Science Operation Concept of Bepicolombo/MMO Based on the MDP Scheme

Proceedings of the International Symposium on Planetary Science 2011, Edited by S. Okano, Y. Kasaba and H. Misawa, pp. 1–11. doi:10.5047/pisps.001 c TERRAPUB, Tokyo, 2019.

Science operation concept of BepiColombo/MMO based on the MDP scheme

Y. Kasaba1, T. Takashima2, M. N. Nishino3, and M. Fujimoto2

1Planetary Plasma and Atmospheric Research Center, Graduate School of Science, Tohoku University, Aramaki-aza-Aoba 6-3, Aoba, Sendai, Miyagi 980-8578, Japan E-mail: [email protected] 2Institute of Space and Astronautical Science, Japan Aerospace Exploration Agency, 3-1-1 Yoshinodai, Chuo-ku, Sagamihara, Kanagawa 252-5210, Japan 3Institute for Space-Earth Environmental Research, Nagoya University, Furo-cho, Chikusa-ku, Nagoya 464-8601, Japan

BepiColombo Mercury Magnetospheric Orbiter (MMO), which was launched in 2018, is dedicated to the first detailed study of magnetic field and the environment of the innermost planet Mercury. This orbiter has wide-range observational capabilities with 5 payload groups; MGF (Magnetic Field Investigation), MPPE (Mercury Plasma Particle Experiment), PWI (Plasma Wave Investigation), MSASI (Mercury Sodium Atmosphere Spectral Imager), and MDM (Mercury Dust Monitor). These payloads are operated through the Mission Data Processor (MDP), and act as an integrated measurement system observing charged and energetic neutral particles, magnetic and electric fields, plasma waves, radio waves, dust, and exospheric constituents. The MDP produces three kinds of data sets, Survey data (L-mode) for continuous monitoring, Nominal data (M-mode) for nominal science analyses with the length of several hours, and Burst data (H-mode) for short data but with the highest quality, with a coordinated manner within our limited telemetry resource. The MDP takes Burst data triggered by preset commands or onboard automatic decision. Although Nom- inal and Burst data cannot fully dumped by the limitation of telemetry bandwidth, partial key terms are dumped by the selection of a scientist team based on Survey or Nominal data. These functions are now implemented and tested. Key words: Mercury, magnetosphere, exosphere, BepiColombo, Mercury Magneto- spheric Orbiter (MMO), Mission Data Processor (MDP)

1 Introduction Although Mercury has been hard to investigate for any scientists in long years, we could have three chances for this innermost planet; Mariner 10, Messenger, and Bepi- Colombo (cf. Balogh et al., 2007). The first, three flyby measurements by Mariner 10 in 1974–1975, revealed the face of this planet and unexpectedly found an intrinsic magnetic field, a magnetosphere (cf. Connerney and Ness, 1988), and high-energy

