditions for every CERES footprint.The cloud algorithms developed by the CERES cloud working group retrieve cloud fraction, cloud optical depth, cloud phase, cloud top and effective temperature/pressure (among other variables) based on MODIS pixel-level measurements (Minnis et al. 2010).
These pixel-level cloud properties are spatially and temporally matched with the CERES footprints and are used to select the scene-dependent ADMs to convert the CERES measured radiances to fluxes.
There is only one CERES instrument on NPP and it has been placed in cross-track scan mode since launch, it is thus not feasible to develop a specific set of ADMs for CERES on NPP.Currently, the Edition 4 Aqua ADMs (Su et al. 2015a) are used to invert fluxes for the CERES measurements on NPP.As mentioned earlier, the CERES footprint size on NPP is larger than that on Aqua.More importantly, the VIIRS channels are not identical to those of MODIS, especially the lack of water vapor (i.e.6.7 µm) and CO2 channels, caused the cloud properties retrieved from MODIS and VIIRS differ from each other.ADMs are scene type dependent, it is important to use consistent scene identification for developing and applying the ADMs.Since the footprint sizes are different between CERES-Aqua and CERES-NPP, will using ADMs developed based on CERES-Aqua measurements for CERES-NPP flux inversion introduce any uncertainties in CERES-NPP flux?Additionally, as the cloud properties retrieved from VIIRS and MODIS differ from each other, the scene identification used to select the ADMs for flux inversion can also be different and thus lead to additional uncertainties in CERES-NPP flux.In this study, we design a simulation study to quantify the CERES-NPP flux uncertainties due to the footprint size difference alone, and due to both the footprint size and cloud property differences.

Method
We cannot answer the above questions by simply differencing the fluxes from CERES-Aqua and CERES-NPP, as the calibrations differ between these two CERES instruments and will be briefly discussed here.The Aqua and NPP orbits fly over each other about every 64 hours.These simultaneous observations from Aqua and NPP are matched to compare SW and LW radiances using CERES Aqua Edition 4 Single Scanner Footprint TOA/Surface Fluxes and Clouds (SSF) product and CERES NPP Edition 1 SSF product.The matching criteria used for SW radiances are that the latitude and longitude differences between the Aqua footprints and the NPP footprints are less than 0.05 degree, solar zenith angle difference is less than 2 degrees, viewing zenith angle and relative azimuth angle differences are less than 5 degrees.Same latitude and longitude matching criteria are used for LW radiances and the viewing zenith angle difference between the Aqua footprints and the NPP footprints is less than 2 degrees.Figure 1 shows the SW, daytime LW, and nighttime LW radiance comparisons between CERES-Aqua and CERES-NPP using matched footprints of 2013 and 2014.The total number of matched footprints, the mean radiances from CERES-Aqua and CERES-NPP, and the root-mean-square errors are summarized in Table 1.The mean SW radiance measured by CERES-NPP is about 1 Wm −2 sr −1 higher than that measured by CERES-Aqua, the daytime mean LW radiance from CERES-NPP is about 0.4 Wm −2 sr −1 lower than that from CERES-Aqua, and the nighttime LW radiance agrees to within 0.1 Wm −2 sr −1 .These differences do not show any view zenith angle dependence.The daytime LW radiance is derived as the difference between total channel and SW channel measurements, and the nighttime LW radiance is simply from the total channel measurements.The differences shown in Table 1 indicate that the calibration of total channels between CERES-Aqua and CERES-NPP agrees very well, and the difference in SW channel calibration could be the cause for the relatively larger daytime LW differences.More research is needed to understand the calibration differences between CERES-Aqua and CERES-NPP.
To quantify the footprint size and cloud retrieval effect on flux inversion without having to account for the calibration differences, we design a simulation study using the MODIS pixel level data.Figure 2 illustrates the process of generating the simulated footprints from the MODIS pixels (represented by the small squares).These pixel-level spectral measurements are used to retrieve cloud properties and aerosol optical depth.These pixel-level imager-derived aerosol and cloud properties, and spectral radiances from MODIS are convolved with the CERES PSF to provide the most accurate aerosol and cloud properties that are spatially and temporally matched with the CERES broadband radiance data.We first 6 To circumvent this issue, we developed narrowband-to-broadband coefficients to convert the MODIS spectral radiances to broadband radiances.
The Edition 4 CERES-Aqua SSF data from July 2002 to September 2007 are used to derive the narrowband-to-broadband regression coefficients separately for SW, daytime LW, and nighttime LW.Seven MODIS spectral bands (0.47, 0.55, 0.65, 0.86, 1.24, 2.13, and 3.7 µm) are used to derive the broadband SW radiances, and the SW regression coefficients are calculated for every calendar month for discrete intervals of solar zenith angle, viewing zenith angle, relative azimuth angle, surface type, snow/non-snow conditions, cloud fraction, and cloud optical depth.Five MODIS spectral bands (6.7, 8.5, 11.0, 12.0, and 14.2 µm) are used to derived the broadband LW radiances, and the LW regression coefficients are calculated for every calendar month for discrete intervals of viewing zenith angle, precipitable water, surface type, snow/none-snow conditions, cloud fraction, and cloud optical depth.The 20 International Geosphere-Biosphere Programme (IGBP) surface types are grouped into 8 surface types: ocean, forest, savanna, grassland, dark desert, bright desert, the Greenland permanent snow, and the Antarctic permanent snow.When there is sea ice over the ocean and snow over the land surface types, regression coefficients for ice and snow conditions are developed (only footprints with 100% sea ice/snow coverage are considered).
These SW and LW narrowband-to-broadband regression coefficients are then applied to

