Introduction

UVSQ-SAT is a LATMOS nanosatellite mission with scientific and technological goals mainly for observing essential climate variables, namely shortwave and longwave radiative fluxes at the top of the atmosphere and UV solar spectral irradiance. However, the UVSQ-SAT pathfinder mission will not provide a continuity of the essential climate variables data records since it is a demonstrator. Another objective of the UVSQ-SAT mission is to provide hands-on experience to UVSQ and Paris-Saclay Universities students in the requirements definition, reliability and quality assurance, cost and risk management, design, construction, spacecraft integration and test, mission operations, and control of complete satellite systems that will serve as the basis for a variety of future space missions for Earth Observation and Astronomy & Astrophysics.
The UVSQ-SAT nanosatellite is a cube of about 11 cm with a mass of up to 1.6 kg and a power consumption of up to 2 W. The launch of the CubeSat is currently targeted in the timeframe of 2020/2021. The choice of the orbit is directly related to scientific goals while taking into account the optimization for launch opportunities as piggyback and the rules governing the space debris mitigation. The selected orbit is a Sun-synchronous (SSO) low Earth orbit (LEO) with a maximum altitude of 600 km and a local time at ascending node (LTAN) of 10:30 hours, which will lead to an atmospheric reentry of the satellite within 25 years. The operational mission life time will be at least of one year on orbit, including the commissioning phase, to achieve the expected UVSQ-SAT scientific objectives.
The first scientific objective of the UVSQ-SAT in-orbit demonstration CubeSat is to measure the incoming solar radiation (total solar irradiance) and the outgoing terrestrial radiation (top of atmosphere outgoing longwave radiation and shortwave . radiation) using twelve miniaturized Earth’s radiative sensors (thermopiles based on advantages of carbon nanotubes and Qioptiq optical solar reflectors).Thus, it might be possible to constrain better the Earth’s radiative balance and, more importantly, the Earth’s energy imbalance (EEI), which is defined as the difference between the absorbed incoming solar radiation and the outgoing terrestrial radiation (longwave and shortwave radiation). The direct determination of EEI is very challenging because EEI is two orders of magnitude smaller than the radiation fluxes in and out of the Earth’s system.
The second scientific objective is to monitor the solar spectral irradiance in the Herzberg continuum (200 – 242 nm) using four photodiodes which benefit from intrinsic advantages of Ga2O3 alloy based sensors grown by pulsed laser deposition. A better understanding of natural factors in climate variability is the essential motivation of the UV solar spectral irradiance measurements. The UV solar variability over time has significant implications for atmospheric chemistry and its modeling. The results concern a spatial reconstruction of Earth radiation budget that might be carried out with UVSQ-SAT data for a given time-period of observation. This analysis will also highlight the interest to implement a satellites constellation in order to improve the determination of the EEI, which is a crucial quantity for testing :w climate models and for predicting the future course of global warming. Today, the implementation of a ‘EEI’ constellation based on small satellites is possible. Indeed, the commercial use of small satellites has started thanks to recent advances in miniaturization and integration. Many fields start benefiting from small satellites: scientific research, technology demonstrations, Earth observations, biological experiments/pharmaceuticals, telecommunications, military applications, etc. Small satellites and the ‘NewSpace’ at Horizon 2020 offer unique opportunities in terms of constellation deployment providing larger simultaneous spatio-temporal coverage of the Earth, which is fundamental for Earth energy imbalance measurements (impacts of aerosols and clouds that are highly variable spatially and temporally).

