Towards spaceborne monitoring of localized CO2 emissions: an instrument concept and first performance assessment

Abstract

The UNFCCC (United Nations Framework Convention on Climate Change) requires the nations of the world to report their carbon dioxide (${\mathrm{CO}}_{\mathrm{2}}$) emissions. The independent verification of these reported emissions is a cornerstone for advancing towards the emission accounting and reduction measures agreed upon in the Paris Agreement. In this paper, we present the concept and first performance assessment of a compact spaceborne imaging spectrometer with a spatial resolution of $\mathrm{50}\times \mathrm{50}$ ${\mathrm{m}}^{\mathrm{2}}$ that could contribute to the “monitoring, verification and reporting” (MVR) of ${\mathrm{CO}}_{\mathrm{2}}$ emissions worldwide. ${\mathrm{CO}}_{\mathrm{2}}$ emissions from medium-sized power plants (1–10 $\mathrm{Mt}{\mathrm{CO}}_{\mathrm{2}}{\mathrm{yr}}^{-\mathrm{1}}$), currently not targeted by other spaceborne missions, represent a significant part of the global ${\mathrm{CO}}_{\mathrm{2}}$ emission budget. In this paper we show that the proposed instrument concept is able to resolve emission plumes from such localized sources as a first step towards corresponding ${\mathrm{CO}}_{\mathrm{2}}$ flux estimates.

Through radiative transfer simulations, including a realistic instrument noise model and a global trial ensemble covering various geophysical scenarios, it is shown that an instrument noise error of 1.1 ppm ($\mathrm{1}\mathit{\sigma }$) can be achieved for the retrieval of the column-averaged dry-air mole fraction of ${\mathrm{CO}}_{\mathrm{2}}$ (X${\mathrm{CO}}_{\mathrm{2}}$). Despite a limited amount of information from a single spectral window and a relatively coarse spectral resolution, scattering by atmospheric aerosol and cirrus can be partly accounted for in the X${\mathrm{CO}}_{\mathrm{2}}$ retrieval, with deviations of at most 4.0 ppm from the true abundance for two-thirds of the scenes in the global trial ensemble.

We further simulate the ability of the proposed instrument concept to observe ${\mathrm{CO}}_{\mathrm{2}}$ plumes from single power plants in an urban area using high-resolution ${\mathrm{CO}}_{\mathrm{2}}$ emission and surface albedo data for the city of Indianapolis. Given the preliminary instrument design and the corresponding instrument noise error, emission plumes from point sources with an emission rate down to the order of 0.3 $\mathrm{Mt}{\mathrm{CO}}_{\mathrm{2}}{\mathrm{yr}}^{-\mathrm{1}}$ can be resolved, i.e., well below the target source strength of 1 $\mathrm{Mt}{\mathrm{CO}}_{\mathrm{2}}{\mathrm{yr}}^{-\mathrm{1}}$. This leaves a significant margin for additional error sources, like scattering particles and complex meteorology, and shows the potential for subsequent ${\mathrm{CO}}_{\mathrm{2}}$ flux estimates with the proposed instrument concept.

1 Introduction

Despite the broad consensus on the negative long-term effects of carbon dioxide (${\mathrm{CO}}_{\mathrm{2}}$) emissions and the efforts to reduce these emissions, the atmospheric ${\mathrm{CO}}_{\mathrm{2}}$ concentrations continue to rise. During the course of 2018, the average ${\mathrm{CO}}_{\mathrm{2}}$ concentration increased from 407 to 410 ppm at the Mauna Loa observatory, representing the fourth-highest annual growth ever recorded at that observatory . ${\mathrm{CO}}_{\mathrm{2}}$ emissions from localized point sources represent a large fraction of the ${\mathrm{CO}}_{\mathrm{2}}$ emitted into the atmosphere. The International Energy Agency (IEA) recently reported that emissions from coal-fired power plants exceeded 10 $\mathrm{Gt}{\mathrm{CO}}_{\mathrm{2}}{\mathrm{yr}}^{-\mathrm{1}}$ for the first time in 2018, hence accounting for approximately 30 % of the global ${\mathrm{CO}}_{\mathrm{2}}$ emissions , mainly due to the continued growth of coal use in Asia and other emerging economies. Figure  depicts the global distribution of reported and estimated annual ${\mathrm{CO}}_{\mathrm{2}}$ emissions from power plants for the year 2009, as provided by the CARMA (Carbon Monitoring for Action) v3.0 database , together with the corresponding cumulative distribution of the power plant emissions. The emission total from 16 898 individual power plants, where exact or approximate coordinates are available, adds up to 9.9 $\mathrm{Gt}{\mathrm{CO}}_{\mathrm{2}}{\mathrm{yr}}^{-\mathrm{1}}$. A large fraction of power plant emissions originates from a relatively small number of large to medium-sized power plants. The CARMA data show that 153 large power plants ($>\mathrm{10}$ $\mathrm{Mt}{\mathrm{CO}}_{\mathrm{2}}{\mathrm{yr}}^{-\mathrm{1}}$) accounted for 24 % of the total annual power plant ${\mathrm{CO}}_{\mathrm{2}}$ emissions, whereas 2111 large and medium-sized power plants ($>\mathrm{1}$ $\mathrm{Mt}{\mathrm{CO}}_{\mathrm{2}}{\mathrm{yr}}^{-\mathrm{1}}$) accounted for as much as 88 % of the power plant ${\mathrm{CO}}_{\mathrm{2}}$ emission budget, clearly demonstrating the significant contribution from medium-sized power plants (1–10 $\mathrm{Mt}{\mathrm{CO}}_{\mathrm{2}}{\mathrm{yr}}^{-\mathrm{1}}$) to the global ${\mathrm{CO}}_{\mathrm{2}}$ emission budget.

To advance towards emission accounting and reduction measures, agreed upon in the Paris Agreement in force since 2016, the independent verification of reported emissions is of high importance. To this end, spaceborne instruments provide a suitable platform with which continuous long-term measurements can potentially be combined with near-global coverage with no geopolitical boundaries.