Results
The monthly mean instantaneous TOA SW fluxes derived using the regression generated broadband radiances for simulated CERES-Aqua are shown in Figure 3 The cloud properties in the simulated CERES-Aqua footprints and in the simulated CERES-NPP footprints are all based upon MODIS retrievals, so the scene identifications used to select ADMs for flux inversion are almost the same for both the CERES-Aqua and the CERES-NPP, except small differences due to differing footprint sizes.However, the cloud properties retrieved using MODIS and VIIRS are different, especially over the polar regions.
Figure 4 shows the daytime cloud fraction and cloud optical depth difference between VIIRS and Aqua-MODIS for April 2013.VIIRS retrieval of cloud fraction is greater than that from MODIS by up to 10% and the VIIRS retrieval of cloud optical depth is smaller than that from MODIS by 2∼3 over part of the Antarctic.VIIRS retrieval of cloud fraction over the northern high-latitude snow regions is smaller than that from MODIS, while the optical depth from VIIRS is higher than that from MODIS.Over the Arctic, cloud optical depth from VIIRS is much higher than that from MODIS.Over the ocean from 60 • S to 60 • N, the differences in cloud fraction seem rather random while the differences in cloud optical depth is mostly positive (VIIRS retrieval is higher than Aqua-MODIS retrieval).Polar region cloud fraction differences are mainly because that VIIRS lacks the water vapor and CO2 channels which affect the polar cloud mask algorithm.VIIRS retrieval also use different parameterization of 1.24 µm reflectance which affects cloud optical depth retrieval over the snow/ice surfaces.
These cloud retrieval differences affect the anisotropy factors selected for flux inversion.
The cloud fraction and cloud optical depth retrievals from MODIS convolved in the simulated CERES-NPP footprints are adjusted to be similar to those from VIIRS retrievals to assess how cloud retrieval differences affect the flux.To accomplish this, daily cloud fraction ratios of VIIRS to MODIS are calculated for each 1 • latitude by 1 • longitude grid box.These ratios are then applied to the cloudy footprints of MODIS retrieval to nudge the MODIS cloud fractions to be nearly the same as those from VIIRS retrieval.Note no adjustment is done for clear footprints.Similarly, daily cloud optical depth ratios of VIIRS to MODIS are calculated using cloudy footprints for each 1 • by 1 • grid box.These ratios are used to adjust the MODIS retrieved cloud optical depth to be close to those from VIIRS retrievals.Over the Arctic Ocean, the cloud optical depth from VIIRS retrieval is much higher than that from the MODIS retrieval while the difference in cloud fraction is relatively small.The polar regions are dominated by oblique views and the anisotropy factors for thick clouds are smaller than those for thin clouds at these oblique angles, which led to large flux increase when using VIIRS cloud properties for flux inversion.
The daytime and nighttime instantaneous LW flux from the simulated CERES-Aqua footprints, LW flux differences due to footprint size difference, and LW flux difference due to both footprint size difference and cloud property difference are shown in Figures 5 and 6