Earth's energy imbalance

Currently, Earth’s surface temperatures are rising by about 0.2 K per decade since 1981 (considering deseasonalised monthly surface temperature anomalies from HadCRUTv4.5). Thus, climate change and global warming pose a severe threat to humanity. Climate processes are controlled by energy exchanges within and among the different components of the Earth system. Monitoring the Earth’s influx and out flux of both longwave and shortwave radiation from all sources is essential to advance our understanding of climate variability and change, and for developing more accurate and reliable climate models and forecasting. Human activities have led to rising levels of heat-trapping greenhouse gases (GHG) in the atmosphere with less terrestrial radiation being able to escape. This unequivocal anthropogenic radiative forcing of the climate system creates an imbalance in the Earth’s energy budget which causes surface and lower atmospheric warming in order to reestablish a balance in the energy budget. For this reason, the Earth’s energy imbalance (EEI) represents a measure of the excess of energy that is being stored in the climate system as a response to anthropogenic forcing. As such, it has been identified as a fundamental diagnostic for analyzing climate variability and anticipating future climate changes. Direct measurements of variations in the energy entering and leaving the Earth system are of primary importance for determining the rate of climate change at regional and global scales . Actually, the most accurate measurement strategy to determine EEI is to monitor the temporal evolution of the ocean heat content since more than 90% of the excess energy that is gained by the Earth in response to the positive EEI accumulates into the ocean in the form of heat. This can be combined with satellite radiation measurements to derive the high-frequency variability in EEI. Indeed, the absolute value EEI can be best estimated from changes in ocean heat content on long timescales, whereas the high spatiotemporal variations in EEI can be provided by satellite observations of net radiation flux variability at the top of atmosphere (TOA).
The information on EEI at high spatial and temporal resolution is crucial for advancing our understanding of the climate change because the Earth’s radiative balance is partly driven by the radiative impacts of aerosols and clouds which are highly variable spatially and temporally and are still relatively poorly quantified(IPCC, 2014). Satellites remote sensing provide a practical and efficient method for mapping Earth Radiative Balance (ERB) components spatially and temporally at different scales. A large satellite constellation would allow a high frequency and sampling in measurements and consequently a more accurate determination of the Earth’s global energy imbalance along with the diurnal and multi-directional sampling needed to capture spatiotemporal scales relevant to aerosol and clouds (e.g. every 3 hours and ideally few km resolution). Advances in small satellite technology now enable the cost effective global solution of monitoring Earth’s environment with a minimum constellation of 15 small satellites. Cloud data sharing is a cost effective solution for collecting the constellation data and providing high quality science data in near realtime. UVSQ-SAT is one of the first in orbit demonstration CubeSat that is intended to demonstrate the ability to build a low-cost satellite with good precision measurements (relative EEI uncertainty at 1σ of ±5 W m−2 during the mission). Recently, the Radiometer Assessment using Vertically Aligned Nanotubes (RAVAN) 3U CubeSat demonstrated technologies for high accuracy measurement of Earth’s radiation budget. The new UVSQ-SAT concept is designed to explore whether it is possible to achieve the EEI required accuracies using broadband small Earth’s radiative sensors (ERS) onboard multiple satellites (constellation). The main goal of the future satellites constellation is to obtain constant flow of direct measurements from space by using miniaturized instruments (volume, mass, power, telemetry) with narrow and broadband sensors to derive EEI at small spatiotemporal scales with an uncertainty at 1σ of ±1 W m−2 for a 1 – 10 km resolution. For longer timescales, EEI direct measurements are also very challenging with a required measurement uncertainty at 1σ of ±0.1 W m−2 during a decade.
Today, best estimates of the EEI long-term timescales are currently derived from temporal changes in Today, best estimates of the EEI long-term timescales are currently derived from temporal changes in +1.0 W m−2, largely consistent with the radiative forcing caused by anthropogenic greenhouse gases. Hansen et al. inferred a planetary energy imbalance of +0.58±0.15 Wm−2 (Earth is absorbing more energy from the Sun than it is radiating to space as heat) during the 6–yr period 2005 – 2010 using ocean heat content. Recently, Johnson et al. estimated EEI at +0.71±0.10 W m−2 for the period 2005.5 – 2015.5 from ocean heat content changes measured by Argo’s automated floats. Satellites measurements from the former generation of Earth Radiation Budget Experiment (ERBE) sensors, along with the current generation of Clouds and the Earth’s Radiant Energy System (CERES) sensors are the basis of a ERB multi-decadal record at the top of atmosphere. Currently, CERES sensors provide the most reliable and stable TOA flux measurements of the ERB components. However, uncertainties in CERES absolute calibration and in the algorithms used to determine ERB from satellite measurements are too large to enable Earth’s energy imbalance to be quantified accurately.