Most of the currently operating, planned and proposed instruments for passive ${\mathrm{CO}}_{\mathrm{2}}$ observations from space measure the reflected shortwave infrared (SWIR) solar radiation in several spectral windows covering the oxygen A (${\mathrm{O}}_{\mathrm{2}}$ A) band near 750 nm as well as the weak and strong ${\mathrm{CO}}_{\mathrm{2}}$ absorption bands near 1600 and 2000 nm, respectively, e.g., GOSAT Greenhouse Gases Observing Satellite;, OCO-2 Orbiting Carbon Observatory-2;, TanSat , GOSAT-2 , OCO-3 , MicroCarb , GeoCarb , CarbonSat and G3E Geostationary Emission Explorer for Europe;. These instruments and instrument concepts further rely on a comparatively high spectral resolution on the order of approximately 0.05–0.3 nm, representing resolving powers (ratio of wavelength over the full-width half-maximum of the instrument spectral response function) ranging from approximately 3600 for the strong ${\mathrm{CO}}_{\mathrm{2}}$ absorption bands near 2000 nm for CarbonSat up to $>\mathrm{20}\mathrm{000}$ for the OCO and GOSAT instruments. Such advanced instruments, like GOSAT and OCO-2 that have been operating since 2009 and 2014, respectively, generally target an accuracy and coverage sufficient to study the natural ${\mathrm{CO}}_{\mathrm{2}}$ cycle on a regional to continental scale e.g., but have also been used to observe and quantify ${\mathrm{CO}}_{\mathrm{2}}$ gradients on the regional scale caused by anthropogenic ${\mathrm{CO}}_{\mathrm{2}}$ emissions in urban areas . OCO-2 data have further been used to observe strong ${\mathrm{CO}}_{\mathrm{2}}$ plumes from localized natural and anthropogenic ${\mathrm{CO}}_{\mathrm{2}}$ sources like volcanoes and coal-fired power plants , demonstrating the capabilities of imaging spectrometers to monitor ${\mathrm{CO}}_{\mathrm{2}}$ from space. The spatial resolution of OCO-2 and similar instruments like OCO-3, TanSat and the planned Copernicus ${\mathrm{CO}}_{\mathrm{2}}$ Monitoring mission CO2M (on the order of 2–4 ${\mathrm{km}}^{\mathrm{2}}$) does, however, pose a difficulty for the routine monitoring of localized power plant ${\mathrm{CO}}_{\mathrm{2}}$ emissions, since the plume is usually only sampled by a handful of pixels in which ${\mathrm{CO}}_{\mathrm{2}}$ plume enhancements cannot be fully separated from the background, making quantitative ${\mathrm{CO}}_{\mathrm{2}}$ emission rate estimates difficult and vulnerable to cloud contamination and instrument noise propagating into ${\mathrm{CO}}_{\mathrm{2}}$ retrieval errors. For this reason CO2M will target isolated large power plants ($\mathit{\gtrsim }\mathrm{10}$ $\mathrm{Mt}{\mathrm{CO}}_{\mathrm{2}}{\mathrm{yr}}^{-\mathrm{1}}$) and large urban agglomerations ($\mathit{\gtrsim }$ Berlin) , and thus a large fraction of the emission total will be missed.

To contribute to closing this gap and expanding on future ${\mathrm{CO}}_{\mathrm{2}}$ monitoring from space, we present the concept and a first performance assessment of a spaceborne imaging spectrometer that could be deployed for the dedicated monitoring of localized ${\mathrm{CO}}_{\mathrm{2}}$ emissions. By targeting power plants with an annual emission rate down to approximately 1 $\mathrm{Mt}{\mathrm{CO}}_{\mathrm{2}}{\mathrm{yr}}^{-\mathrm{1}}$, a substantial fraction of the ${\mathrm{CO}}_{\mathrm{2}}$ emissions from power plants and hence a significant part of the global man-made ${\mathrm{CO}}_{\mathrm{2}}$ emission budget in total could be resolved (given global coverage through a fleet of instruments). As shown in Fig. , it is of key importance to also cover medium-sized power plants (1–10 $\mathrm{Mt}{\mathrm{CO}}_{\mathrm{2}}{\mathrm{yr}}^{-\mathrm{1}}$) as they alone contributed to approximately 64 % of the ${\mathrm{CO}}_{\mathrm{2}}$ emissions from power plants in 2009 according to the CARMA v3.0 data. To achieve this, the proposed instrument has an envisaged spatial resolution of $\mathrm{50}\times \mathrm{50}$ ${\mathrm{m}}^{\mathrm{2}}$. With such a high spatial resolution and large amount of ground pixels per unit area, averaging plume enhancements and background concentration fields is avoided. This leads to an enhanced contrast compared to a coarser spatial resolution. To increase the number of collected photons and hence the signal-to-noise ratio (SNR) and relative precision of the ${\mathrm{CO}}_{\mathrm{2}}$ concentration retrievals, such a high spatial resolution has to be compensated for with a rather coarse spectral resolution. To further compensate for the limited spatial coverage of a single instrument, a comparatively compact and low-cost instrument design is an important aspect, as it would allow for a fleet of instruments to be deployed, increasing the spatial coverage.

recently demonstrated that atmospheric ${\mathrm{CO}}_{\mathrm{2}}$ concentrations can be retrieved with an accuracy $<\mathrm{1}$ % using such a comparatively simple spectral setup with one single spectral window and a relatively coarse spectral resolution of approximately 1.3–1.4 nm (resolving power of 1400–1600). demonstrated the ability to resolve and quantify methane (${\mathrm{CH}}_{\mathrm{4}}$) plumes, which pose a similar remote sensing challenge as ${\mathrm{CO}}_{\mathrm{2}}$, using data from the spaceborne Hyperion imaging spectrometer with a spectral and spatial resolution of 10 nm (resolving power around 230) and 30 m, respectively. The observation of emission plumes, from plume detection to enhancement quantification and flux estimation, using imaging spectroscopy with a single narrow spectral window and a spectral resolution as coarse as 5 to 10 nm (resolving power around 200–500) has further been repeatedly demonstrated using airborne imaging spectroscopy data for both ${\mathrm{CO}}_{\mathrm{2}}$ and ${\mathrm{CH}}_{\mathrm{4}}$ . For an airborne instrument primarily dedicated to the quantitative imaging of ${\mathrm{CH}}_{\mathrm{4}}$, but also ${\mathrm{CO}}_{\mathrm{2}}$ plumes, proposed a single spectral window and a spectral resolution of 1.0 nm (resolving power around 2000–2400), again coarse enough to reach a spatial resolution on the order of 10–100 m. The commercial instrument GHGSat-D operated by the Canadian company GHGSat Inc. was launched in 2016 as a demonstrator for a satellite constellation concept targeting the detection of ${\mathrm{CH}}_{\mathrm{4}}$ plumes from individual point sources within selected $\approx \mathrm{10}\times \mathrm{10}$ km${}^{\mathrm{2}}$ target regions at a spectral and spatial resolution of 0.1 nm (resolving power around 16 000) and 50 m, respectively . recently showed how anomalously large ${\mathrm{CH}}_{\mathrm{4}}$ point sources can be discovered with GHGSat-D observations.

Figure 2

Ray-tracing diagram of the preliminary optical design assuming a three-mirror anastigmat (TMA) telescope combined with an Offner-type spectrometer.

[Figure omitted. See PDF]

Given the results from previous studies and the technology at hand, we are confident that the proposed instrument concept presented here could be realized and that it would be an important complement to the fleet of current and planned spaceborne ${\mathrm{CO}}_{\mathrm{2}}$ instruments, allowing for the routine quantitative monitoring of ${\mathrm{CO}}_{\mathrm{2}}$ emissions from large and medium-sized power plants. The proposed instrument concept would also serve as a good complement and companion to CO2M by also targeting medium-sized power plants and providing high-resolution images with finer ${\mathrm{CO}}_{\mathrm{2}}$ plume structures. The added value of such an instrument would be of interest both in terms of advancing science and providing independent emission estimates that could be used to verify reported ${\mathrm{CO}}_{\mathrm{2}}$ emission rates at the facility level and inform policy makers on the progress of reducing man-made ${\mathrm{CO}}_{\mathrm{2}}$ emissions. The proposed instrument concept is described in Sect. , followed by a description of the instrument noise model in Sect. . A global performance assessment addressing instrument noise and the errors introduced by atmospheric aerosol is presented in Sect. . The ability to resolve single ${\mathrm{CO}}_{\mathrm{2}}$ emission plumes at an urban scale is further simulated in Sect. . A short summary and our concluding remarks are finally presented in Sect. .