Summary and discussion
The scene-type dependent ADMs are used to convert the radiances measured by the CERES instruments to fluxes.Specific empirical ADMs were developed for CERES instruments on TRMM, Terra, and Aqua (Loeb et al. 2003(Loeb et al. , 2005;;Su et al. 2015a).As there is only one CERES instrument on NPP and it has being placed in cross track mode since launch, it is not possible to construct a set of ADMs specific for CERES on NPP.Edition 4 Aqua ADMs (Su et al. 2015a) are thus used for flux inversions for CERES-NPP measurements.However, the altitude of the NPP orbit is higher than that of the Aqua orbit resulting in a larger CERES footprint size on NPP than on Aqua.Given that the footprint size of CERES-NPP is different from that of CERES-Aqua, we need to quantify the CERES-NPP flux uncertainty caused by using the CERES-Aqua ADMs.Furthermore, there are some differences between the imagers fly alongside CERES-Aqua (MODIS) and CERES-NPP (VIIRS), as VIIRS lacks the water vapor and CO2 channels.These spectral differences and algorithm differences lead to notable cloud property differences retrieved from MODIS and VIIRS.
As the anisotropy factors are scene-type dependent, differences in cloud properties will also introduce uncertainties in flux inversion.
Atmos.Meas.Tech.Discuss., doi:10.5194/amt-2017-75,2017 Manuscript under review for journal Atmos.Meas.Tech.Discussion started: 13 April 2017 c Author(s) 2017.CC-BY 3.0 License.use the CERES-Aqua PSF to convolve the aerosol/cloud properties, and the spectral radiances (and other ancillary data) into Aqua-size footprints (orange ovals of the top figure), as is done for the standard CERES-Aqua SSF product.We then increase the footprint size to be that of NPP (orange ovals of the bottom figure) and use the CERES-NPP PSF to average cloud/aerosol properties, spectral radiances, and other ancillary data into the simulated NPP footprints.Four months (July 2012, October 2012, January 2013, and April 2013) of simulated CERES-Aqua and CERES-NPP data were created.For every CERES-Aqua footprint, it contains the broadband SW and LW radiances measured by the CERES instrument.The simulated NPP footprints however do not contain broadband radiances.
(a) for April 2013.Note these fluxes are different from those in the Edition 4 Aqua SSF product as the CERES measured radiances differ from those inferred using narrowband-to-broadband regression coefficients.The flux differences caused by the footprint size difference between the simulated CERES-Aqua and the simulated CERES-NPP are shown in Figure 3(b).Grid boxes in white indicate that the number of footprints with valid SW fluxes differ by more than 2% between simulated CERES-Aqua and CERES-NPP, as the narrowband-to-broadband regressions are only applied to footprints that consist with the same surface types which result in less 8 Atmos.Meas.Tech.Discuss., doi:10.5194/amt-2017-75,2017 Manuscript under review for journal Atmos.Meas.Tech.Discussion started: 13 April 2017 c Author(s) 2017.CC-BY 3.0 License.footprints with valid fluxes for CERES-NPP than for CERES-Aqua.The footprint size difference between CERES-Aqua and CERES-NPP introduces an uncertainty that rarely exceeds 4.0 Wm −2 in monthly gridded CERES-NPP instantaneous SW fluxes.For global monthly mean instantaneous SW flux, the simulated CERES-NPP has a low bias of 0.4 Wm −2 and an uncertainty of 0.8 Wm −2 .Results from the other three months are very similar to April 2013 (not shown).

Figure 3
Figure 3(c) shows the SW flux difference caused by both the footprint size and cloud property differences.Adding the cloud property differences increase the CERES-NPP flux uncertainty compared to when only footprint size differences are considered (Figure 3(b)), regional instantaneous flux uncertainty over the Arctic ocean can exceed 20 Wm −2 .Accounting for cloud property differences, the global mean instantaneous SW flux from simulated CERES-NPP has a high bias of 1.1 Wm −2 and the uncertainty is increased to 2.4 Wm −2 .
. The effect of footprint size on instantaneous LW flux uncertainty is generally within 1.0 Wm −2 for gridded monthly mean LW flux, and the uncertainty of global monthly mean LW flux is within 0.2 Wm −2 .When cloud property differences are also considered, the uncertainty of monthly gridded LW fluxes increases to about 2 Wm −2 with the uncertainty of global monthly mean LW flux of about 0.3 Wm −2 .The instantaneous LW fluxes showed much less sensitivity to cloud property changes than the SW fluxes, especially over the Arctic Ocean where cloud optical depth changed significantly.This is because the LW ADMs over the snow/ice surfaces have very little sensitivity to cloud optical depth (Su et al. 2015a), but they were developed for discrete cloud fraction intervals and larger flux changes are noted in regions experiencing large cloud fraction changes.
To quantify the flux uncertainties due to the footprint size difference between CERES-Aqua and CERES-NPP, and due to both the footprint size difference and cloud property difference, we use the MODIS pixel level data to simulate the CERES-Aqua and CERES-NPP footprints.The simulation is designed to isolate the effects of footprint size difference and cloud property difference on flux uncertainty from calibration difference between CERES-NPP and CERES-Aqua.Comparisons using two years of collocated CERES-Aqua and CERES-NPP footprints indicate that the SW radiances from CERES-NPP is about 1.5% higher than those from CERES-Aqua, the daytime LW radiance from CERES-NPP is about 0.5% lower than those from CERES-Aqua, and the nighttime LW radiances agree to within 0.1%.The pixel-level MODIS spectral radiances, the imager-derived aerosol and cloud properties, and other ancillary data are first convolved with the CERES Aqua PSF to generate the simulated CERES-Aqua footprints, and then convolved with the CERES NPP PSF to generate the simulated CERES-NPP footprints.Broadband radiances within the simulated CERES-Aqua and CERES-NPP footprints are derived using the MODIS spectral bands based upon narrowband-to-broadband regression coefficients developed using five-years of Aqua data, thus ensure consistency between broadband radiances from simulated CERES-Aqua and CERES-NPP.These radiances are then converted to fluxes using the CERES-Aqua ADMs.The footprint size difference between CERES-Aqua and CERES-NPP introduces instantaneous flux uncertainties in monthly gridded CERES-NPP of less than 4.0 Wm −2 for SW, and less than 1.0 Wm −2 for both daytime and nighttime LW.Area-weighted monthly gridded instantaneous flux differences and the absolute flux differences are used to quantify the global monthly mean instantaneous flux bias and uncertainty.The global monthly mean instantaneous SW flux from simulated CERES-NPP has a low bias of 0.4 Wm −2 and an uncertainty of 0.8 Wm −2 , the LW has a bias of about 0.1 Wm −2 and an uncertainty of 0.2 Wm −2 .The cloud properties in the simulated CERES-Aqua footprints and in the simulated CERES-NPP footprints are all based upon MODIS retrievals, but in reality cloud properties retrieved from VIIRS differ from those from MODIS.To assess the flux uncertainty from scene identification differences, cloud fraction and cloud optical depth in the simulated CERES-NPP footprints are perturbed to be more like the VIIRS retrievals.When both footprint size and cloud property differences are considered, the uncertainties of monthly gridded CERES-NPP SW flux can be up to 20 Wm −2 in the Arctic regions where cloud