The CERES data products are more useful for providing the spatial and temporal variability of EEI. Actually, there is a risk of a gap in the ERB data since all current CERES missions are close to the end of their life time after 2026 when only tropical missions such as the Franco-Indian Megha-Tropiques or short-duration missions as ESA-JAXA’s EarthCARE carry ERB instruments. Indeed, a project intended to complete and replace the CERES instruments by the Radiation Budget Instrument (RBI) mission in 2021, in 2026, and in 2031 was canceled by NASA. Since then, the possibility of a constellation of small satellites in orbit before 2026 is being explored. It represents a major challenge and could meet the RBI measurement requires for continuity of the climate data records.
To conclude, a measure of the energy imbalance at the top of the atmosphere is crucial but extremely difficult. It is a key step in the chain linking climate warming to the increase in greenhouse gases. This would be an additional element in the scientific basis for climate change mitigation, notably the magnitude of reduction in GHG emissions required to limit global warming (e.g. 2◦C). Ideally, accurate long-term direct measurements of EEI would confirm the extent of climate warming. Short-term measurements of EEI at high spatiotemporal resolution would allow to constrain better poorly known radiative forcings associated with aerosols, aerosol-cloud interactions, surface albedo, UV solar irradiance, etc. The accurate measurements of solar and terrestrial radiative fluxes at TOA over a wide range of surfaces and conditions (e.g. clear-sky, with/without specific clouds or aerosols) would enable a better evaluation of the overall radiative effects of clouds and aerosols and their representations in climate models. Indeed, aerosols and cloud feedbacks arguably remain the dominant source of uncertainties in climate modelling and of its more societally relevant aspects (e.g. changes in precipitation, etc.) as explained in. The incoming and outgoing shortwave measurements required to calculate EEI also can be used to derive the albedo. Spatially and temporally resolved albedo measurements allow us to observe the impact of changes in land use, aerosols and clouds, in terms of reflection of incident solar radiation back to space - essential for Earth radiation budget and therefore for climate. The spatial and temporal resolution of the radiative measurements determines the scales of the targeted processes. Ideally, the resolution should be high enough to investigate fine-scale processes associated with aerosols and clouds, possibly the most important source of divergence between climate models. A 1 – 10 km resolution would be appropriate for studying local aerosol plumes and clouds. In terms of temporal resolution, being able to follow, even in a crude way, diurnal variations would be a major step forward, in particular for diurnal cycles of clouds or the formation of secondary aerosols (e.g. sulfur, nitrates which are formed by photochemistry). It is worth pointing that the albedo issue is at the heart of geo-engineering (or rather climatic intervention) by solar radiation management, notably using the injection of aerosols or precursors in the atmosphere. High resolution radiative measurements would help to characterize to what extent aerosol affect directly the albedo and indirectly cloud properties on small scales, a sort of today analogues for geo-engineering. More generally, these measurements would help to carry out process studies on the relationship between initial perturbations and atmospheric response at local scales in terms of shortwave and long wave radiation .

Solar spectral irradiance in the Herzberg continuum

The role of solar variability in climate variability remains a topic of considerable scientific and societal importance. Solar radiation is the energy source and is important for climate. The incoming solar flux or/and its spectral distribution at the top of the atmosphere (due to changes in solar activity or in the Earth’s orbital parameters) fluctuate over a wide range of temporal scales, from the 27–day rotational cycle to multi-thousands of years. It also includes 11–yr solar cycles and cycles of the order of hundreds of years, called “grand solar minima” and “grand solar maximai”.
The solar spectrum and its variability represent key inputs not only for solar physics but also for climate physics. Climate models require time-varying solar spectra as forcing with the available information often based on solar reconstructions and solar models. There are multiples lines of evidence showing that solar variability has been a key forcing in the history of the Earth’s climate. Correlations between solar proxies and atmospheric/climate indicators have been established in present-day datasets and in sedimentary and ice core archives. However, most of the apparent correlations and associated solar signals tend to be very variable and intermittent. Some are also very difficult to reproduce in climate models.