2 Mission and instrument concept

The instrument concept presented in this paper is based on a spaceborne push-broom imaging grating spectrometer measuring spectra of reflected solar radiation in one single SWIR spectral window, from which the column-averaged dry-air mole fraction of ${\mathrm{CO}}_{\mathrm{2}}$ (X${\mathrm{CO}}_{\mathrm{2}}$) can be retrieved. With an expected instrument mass of approximately 90 kg, it is suitable for deployment on small satellite buses. Since the proposed instrument targets the quantification of localized ${\mathrm{CO}}_{\mathrm{2}}$ emissions from, e.g., coal-fired power plants, a high spatial resolution of $\mathrm{50}\times \mathrm{50}$ ${\mathrm{m}}^{\mathrm{2}}$ is envisaged. The instrument is designed to fly in a sun-synchronous orbit at an altitude of 600 km and a local equatorial crossing time at 13:00 LT. This orbit is chosen in order to have a well-developed boundary layer at overpass together with good radiometric performance (high SNR).

Table 1

Mission and instrument design parameters of the proposed spaceborne ${\mathrm{CO}}_{\mathrm{2}}$ monitoring instrument concept.

[Figure omitted. See PDF]

Figure a shows the continuum SNR (calculated with Eqs. –) as a function of the scene brightness for the two prospective spectral setups SWIR-1 and SWIR-2. The scene brightness describes the conversion from incident solar irradiance to reflected solar radiance and is calculated as the product of the surface albedo and the cosine of the SZA divided by $\mathit{\pi }$, hence assuming a Lambertian surface. For the reference scene (albedo 0.1, SZA 70${}^{\circ }$), the continuum SNR is approximately 180 and 100 for SWIR-1 and SWIR-2, respectively. The consistently higher SNR for SWIR-1 compared to SWIR-2 is mainly the result of higher solar radiance (see Fig. ) and generally higher surface albedo (see, e.g., Fig. 7 in ) in SWIR-1. Looking at the individual contributions from the different instrument noise sources in Fig. b, it is clear that the readout noise and signal shot noise are the major contributors, whereas the noise arising from quantization errors, dark current and thermal background radiation has a small or even negligible contribution in comparison. The signal shot noise is, however, smaller than the dark current, readout noise and quantization noise inside the ${\mathrm{CO}}_{\mathrm{2}}$ absorption bands, in which the signal, and hence the signal shot noise, decreases. Note that all noise terms except for the signal shot noise, ${\mathit{\sigma }}_{\mathrm{ss}}$, are constant.

4 Generic performance evaluation

In this section we conduct a first performance evaluation of the proposed instrument concept by assessing the X${\mathrm{CO}}_{\mathrm{2}}$ retrieval errors expected on a global scale. Such errors arise due to instrument noise and because of inadequate knowledge about the light path through the atmosphere due to scattering aerosol and cirrus particles. For this purpose we use a global trial ensemble with a large collection of geophysical scenarios with varying atmospheric gas concentrations, meteorological conditions, surface albedo, SZA, and aerosol and cirrus compositions that can be expected to be observed by a polar-orbiting instrument. The same methodology and dataset have been used in several previous studies to assess the greenhouse gas retrieval performance of different satellite instruments .

The global trial ensemble contains geophysical data representative for the months of January, April, July and October. Atmospheric gas concentrations stem from the CarbonTracker model ${\mathrm{CO}}_{\mathrm{2}}$ for the year 2010;, the Tracer Model 4 ${\mathrm{CH}}_{\mathrm{4}}$ for the year 2006; and the ECHAM5-HAM model ${\mathrm{H}}_{\mathrm{2}}\mathrm{O}$;. Surface albedo data representative for the SWIR-1 and SWIR-2 windows stem from the MODIS (Moderate Resolution Imaging Spectroradiometer) MCD43A4 product . Aerosol optical properties are calculated (assuming Mie scattering) for an aerosol size distribution superimposed from seven lognormal size distributions and five chemical types at 19 vertical layers, as provided by the ECHAM5-HAM model . Cirrus optical properties are calculated for randomly oriented hexagonal columns and plates following the ray-tracing model of and . In total the global trial ensemble consists of approximately 10 000 scenes with X${\mathrm{CO}}_{\mathrm{2}}$ ranging from 340 to 400 ppm with an average of 382 ppm, albedo ranging from 0 to 0.7 with an average of 0.13 (SWIR-2 window), aerosol optical thickness (AOT) ranging from 0 to 1.0 with an average of 0.05 (SWIR-2 window) and cirrus optical thickness (COT) ranging from 0 to 0.8 with an average of 0.13 (SWIR-2 window). Thus, the global trial ensemble contains challenging scenes with scattering loads that would be filtered out by current satellite retrievals, such as those applied to OCO-2 and GOSAT data, which typically screen scenes with scattering optical thickness greater than 0.3 (at the ${\mathrm{O}}_{\mathrm{2}}$ A band around 760 $\mathrm{nm}$). All data in the global trial ensemble are re-gridded to a spatial resolution of approximately 2.8${}^{\circ }$ $\times$ 2.8${}^{\circ }$. This is, of course, much coarser than the envisaged $\mathrm{50}\times \mathrm{50}$ ${\mathrm{m}}^{\mathrm{2}}$, but for investigating the propagation of instrument noise into the target quantity X${\mathrm{CO}}_{\mathrm{2}}$ on a global scale, this dataset serves its purpose. See previous studies e.g., for further details on the content of the global trial ensemble.

The geophysical data for each scene are fed to the radiative transfer software RemoTeC in order to simulate corresponding synthetic measurements. The measurement noise is calculated by propagating the instrument's SNR (Sect. ) into a statistical error estimate according to the rules of Gaussian error propagation . Simulations are conducted globally for the 16th day of each of the four months January, April, July and October, hence covering SZA conditions ranging from 0 to 86${}^{\circ }$.

By retrieving X${\mathrm{CO}}_{\mathrm{2}}$ from the simulated synthetic spectra, the range of X${\mathrm{CO}}_{\mathrm{2}}$ retrieval errors that can be expected with the proposed instrument concept can be estimated, as can the ability to account for atmospheric aerosol. The RemoTeC retrieval algorithm e.g., is based on a Philipps–Tikhonov regularization scheme that uses the first-order difference operator as a side constraint to retrieve the ${\mathrm{CO}}_{\mathrm{2}}$ partial column profiles, from which X${\mathrm{CO}}_{\mathrm{2}}$ can be determined. Additional retrieval parameters are the total column concentrations of ${\mathrm{H}}_{\mathrm{2}}\mathrm{O}$ and ${\mathrm{CH}}_{\mathrm{4}}$ (only for SWIR-1), surface albedo (as a second-order polynomial), spectral shift, solar shift, and possibly information on scattering aerosol. Here we assume knowledge about the air mass (needed to calculate X${\mathrm{CO}}_{\mathrm{2}}$), but in reality meteorological and topography data would be required to estimate the air mass.

4.1

(a, b) X${\mathrm{CO}}_{\mathrm{2}}$ retrieval errors as a function of the total particulate optical thickness $\text{AOT}+\text{COT}$ for scattering (a) and non-scattering (b) RemoTeC retrievals. (c, d) X${\mathrm{CO}}_{\mathrm{2}}$ retrieval errors as a function of SWIR-2 surface albedo for scattering (c) and non-scattering (d) RemoTeC retrievals. $N$ denotes the number of retrievals. Note the different ranges of the $y$ axes in the upper and lower panels.