1
Radiance comparisons between matched CERES-Aqua and CERES-NPP footprints, (a) SW; (b) daytime LW; and (c) nighttime LW using data of 2013 and 2014.20 2 Scheme of convoluting the MODIS pixels into the Aqua and NPP footprints.21 3 The monthly gridded mean TOA instantaneous SW fluxes derived based upon the broadband radiances from regression coefficients for the Aqua footprints (a), the flux differences caused by footprint size difference between simulated NPP and Aqua (b), and the flux differences caused by both footprint size and cloud property differences (c) using April 2013 data.Regions shown in white have large sample number differences between Aqua and simulated NPP.22 4 Cloud fraction (a) and cloud optical depth (b) differences between VIIRS and MODIS retrievals for April 2013.23 5 The monthly gridded mean TOA daytime LW fluxes derived based upon the broadband radiances from regression coefficients for the Aqua footprints (a), the flux differences caused by footprint size difference between simulated NPP and Aqua (b) , and the flux differences caused by both footprint size and cloud property differences (c) using April 2013 data.Regions shown in white have large sample number differences between Aqua and simulated NPP.24 6 The monthly gridded mean TOA nighttime LW fluxes derived based upon the broadband radiances from regression coefficients for the Aqua footprints (a), the flux differences caused by footprint size difference between simulated NPP and Aqua (b), and the flux differences caused by both footprint size and cloud property differences (c) using April 2013 data.Regions shown in white have large sample number differences between Aqua and simulated NPP. 25

Fig. 2 .
Fig. 2. Scheme of convoluting the MODIS pixels into the Aqua and NPP footprints.

Fig. 3 .
Fig. 3.The monthly gridded mean TOA instantaneous SW fluxes derived based upon the broadband radiances from regression coefficients for the Aqua footprints (a), the flux differences caused by footprint size difference between simulated NPP and Aqua (b), and the flux differences caused by both footprint size and cloud property differences (c) using April 2013 data.Regions shown in white have large sample number differences between Aqua and simulated NPP.

Fig. 5 .
Fig. 5.The monthly gridded mean TOA daytime LW fluxes derived based upon the broadband radiances from regression coefficients for the Aqua footprints (a), the flux differences caused by footprint size difference between simulated NPP and Aqua (b) , and the flux differences caused by both footprint size and cloud property differences (c) using April 2013 data.Regions shown in white have large sample number differences between Aqua and simulated NPP.

Fig. 6 .
Fig. 6.The monthly gridded mean TOA nighttime LW fluxes derived based upon the broadband radiances from regression coefficients for the Aqua footprints (a), the flux differences caused by footprint size difference between simulated NPP and Aqua (b), and the flux differences caused by both footprint size and cloud property differences (c) using April 2013 data.Regions shown in white have large sample number differences between Aqua and simulated NPP.