Establishing a quantitative forcing-response relationship for the Sun-Earth link is problematic without a clear understanding of the key mechanisms engaged in the action of solar variability on the atmosphere and climate, notably at regional scales. There is no general consensus on those mechanisms. The overall response of the atmosphere and surface climate to solar variability involves a wide range of coupled chemical, dynamical, and radiative processes and the interactions between different atmospheric layers and between the atmosphere and the ocean. It is worth stressing that the issue of the solar impacts is not just critical for paleoclimate. It is also highly relevant for the present-day climate evolution, which is driven by the GHG rising concentrations. Climate change is a major and growing threat to natural, managed and human systems. There is already growing evidence for its adverse impacts on the natural environment and human societies e.g. ecosystems, biological diversity, water resources and economy). There are several sources of uncertainties in climate simulations, in particular in the projections that are used by decision makers to design differentiated mitigation and adaptation strategies. Some of the uncertainties originate from the difficulty to separate the anthropogenic contribution from the natural variability. Quantifying accurately the anthropic contribution and projecting future changes requires understanding and quantifying the natural climate variability including the solar-driven variations. It has even been suggested that a new grand solar minimum might occur in the 21st century and even last until the end of the 22nd century.
The uncertainties are not limited to the mechanisms. They also pertain to the solar variability itself, especially the spectral variations. Indeed, solar forcing is not simply limited to a change in total energy flux. Spectral variations are also important. The relative variations in incoming solar spectral irradiance (SSI) increases very rapidly with decreasing wavelength in the UV range and below. For instance, over an 11–yr cycle, the total solar irradiance (TSI) fluctuates by about 0.1% (∼1.4 W m−2) whereas, in contrast, the radiative flux in the 200 nm region, a key spectral window for stratospheric ozone photochemistry, varies by several %. This has important implications for the way variations in incoming solar energy are redistributed among the different atmospheric layers. The choice of solar UV irradiance variability used to force the models is critical for the solar perturbations of the middle atmosphere.
The exceptionally weak solar cycle 24 and the future solar cycle 25 (expected to begin in late 2019) are interesting periods in this context as they might possibly imply the beginning of a general negative solar forcing which would be expected to be vastly outweighed by the global anthropogenic positive forcing. It is also time to clarify better the mechanisms involved in the solar forcing and atmospheric response. The idea is to investigate carefully processes affecting several atmospheric layers. Historically, the impact of solar variability on surface climate has often been seen as resulting only from the direct radiative effects on the Earth’s surface and the lower atmosphere. In this framework, the drivers are variations in incoming total solar energy (TSI) in wavelength ranges where the middle atmosphere is more or less transparent, i.e. wavelengths longer than 320 nm, corresponding mostly soft UV (UVA), visible, and near infra-red (IR) ranges.
They directly cause changes in the heating rate of the Earth’s surface and the lower atmosphere, modifying surface temperatures and climate. At first order, the change in global temperature is essentially due to this direct effect. However, there is also an indirect effect, the so-called ‘top-down’ mechanism (by opposition to the direct effect referred as the ‘bottom-up’ mechanism). In that case, the drivers are variations in the incoming UV flux (below 320 nm) and energetic particles whose energies are almost entirely absorbed by the middle atmosphere. They cause photochemical and dynamic perturbations of the middle atmosphere which then propagate to the troposphere via stratospheric-tropospheric couplings and result in modifications of surface climate, notably on regional patterns. Both mechanisms (‘top-down’, ‘bottom-up’) operate at the same time 
in reality and influence the middle atmosphere and surface climate. An additional complication in studying the ‘top-down’ mechanism is the fact that UV variations impact the middle atmosphere not only directly via changes in radiative heating but also indirectly via photochemically driven changes in ozone, the key chemical species and UV absorber in the stratosphere. As a result, the stratospheric temperature response to UV changes is amplified by about a factor 2 in a chemistry-climate model (with ozone calculated interactively) compared to the response in climate model with specified constant ozone. Clearly, the ozone response to solar variability needs to be accounted for in models by treating ozone like temperature, as a variable of the system instead of an input of the model. Only chemistry-climate models can simulate this interaction.