[Figure omitted. See PDF]

Figure a shows the difference between the X${\mathrm{CO}}_{\mathrm{2}}$ retrieved (“retr”) from the synthetic measurements and the corresponding “true” X${\mathrm{CO}}_{\mathrm{2}}$ used as input to simulate these synthetic measurements. This deviation from the truth contains information on both random instrument noise error (Sect. ) and systematic errors arising from insufficient modeling of the aerosol and cirrus properties. For comparison, Fig. b shows the corresponding results achieved when using a non-scattering retrieval, i.e., when the scattering by atmospheric aerosol and cirrus, now present in the atmosphere and the simulated synthetic spectra, is neglected (similar to Sect. ). The retrieval errors are strongly reduced when the RemoTeC retrieval algorithm accounts for the scattering by atmospheric aerosol. When scattering is considered, half of the X${\mathrm{CO}}_{\mathrm{2}}$ retrievals deviate from the true abundance by less than 2.5 ppm, while two-thirds of the retrievals deviate by less than 4 ppm (approx. 1 %), with no clear error correlation with the optical thickness of the scattering particles. For the non-scattering retrieval, the corresponding numbers are 16 and 28 ppm, with a mean bias of $-\mathrm{25}$ ppm that increases with optical thickness, exposing the necessity of accounting for atmospheric aerosol and cirrus when retrieving the X${\mathrm{CO}}_{\mathrm{2}}$.

Scattering particles can modify the light path and hence the X${\mathrm{CO}}_{\mathrm{2}}$ retrieval in primarily two ways. Firstly, an elevated layer of aerosol or cirrus will scatter parts of the incoming solar radiation towards the observing sensor at a higher altitude compared to the Earth's surface, leading to a reduced light path. Secondly, aerosol and cirrus will extend the light path to some degree as a result of multiple scattering between scattering particles and the surface. Such modifications of the light path will be understood as ${\mathrm{CO}}_{\mathrm{2}}$ concentrations in the atmosphere that are either too low (overall reduced light path) or too high (overall extended light path) if scattering cannot be accounted for in the retrieval. Which effect dominates is primarily driven by the surface albedo. This is visualized in Fig. d that shows the difference between retrieved and true X${\mathrm{CO}}_{\mathrm{2}}$ as a function of the surface albedo when scattering by aerosol and cirrus is neglected in the retrieval. Over darker surfaces for which the effect of multiple scattering between aerosol and the surface is limited, aerosol and cirrus particles scattering the incoming solar radiation towards the sensor higher up in the atmosphere become the dominating effect, leading to a reduced light path and underestimation of the X${\mathrm{CO}}_{\mathrm{2}}$. Over brighter surfaces for which the effect of multiple scattering becomes dominant, the non-scattering retrieval is more likely to overestimate the ${\mathrm{CO}}_{\mathrm{2}}$ abundance because the loss of radiation due to an extended light path, resulting from the multiple scattering, is assumed to be caused by more absorbing ${\mathrm{CO}}_{\mathrm{2}}$ molecules in the atmosphere. Figure c shows the difference between retrieved and true X${\mathrm{CO}}_{\mathrm{2}}$ as a function of the surface albedo when scattering by aerosol and cirrus is accounted for when retrieving X${\mathrm{CO}}_{\mathrm{2}}$ from the synthetic measurements of the proposed satellite concept. It is clear that when aerosol properties are retrieved alongside the ${\mathrm{CO}}_{\mathrm{2}}$ abundance, the curve-shaped relationship between the X${\mathrm{CO}}_{\mathrm{2}}$ error and surface albedo vanishes with no clear error correlation other than X${\mathrm{CO}}_{\mathrm{2}}$ errors increasing with decreasing albedo (and thus SNR). Note that errors arising from the Lambertian albedo assumption (BRDF – bidirectional reflectance distribution function – effects) are neglected in the scattering simulations.

Although layers of aerosol and cirrus can be partly accounted for in the retrieval, scenes with thicker clouds and aerosol layers will have to be identified and filtered out in the data processing chain. Such a cloud filter could exploit the different optical depths of the two ${\mathrm{CO}}_{\mathrm{2}}$ bands in the SWIR-2 window by retrieving X${\mathrm{CO}}_{\mathrm{2}}$ from the two ${\mathrm{CO}}_{\mathrm{2}}$ bands independently (assuming a non-scattering atmosphere) and filtering for discrepancies.

5 Performance evaluation for an urban case study

While the previous section assessed X${\mathrm{CO}}_{\mathrm{2}}$ errors for the range of geophysical conditions expected to be encountered on a global scale, this section evaluates the ${\mathrm{CO}}_{\mathrm{2}}$ monitoring capabilities at an urban scale using high-resolution ${\mathrm{CO}}_{\mathrm{2}}$ concentration and surface albedo data. Similar to Sect. , the high-resolution data are used to simulate synthetic measurements, from which synthetic X${\mathrm{CO}}_{\mathrm{2}}$ abundances can be retrieved in order to make a first assessment of the ${\mathrm{CO}}_{\mathrm{2}}$ monitoring ability of the proposed instrument concept in terms of resolving ${\mathrm{CO}}_{\mathrm{2}}$ emission plumes.

5.1 Datasets

5.1.1

${\mathrm{CO}}_{\mathrm{2}}$ concentration field from the Hestia dataset

To compute a high-resolution three-dimensional field of ${\mathrm{CO}}_{\mathrm{2}}$ concentrations to be used as input for the radiative transfer simulations, annual estimates of fossil fuel ${\mathrm{CO}}_{\mathrm{2}}$ emissions for the city of Indianapolis in the year 2015 are used. These data are generated by the Hestia Project for which the fossil fuel ${\mathrm{CO}}_{\mathrm{2}}$ emissions are quantified in urban areas down to the scale of individual buildings and streets using a bottom-up approach. The results for the city of Indianapolis are gridded and archived at a spatial resolution of $\mathrm{200}\times \mathrm{200}$ ${\mathrm{m}}^{\mathrm{2}}$. For this study, however, the Hestia Project dataset was gridded to $\mathrm{50}\times \mathrm{50}$ ${\mathrm{m}}^{\mathrm{2}}$ via request to the Hestia research team in order to match the envisaged spatial resolution of the proposed instrument concept. The fossil fuel ${\mathrm{CO}}_{\mathrm{2}}$ emission rates for Indianapolis at $\mathrm{50}\times \mathrm{50}$ ${\mathrm{m}}^{\mathrm{2}}$ resolution can be seen in Fig. a. ${\mathrm{CO}}_{\mathrm{2}}$ emissions from different sources and sectors like road traffic and point sources (single yellow pixels) can be seen. There is also an apparent emission gradient, with stronger emissions in the city center and weaker emissions towards the suburbs. Hence, the Hestia ${\mathrm{CO}}_{\mathrm{2}}$ emission data for Indianapolis provide a realistic emission scenario for evaluating the ${\mathrm{CO}}_{\mathrm{2}}$ monitoring capabilities of the proposed instrument concept. Moreover, the area of the Hestia domain (approx. $\mathrm{34}\times \mathrm{33}$ ${\mathrm{km}}^{\mathrm{2}}$) is comparable to what the prospective tile size of each observation target area could be.

The Hestia ${\mathrm{CO}}_{\mathrm{2}}$ emission data are used as input to a Gaussian dispersion model in order to compute a three-dimensional ${\mathrm{CO}}_{\mathrm{2}}$ concentration field. For a given ${\mathrm{CO}}_{\mathrm{2}}$ emission rate $Q$ ($\mathrm{g}{\mathrm{s}}^{-\mathrm{1}}$), the ${\mathrm{CO}}_{\mathrm{2}}$ concentration $C$ ($\mathrm{g}{\mathrm{m}}^{-\mathrm{3}}$) at a given position ($x,y,z$) downwind of the emitter is calculated as 6$$\begin{array}{rl}C(x,y,z)=& {\displaystyle \frac{Q}{\mathrm{2}\mathit{\pi }u{\mathit{\sigma }}_y{\mathit{\sigma }}_z}}\exp \left({\displaystyle \frac{-y^{\mathrm{2}}}{\mathrm{2}{\mathit{\sigma }}_y^{\mathrm{2}}}}\right)\\ & \left[\exp \right({\displaystyle \frac{-(z-h{)}^{\mathrm{2}}}{\mathrm{2}{\mathit{\sigma }}_z^{\mathrm{2}}}})+\exp ({\displaystyle \frac{-(z+h{)}^{\mathrm{2}}}{\mathrm{2}{\mathit{\sigma }}_z^{\mathrm{2}}}}\left)\right],\end{array}$$ where $u$ is the horizontal wind speed in the $x$ direction (along-wind), $h$ is the height of the emitting source (in meters above ground level), and ${\mathit{\sigma }}_y$ and ${\mathit{\sigma }}_z$ are the standard deviations of the concentration distribution (in meters) in the horizontal across-wind and vertical dimension, respectively; ${\mathit{\sigma }}_y$ and ${\mathit{\sigma }}_z$, and hence the spread of the emission plume, depend on the atmospheric instability, i.e., the degree of atmospheric turbulence and the downwind distance $x$ from the emitting source. Here, we calculate ${\mathit{\sigma }}_y$ and ${\mathit{\sigma }}_z$ assuming the Pasquill–Gifford stability class C (slightly unstable atmosphere). Furthermore, a constant wind speed $u=\mathrm{3}$ $\mathrm{m}{\mathrm{s}}^{-\mathrm{2}}$ and an emitting source height $h=\mathrm{75}$ m (for all sources) are assumed. This model setup is comparable to similar studies e.g.,.

Figure 7

(a) Hestia fossil fuel ${\mathrm{CO}}_{\mathrm{2}}$ emission data for Indianapolis in 2015 at $\mathrm{50}\times \mathrm{50}$ ${\mathrm{m}}^{\mathrm{2}}$ spatial resolution. (b) Corresponding field of vertically integrated X${\mathrm{CO}}_{\mathrm{2}}$ enhancements at $\mathrm{50}\times \mathrm{50}$ ${\mathrm{m}}^{\mathrm{2}}$ spatial resolution with respect to a constant background, computed using the Hestia ${\mathrm{CO}}_{\mathrm{2}}$ emission data and a Gaussian dispersion model. Panel (c) is the same as (b), but at $\mathrm{2}\times \mathrm{2}$ ${\mathrm{km}}^{\mathrm{2}}$ spatial resolution. (d) Per-pixel X${\mathrm{CO}}_{\mathrm{2}}$ enhancements for three along-track excerpts centered at 400, 1500 and 4000 m downwind of the emitter at $\mathrm{50}\times \mathrm{50}$ ${\mathrm{m}}^{\mathrm{2}}$ and $\mathrm{2}\times \mathrm{2}$ ${\mathrm{km}}^{\mathrm{2}}$ spatial resolution. The respective positions of the along-track excerpts are indicated by the small grey lines in (b) and (c). The $x$ and $y$ dimensions of the Hestia Indianapolis domain are illustrated as hypothetical satellite across-track and along-track dimensions, respectively.

[Figure omitted. See PDF]

Downwind ${\mathrm{CO}}_{\mathrm{2}}$ concentrations from each emitting source (pixel) in the Hestia dataset are calculated across an equidistant grid at 50 m resolution in all dimensions, and the contributions from all individual emitting sources (pixels) are subsequently combined to form a three-dimensional ${\mathrm{CO}}_{\mathrm{2}}$ concentration field over Indianapolis. Figure b shows the resulting (vertically integrated) two-dimensional field of (noiseless) X${\mathrm{CO}}_{\mathrm{2}}$ enhancements at $\mathrm{50}\times \mathrm{50}$ ${\mathrm{m}}^{\mathrm{2}}$ spatial resolution over a constant background with a surface pressure of 1013 hPa. While weaker diffuse sources like streets cannot be identified, the plumes from stronger point sources are clearly pronounced given the high spatial resolution that allows for a detailed mapping of the plumes. For comparison, Fig. c shows the corresponding X${\mathrm{CO}}_{\mathrm{2}}$ enhancements assuming a coarser spatial resolution of $\mathrm{2}\times \mathrm{2}$ ${\mathrm{km}}^{\mathrm{2}}$. Although the stronger plumes can still be identified at the coarser resolution, the X${\mathrm{CO}}_{\mathrm{2}}$ enhancements are significantly lower and each plume is only sampled by a few pixels. Figure d further shows these X${\mathrm{CO}}_{\mathrm{2}}$ enhancements in more detail for three along-track excerpts centered at 400, 1500 and 4000 m downwind of the strongest emitter in Indianapolis, with an annual emission rate of 3.24 $\mathrm{Mt}{\mathrm{CO}}_{\mathrm{2}}{\mathrm{yr}}^{-\mathrm{1}}$ in 2015. The positions of the three along-track excerpts are indicated with grey lines in Fig. b and c. With a spatial resolution of $\mathrm{2}\times \mathrm{2}$ ${\mathrm{km}}^{\mathrm{2}}$, the along-track plume excerpts are only sampled by one pixel each, with a maximum X${\mathrm{CO}}_{\mathrm{2}}$ enhancement of 1.2 ppm. With the envisaged $\mathrm{50}\times \mathrm{50}$ ${\mathrm{m}}^{\mathrm{2}}$ spatial resolution, however, the plume is sampled by 7, 15 and 29 pixels in the along-track dimension 400, 1500 and 4000 m downwind of the emitter, respectively, with maximum X${\mathrm{CO}}_{\mathrm{2}}$ enhancements reaching approximately 18, 6 and 3 ppm, respectively. This clearly demonstrates the benefit of an instrument with a high spatial resolution when resolving ${\mathrm{CO}}_{\mathrm{2}}$ emission plumes from space.