Thus, it is necessary to have continuous measurements of the UV solar spectral irradiance with a good accuracy, and particularly in the Herzberg continuum (expected SSI uncertainty at 1σ of ±0.1% per decade) for its influence on stratospheric ozone chemistry. Several missions (Compact Spectral Irradiance Monitor (CSIM), UVSQ-SAT) aim to test the efficacy of a CubeSat doing accurate SSI measurements of a much bigger and more expensive satellite. Indeed, the new UVSQ-SAT concept is designed to see if it is 
possible to achieve the SSI required accuracies. The first step of this strategy is to demonstrate the ability to build a sensor, which is a compact/robust/radiation-resistant solid-state photodetector that do not require cooling. This sensor needs to be able to have a functional lifetime highest than classical space-based UV sensors, which have a limited scientific operating lifetime in weeks rather than months or years due to contaminant trapping by their cooled surfaces. During the UVSQ-SAT lifetime mission, the sensor will have to measure the UV irradiance variability in the Herzberg continuum with an uncertainty at 1σ better than ±0.5% and to increase in the future the reliability in long-term data record .
To conclude, there is a need for a better understanding of how the Sun affects the climate, particularly for the UV radiation affecting ozone (the Herzberg continuum) since it links stratospheric ozone with regional effects. The Herzberg continuum corresponds to a spectral region (200 – 242 nm) where atmospheric absorption is relatively low and, hence, solar UV radiation penetrates deeply in the atmosphere, down to the lower stratosphere, where it photolysis molecular oxygen (O2) to produce ozone (O3). Absolute solar spectral irradiance and variability in the Herzberg continuum are necessary 
to better understand the stratospheric ozone response to solar UV irradiance changes. This is important because the Sun has long-term and short-term variations and we need to know how these interact with anthropogenic effects. It is also important to understand natural factors in climate variability to give a basis for a future where it might be predicted. The accurate measurements of the solar spectrum at the top of the atmosphere and its variability are fundamental inputs for Earth’s climate (climate’s modeling) and terrestrial atmospheric photochemistry. This is also important for long-term variations of solar cycle minima, which are of fundamental importance for Solar Physics modeling (dynamo, energy transfer, magnetic and 11–yr cycles, etc.). Thus, it is necessary to monitor continuously the Herzberg continuum region over years. One of the objectives of the UVSQ-SAT mission is to validate a new technology for future continuous UV observations using small satellites.

Scientific requirement

As explained in details in Section 2.1, measuring the absolute value of the Earth’s energy imbalance and its variability over time appear to be a very difficult challenge. The relevant scientific goal is to be able to detect any long-term trend with a target accuracy of 1/10 of the expected signal of 0.5 – 1.0 Wm−2 in the global mean during a decade. This issue could be solved through better absolute calibration of the sensors since several satellites will be needed to carry out these measurements with satellites temporal overlap to realize inter-calibrations. Table 1 presents the scientific objectives to be achieved by future space-based instrumentations onboard small satellites with on-board calibration systems for EEI observations. These EEI scientific objectives are extremely relevant and have not been achieved so far. At the present stage, the UVSQ-SAT CubeSat is a demonstrator, expecting future developments and improvement that would then really allow to make use of CubeSat technology for these scientific purposes. EEI expected performances of the UVSQ-SAT CubeSat are given in Table 1. The absolute value of UV SSI and its variability during more than one decade are also challenging. Accurate observations are fundamental to consolidate the reconstruction models of the solar spectral irradiance. Spectral And Total Irradiance REconstruction for the Satellite Era (SATIRE-S) highlights a weak long-term trend (Figure 1) of UV solar spectral irradiance over the past 40 years for solar minimas (inter-cycles), which can be real or not. The relevant scientific goal is to be able to detect any long-term trend with a target stability per decade of ±3.4 10−5 Wm−2nm−1 as example at 215 nm (Table 1). These solar observations with satellites temporal overlap to realize inter-calibrations are important since an analysis of radionuclides concludes that the Sun will enter a state of significantly lower activity within the next 50 to 100 years. These accurate observations are also important for long-term reconstructions over centuries where only proxies of solar activity are available as input for the reconstruction models. Indeed, the physical assumptions that go into the models lead to considerable discrepancies. At the present stage, the UVSQ-SAT CubeSat is a demonstrator that must show that these accurate SSI continuous measurements are possible with small satellites using new compact and robust disruptive technologies. UV solar spectral irradiance in the Herzberg continuum expected performances of the UVSQ-SAT CubeSat are given in Table 1.