5.1.2 Surface albedo data from Sentinel-2

To accurately simulate the instrument SNR and hence the measurement noise, it is important to know how large a fraction of the solar radiation incident on the Earth's surface is reflected back towards space. To get realistic estimates of the surface albedo within the Hestia Indianapolis domain, data from the European Sentinel-2 satellites are used. The multispectral instrument aboard Sentinel-2 measures the TOA radiance in 13 spectral bands with a spatial resolution ranging from $\mathrm{10}\times \mathrm{10}$ ${\mathrm{m}}^{\mathrm{2}}$ to $\mathrm{60}\times \mathrm{60}$ ${\mathrm{m}}^{\mathrm{2}}$. For this study, we use the Sentinel-2 L1C radiances measured in spectral band 12 (centered at approx. 2200 nm) at a spatial resolution of $\mathrm{20}\times \mathrm{20}$ ${\mathrm{m}}^{\mathrm{2}}$. The software Sen2Cor is employed to compute the corresponding L2 surface reflectances from the L1C TOA radiances through a so-called atmospheric correction.

Surface reflectance data for the month of July 2018 are computed and re-gridded (using nearest neighbor) to the envisaged spatial resolution of $\mathrm{50}\times \mathrm{50}$ ${\mathrm{m}}^{\mathrm{2}}$. The surface reflectance for Sentinel-2 pixels classified as vegetation is scaled by a factor of 0.82 in order to account for the generally lower reflectance by vegetation in the SWIR-2 window compared to Sentinel-2's band 12. The scaling factor has been derived using spectral reflectance data from the ECOSTRESS spectral library . Figure b shows the gridded surface reflectance data for Indianapolis together with a corresponding red–green–blue (RGB) composite (Fig. a) using Sentinel-2 data from the bands centered at red, green and blue wavelengths as a reference. The scaled and gridded Sentinel-2 surface reflectance data are taken as representative for a constant Lambertian surface albedo within the entire SWIR-2 window. Figure S1 in the Supplement shows spectral reflectances in the SWIR spectral region for various urban materials using data from the Spectral Library of impervious Urban Materials SLUM;. The small spectral variations within the SWIR-2 spectral window used in this study indicate that the true spectral signatures of the surface albedo could be fitted with sufficient precision using the second-order polynomial during the retrieval, and hence the assumption of a constant albedo is reasonable for this study.

The average surface reflectance within the Hestia domain is 0.13. Despite annual variability in surface reflectance, mainly due to changes in vegetation and/or crops, this is a value representative throughout most of the year. For comparison, average surface reflectances from the same source for January (snow-free days), April and October 2018 amount to 0.11, 0.17 and 0.11, respectively.

5.1.3 Background data from CarbonTracker

Background data, including vertical profiles of ${\mathrm{CO}}_{\mathrm{2}}$, ${\mathrm{H}}_{\mathrm{2}}\mathrm{O}$, temperature and pressure, are taken for 15 July 2016 from the CarbonTracker CT2017 dataset with updates documented at http://carbontracker.noaa.gov, last access: 18 March 2019. The CarbonTracker CT2017 data over Indianapolis are provided at a spatial resolution of 1${}^{\circ }$ $\times$ 1${}^{\circ }$, meaning that the entire Hestia Indianapolis domain is covered by one single CarbonTracker pixel, leading to a constant background data field.

5.2

(a) X${\mathrm{CO}}_{\mathrm{2}}$ enhancements with respect to the background retrieved from the simulated synthetic measurements over the Hestia Indianapolis domain under non-scattering conditions. Locations of the four strongest point sources are labeled with $E_{\mathrm{1}-\mathrm{4}}$. (b) Corresponding deviations between retrieved (“retr”) and true X${\mathrm{CO}}_{\mathrm{2}}$. Dark scenes with albedo $<\mathrm{0.05}$ have been filtered out due to unreliable X${\mathrm{CO}}_{\mathrm{2}}$ retrievals.

[Figure omitted. See PDF]

Figure a shows the retrieved field of X${\mathrm{CO}}_{\mathrm{2}}$ enhancements with respect to the retrieved background X${\mathrm{CO}}_{\mathrm{2}}$ over the Hestia domain. The ${\mathrm{CO}}_{\mathrm{2}}$ plume from the strongest point source, $E_{\mathrm{1}}$, with an annual ${\mathrm{CO}}_{\mathrm{2}}$ emission rate of $Q_{\mathrm{1}}=\mathrm{3.24}$ $\mathrm{Mt}{\mathrm{CO}}_{\mathrm{2}}{\mathrm{yr}}^{-\mathrm{1}}$, is clearly resolved with local X${\mathrm{CO}}_{\mathrm{2}}$ enhancements well above 100 ppm close to the emitting source. Although they emit considerably less ${\mathrm{CO}}_{\mathrm{2}}$, the plumes from the second- and third-strongest point sources, $E_{\mathrm{2}}$ and $E_{\mathrm{3}}$, with annual ${\mathrm{CO}}_{\mathrm{2}}$ emission rates of $Q_{\mathrm{2}}=\mathrm{0.55}$ $\mathrm{Mt}{\mathrm{CO}}_{\mathrm{2}}{\mathrm{yr}}^{-\mathrm{1}}$ and $Q_{\mathrm{3}}=\mathrm{0.48}$ $\mathrm{Mt}{\mathrm{CO}}_{\mathrm{2}}{\mathrm{yr}}^{-\mathrm{1}}$, respectively, can be clearly separated from the background as well. The plume from the fourth-strongest point source, $E_{\mathrm{4}}$, with an annual ${\mathrm{CO}}_{\mathrm{2}}$ emission rate of $Q_{\mathrm{4}}=\mathrm{0.32}$ $\mathrm{Mt}{\mathrm{CO}}_{\mathrm{2}}{\mathrm{yr}}^{-\mathrm{1}}$, can also be observed but is partly obscured by filtered-out dark surface areas for which retrieval errors are too high. Plumes from weaker point sources ($\mathit{\lesssim }\mathrm{0.1}$ $\mathrm{Mt}{\mathrm{CO}}_{\mathrm{2}}{\mathrm{yr}}^{-\mathrm{1}}$) and other sources like streets and highways cannot be identified given the spatial resolution and instrument noise errors of the proposed instrument.

Figure 10

Upper panels: retrieved two-dimensional fields of X${\mathrm{CO}}_{\mathrm{2}}$ enhancements in the vicinity of the four strongest ${\mathrm{CO}}_{\mathrm{2}}$ emitters $E_{\mathrm{1}}$, $E_{\mathrm{2}}$, $E_{\mathrm{3}}$ and $E_{\mathrm{4}}$ within the Hestia Indianapolis dataset. Lower panels: corresponding per-pixel (circles) and average (solid lines) along-track X${\mathrm{CO}}_{\mathrm{2}}$ enhancements within the area 200 to 2200 m downwind and $-\mathrm{1000}$ to 1000 across-wind of the respective emitters. The blue rectangles in the upper panels show the areas from which the corresponding per-pixel and average along-track X${\mathrm{CO}}_{\mathrm{2}}$ enhancements, depicted in the respective lower panels, are extracted and calculated. The color of the circles follows the color bars in the respective upper panels.

[Figure omitted. See PDF]

One concern with high-resolution ${\mathrm{CO}}_{\mathrm{2}}$ remote sensing is the impact of albedo heterogeneity at an urban scale at such a high spatial resolution. For the non-scattering scenario simulated here, the albedo fitted by the retrieval algorithm matches the reference input albedo with an average (absolute) deviation of 0.14 %, and there is consequently no spatial variability in the accuracy of the albedo retrieval that in turn affects the X${\mathrm{CO}}_{\mathrm{2}}$ retrieval accuracy. There is, however, the evident effect that a higher albedo leads to a higher SNR and hence a generally lower noise error. This is evident from Fig. b, showing the difference between the retrieved and true X${\mathrm{CO}}_{\mathrm{2}}$, thus illustrating an instantaneous noise error field that would be expected for a single satellite overpass. Generally, the deviations from the true X${\mathrm{CO}}_{\mathrm{2}}$ are smaller over areas of brighter surfaces like concrete, whereas the deviations are larger over dark surfaces like forests (see also Fig. ). The effect of albedo heterogeneity in combination with scattering particles is not addressed in this paper and will have to be analyzed in future studies.

Across the entire Hestia domain (but excluding dark scenes with albedo $<\mathrm{0.05}$) 68 % and 95 % of the X${\mathrm{CO}}_{\mathrm{2}}$ retrievals deviate from the true X${\mathrm{CO}}_{\mathrm{2}}$ by less than 1.1 and 2.3 ppm, respectively. This is comparable to the noise error obtained for the global trial ensemble in Sect.  (1.1 and 2.0 ppm, respectively).

Figure  shows close-ups of the simulated X${\mathrm{CO}}_{\mathrm{2}}$ enhancement field (upper panels) in the vicinity of the four strongest emitters in the Hestia Indianapolis dataset ($E_{\mathrm{1}}$–$E_{\mathrm{4}}$ in Fig. a), along with the corresponding per-pixel and average along-track X${\mathrm{CO}}_{\mathrm{2}}$ enhancements (lower panels) for the range 200–2200 m downwind of the respective emitting sources. Enhancements from the 200 m closest to each emitting source are excluded as those scenes could likely be obscured by condensate in a real situation.

The plume of the strongest emitter $E_{\mathrm{1}}$ in Indianapolis with an annual emission rate of $Q_{\mathrm{1}}=\mathrm{3.24}$ $\mathrm{Mt}{\mathrm{CO}}_{\mathrm{2}}{\mathrm{yr}}^{-\mathrm{1}}$ is clearly resolved. Within the area 200–2200 m downwind of the emitting source (blue square), maximum enhancements exceed 25 ppm, and in total, approximately 200 (60) pixels have enhancements above 4 ppm (8 ppm), representing enhancements of approximately 1 % (2 %) with respect to the background. The average along-track X${\mathrm{CO}}_{\mathrm{2}}$ enhancement 200–2200 m downwind of the emitting source (blue and white line) reaches 12 ppm. The plumes from the second- and third-strongest emitters, $E_{\mathrm{2}}$ and $E_{\mathrm{3}}$, which are approximately 6 times weaker than $E_{\mathrm{1}}$ with annual emission rates of $Q_{\mathrm{2}}=\mathrm{0.55}$ and $Q_{\mathrm{3}}=\mathrm{0.48}$ $\mathrm{Mt}{\mathrm{CO}}_{\mathrm{2}}{\mathrm{yr}}^{-\mathrm{1}}$, respectively, have considerably lower X${\mathrm{CO}}_{\mathrm{2}}$ enhancements but can nevertheless clearly be separated from the background with distinct increments in both per-pixel and average X${\mathrm{CO}}_{\mathrm{2}}$ enhancements within the area 200–2200 m downwind of the emitters (blue squares). While the background fields vary from approximately $-\mathrm{1}$ to 1 ppm due to instrument noise, the per-pixel plume enhancements vary from approximately 0.5 to 3 ppm, with single enhancements exceeding 4 ppm close to the emitting source. The average along-track X${\mathrm{CO}}_{\mathrm{2}}$ enhancements 200–2200 m downwind of the emitting sources (blue and white lines) reach 1.9 and 1.5 ppm for $E_{\mathrm{2}}$ and $E_{\mathrm{3}}$, respectively. Despite being partly obscured by filtered-out dark surfaces (water), the plume from the fourth-strongest emitter $E_{\mathrm{4}}$, with an annual emission rate of $Q_{\mathrm{4}}=\mathrm{0.32}$ $\mathrm{Mt}{\mathrm{CO}}_{\mathrm{2}}{\mathrm{yr}}^{-\mathrm{1}}$, can also be separated from the background both when looking at the two-dimensional field and the per-pixel enhancements within the area 200–2200 m downwind of the emitter. With maximum average X${\mathrm{CO}}_{\mathrm{2}}$ enhancements of at most approximately 1.3 ppm, the proposed instrument concept is, however, approaching the limit of what it could achieve in terms of ${\mathrm{CO}}_{\mathrm{2}}$ plume observations under favorable conditions, i.e., when the effect of aerosol-induced errors is neglected and the SZA is relatively low. A second peak in the average along-track X${\mathrm{CO}}_{\mathrm{2}}$ enhancements is observed approximately 850 m above (north of) the fourth-strongest emitter $E_{\mathrm{4}}$. This enhancement stems from the ${\mathrm{CO}}_{\mathrm{2}}$ plume from the seventh-strongest emitter in Indianapolis (labeled $E_{\mathrm{7}}$ in the top right panel of Fig. ), with an annual emission rate of $Q_{\mathrm{7}}=\mathrm{0.1}$ $\mathrm{Mt}{\mathrm{CO}}_{\mathrm{2}}{\mathrm{yr}}^{-\mathrm{1}}$. Quantifying the ${\mathrm{CO}}_{\mathrm{2}}$ emission rate from such a weak source is, however, not realistic given the low sampling density (especially further downwind) in combination with the weak per-pixel enhancements.

6 Conclusions

To follow the progress on reducing anthropogenic ${\mathrm{CO}}_{\mathrm{2}}$ emissions worldwide, independent monitoring systems are of key importance. In this paper, we present the concept of a compact spaceborne imaging spectrometer with a high spatial resolution of $\mathrm{50}\times \mathrm{50}$ ${\mathrm{m}}^{\mathrm{2}}$ targeting the monitoring of localized ${\mathrm{CO}}_{\mathrm{2}}$ emissions. We further demonstrate how the instrument concept could resolve ${\mathrm{CO}}_{\mathrm{2}}$ emission plumes from localized point sources like medium-sized power plants, thus having the potential to contribute to the independent large-scale verification of reported ${\mathrm{CO}}_{\mathrm{2}}$ emissions at the facility level.

Through radiative transfer simulations using a global trial ensemble, a preliminary yet realistic instrument design and an instrument noise model, we show that the expected instrument-noise-induced X${\mathrm{CO}}_{\mathrm{2}}$ errors are smaller than 1.1 and 2.0 ppm for 68 % and 95 % of the retrievals, respectively, using the SWIR-2 spectral setup covering the ${\mathrm{CO}}_{\mathrm{2}}$ absorption bands near 2000 nm. For the SWIR-1 spectral setup covering the weaker ${\mathrm{CO}}_{\mathrm{2}}$ absorption bands near 1600 nm, the instrument-noise-induced X${\mathrm{CO}}_{\mathrm{2}}$ errors are significantly higher, making it inadequate for the proposed instrument concept. Although the main focus in this paper is on the performance of the proposed ${\mathrm{CO}}_{\mathrm{2}}$ monitoring instrument concept, we could also show that despite the usage of a single spectral window and a relatively coarse spectral resolution of $\mathrm{1.29}$ nm, scattering by highly complex atmospheric aerosol compositions can be partly accounted for during X${\mathrm{CO}}_{\mathrm{2}}$ retrievals on the global scale, limiting the deviation from the true X${\mathrm{CO}}_{\mathrm{2}}$ to at most 4.0 ppm for two-thirds of the retrievals. This gives us confidence that accurate two-dimensional fields of X${\mathrm{CO}}_{\mathrm{2}}$ enhancements could be retrieved from real spectra measured by the proposed instrument concept. A reasonable a priori state vector with respect to the aerosol properties (e.g., provided through models or a companion aerosol instrument; ) would, however, still be important. As an example, a multi-angle polarimeter instrument is planned to fly together with the ${\mathrm{CO}}_{\mathrm{2}}$ instrument onboard the CO2M mission in order to minimize systematic X${\mathrm{CO}}_{\mathrm{2}}$ errors .

Using high-resolution ${\mathrm{CO}}_{\mathrm{2}}$ emission data for the city of Indianapolis together with a Gaussian dispersion model, corresponding high-resolution albedo data and additional radiative transfer simulations, we have clearly demonstrated that the instrument is well suited for the task of the spaceborne ${\mathrm{CO}}_{\mathrm{2}}$ monitoring of large and medium-sized power plants and can (only limited by its own instrument noise) resolve emission plumes from point sources with an emission source strength down to the order of 0.3 $\mathrm{Mt}{\mathrm{CO}}_{\mathrm{2}}{\mathrm{yr}}^{-\mathrm{1}}$. This is well below the target emission source strength of 1 $\mathrm{Mt}{\mathrm{CO}}_{\mathrm{2}}{\mathrm{yr}}^{-\mathrm{1}}$, hence leaving a significant margin for additional error sources and aspects not yet addressed here.

Given the results from this first performance assessment, the proposed instrument concept demonstrates clear potential for the independent quantification of ${\mathrm{CO}}_{\mathrm{2}}$ emissions from medium-sized power plants (1–10 $\mathrm{Mt}{\mathrm{CO}}_{\mathrm{2}}{\mathrm{yr}}^{-\mathrm{1}}$), which are currently not targeted by other planned spaceborne ${\mathrm{CO}}_{\mathrm{2}}$ monitoring missions. On the local scale (Indianapolis), we have constrained the present analysis to a day in July using a rather simplistic Gaussian dispersion model that assumes constant atmospheric stability and (unidirectional) horizontal wind speed. It might be that the ability to resolve ${\mathrm{CO}}_{\mathrm{2}}$ emission plumes will become more, perhaps even too, challenging under certain more realistic conditions. Nevertheless, these first results are certainly promising and encourage further studies.

The high spatial resolution needed to resolve emission plumes from localized sources like medium-sized power plant does, however, imply limitations in terms of spatial coverage arising from the narrow swath (50 km assuming 1000 detector pixels in the spatial dimension) and the forward motion compensation. Hence, a single satellite with the proposed instrument concept could not quantify ${\mathrm{CO}}_{\mathrm{2}}$ emissions at the local to regional scale with dense global coverage and high temporal resolution but would have to be restricted to some predefined targets. The relatively compact design with a single spectral window could, however, allow for the deployment of a fleet of instruments and hence the independent monitoring of localized ${\mathrm{CO}}_{\mathrm{2}}$ emissions on a larger scale with high temporal resolution. As an alternative to a fleet of satellites, the proposed instrument concept could also prove useful in synergy with a spaceborne ${\mathrm{CO}}_{\mathrm{2}}$ lidar e.g.,; the passive spectrometer would benefit from the lidar's accuracy and knowledge of the light path, and the lidar would benefit from the spectrometer's imaging capability.

With the successful demonstration in this paper, i.e., that ${\mathrm{CO}}_{\mathrm{2}}$ emission plumes from medium-sized power plants can be resolved from space with a compact yet realistic instrument design, the next step will be to analyze the ability to quantify the corresponding ${\mathrm{CO}}_{\mathrm{2}}$ emission rates from the two-dimensional fields of synthetically retrieved X${\mathrm{CO}}_{\mathrm{2}}$ enhancements. This follow-up study will be conducted for different seasons (with varying surface albedo and solar zenith angles), meteorological conditions and emission source strengths using large eddy, rather than Gaussian, modeling of the ${\mathrm{CO}}_{\mathrm{2}}$ plume dispersion. Although the effect of aerosols has partly been assessed on the global scale in this study, information on the properties and distribution of aerosols should also be included in the local-scale simulations in order to better understand the instrument's ability to resolve and quantify localized ${\mathrm{CO}}_{\mathrm{2}}$ emissions under more realistic conditions. Such an in-depth aerosol analysis is, however, the task of further future studies.

Data availability

Hestia Project data at $\mathrm{50}\times \mathrm{50}$ ${\mathrm{m}}^{\mathrm{2}}$ spatial resolution are available from KG upon request (Hestia Project data at original $\mathrm{200}\times \mathrm{200}$ ${\mathrm{m}}^{\mathrm{2}}$ spatial resolution are available at 10.18434/T4/1503341; ). Sentinel-2 data are available at https://scihub.copernicus.eu/dhus/#/home (last access: 9 July 2019). ECOSTRESS Spectral Library data are available at https://speclib.jpl.nasa.gov (last access: 12 February 2019). CarbonTracker CT2017 data are available at http://carbontracker.noaa.gov (last access: 18 March 2019). SLUM data are available at http://www.met.reading.ac.uk/micromet/LUMA/SLUM.html (last access: 18 March 2020).

The supplement related to this article is available online at: https://doi.org/10.5194/amt-13-2887-2020-supplement.

Author contributions

JS developed the instrument noise model, performed the simulations and wrote most of the paper. DK led the instrument design work. JW performed the spectral sizing. CP assisted in developing the instrument noise model and performed the FMC analysis. IS did the optical design. KRG and JL developed and provided the Hestia Project ${\mathrm{CO}}_{\mathrm{2}}$ emission data. JS, DK, JW, CP, IS, AR and AB defined the mission concept and instrument design. AR and AB led the study.

Competing interests

The authors declare that they have no conflict of interest.

Acknowledgements

We kindly acknowledge all persons and institutions that made their data available to us for this study. CARMA v3.0 ${\mathrm{CO}}_{\mathrm{2}}$ emission data for power plants worldwide were provided by Kevin Ummel. Sentinel-2 data used to derive high-resolution surface reflectance data were provided by the ESA trough the Copernicus Open Acess Hub. Spectral reflectance data used to scale the Sentinel-2 surface reflectance data were reproduced from the ECOSTRESS Spectral Library through the courtesy of the Jet Propulsion Laboratory, California Institute of Technology, Pasadena, California, USA. CarbonTracker CT2017 data were provided by NOAA ESRL, Boulder, Colorado, USA. Spectral reflectance data for urban materials (see the Supplement) were reproduced from the Spectral Library of impervious Urban Materials (SLUM) through the courtesy of the University of Reading, UK. We also thank Peter Haschberger, Claas Köhler, Günter Lichtenberg, Andreas Baumgartner, Christoph Kiemle, Luca Bugliaro, Julian Kostinek and Andreas Luther for valuable input on the mission concept, instrument design and/or a previous version of this paper.

Financial support

The article processing charges for this open-access publication were covered by a Research Centre of the Helmholtz Association.

Review statement

This paper was edited by Andreas Richter and reviewed by Luis Guanter and one anonymous referee.