1 2 Y. Kasaba et al. electron events (cf. Christon et al., 1987). We are just having the second, the Mes- senger spacecraft (cf. Solomon et al., 2007), whose orbit insertion was on 18 March 2011. Payloads aboard the spacecraft will provide a lot of new discoveries. BepiColombo, the first ESA (European Space Agency) and JAXA (Japanese Aerospace Exploration Agency) joint planetary mission, will be the third and the most progressive project (cf. Benkhoff et al., 2010). The mission consists of two spacecraft with different orbits and attitude control systems, in order to maximize the comprehensive studies, i.e., its structure, evolution, and surrounding environment. One spacecraft, the Mercury Planetary Orbiter (MPO) led by ESA, has eleven payload groups mainly focusing on the planetary surface and interior. Another spacecraft, the Mercury Magnetosphere Orbiter (MMO) led by JAXA, provides five payload groups mainly for the exosphere and magnetosphere. Combined observation with two orbiters also enables us to investigate fine magnetic field structure and planetary environments (cf. Milillo et al., 2010). Both spacecraft was launched together in Oct. 2018. After long cruise, both spacecraft will be into their polar orbits (the MPO: 400 × 1508 km; the MMO: 400 × 11, 824 km) in the end of 2025 in the plan. The MMO spacecraft, a spinning platform with the rate of 4–5 sec, is mostly dedicated to the first detailed study of magnetic field, waves, and particle environment of the planet Mercury (cf. Yamakawa et al., 2004; Hayakawa et al., 2004). Four main scientific targets are set from the BepiColombo mission objectives, significantly advance comparative studies of the magnetic fields and magnetospheres of terrestrial planets: (1) Structure and origin of Mercury’s magnetic field, (2) Structure, dynamics, and physical processes in Mercury’s magnetosphere, (3) Structure, variation, and origin of Mercury’s exosphere, and (4) The inner solar system (cf. Fujimoto et al., 2007). In order to achieve those, payload groups were selected in 2005 covering wide observational capabilities of charged and energetic neutral particles, magnetic and electric fields, plasma and radio waves, dust, and exospheric constituents (cf. Mukai et al., 2006). MGF (MaGnetic Field investigation) for magnetic field with 2 sub instruments (Baumjohann et al., 2010), MPPE (Mercury Plasma Particle Experiment) for plasma and neutral particles with 7 sub instruments (Saito et al., 2010), and PWI (Plasma Wave Investigation) for electric field, plasma waves, and radio waves with 7 sub instruments (Kasaba et al., 2010) were provided by large consortia from Japan, Europe and other countries. Those payload packages will perform in-situ and remote measurements of the magnetosphere, exosphere, and solar wind environments. MSASI (Mercury Sodium Atmosphere Spectral Imager), an imaging system is also included for the study of the sodium exosphere (Yoshikawa et al., 2010). MDM (Mercury Dust Monitor) covers the dust information around Mercury and the inner heliosphere (Nogami et al., 2010). The combination of all with the MPO payloads will observe the unique environments of Mercury, characterized by the weak intrinsic magnetic fields and the high dynamic pressure of the solar wind. In order to fulfill the science objectives under the strong restriction of mass, power, and telemetry resources, all payload groups are under the unified and coordinated controls of observational mode and time resolution, conducted by MDP Science operation concept of BepiColombo/MMO based on the MDP scheme 3

(Mission Data Processor) (Kasaba et al., in preparation). It enables us to do the scientiﬁc operation of the MMO as an integrated measurement system consisting of these ﬁve instruments groups.

2 MMO Operation Concept with the Mission Data Processor There are mainly two restrictions in the MMO science operation. One is the telemetry bandwidth to the ground. Although the raw data rate from all payloads is over 8.5 Mbps (mainly from MPPE—particle energy and counts, PWI—waveforms and spectrum, and MSASI—imaging), the telemetry bandwidth to the ground is few to 32 kbps depending on the distance between Earth and Mercury. It is also limited by the operation time of the ground station, JAXA Usuda Deep Space Center. The- fore, the expected telemetry rate is less than 4 kbps in average. Another is thermal harshness. In the phase close to the perihelion, the base temperature inside the spacecraft is close to 60◦C at the periherm, which requests frequent and synchronous mode change operations for all payloads. In order to shrink the amount of dumped data but maximize the science output under these restrictions, the observations of and the data production from all payloads are coordinated and integrated. For this spacecraft, we adopted the Mission Data Processor (MDP) Scheme, based on following items. This scheme defines not only the design requirement of the spacecraft orbiting Mercury but also the role of scientists on the ground. The load and capability of Scientists looking the data is a key of this concept: (1) All data commu- nications of the payloads are through the MDP. (2) Telecommand to each payload can be sent not only from the spacecraft system but also from the MDP, for the integrated control. (3) Housekeeping (HK) data from each payload can be monitored by the MDP, and be sent to the spacecraft system, for the integrated status monitoring. (4) Scientific data from each payload is accumulated by the MDP in the static format, as raw as possible, for the integrated telemetry production. The MDP keeps them with specific interval inside the own data storage. (5) In the MDP, scientific telemetry is produced from the data in the MDP storage to the system recorder, with a synchronization of related instruments. It produces three kinds of data sets; Survey mode with low rate (L mode), Nominal mode with medium rate (M mode), and Burst mode with the highest rate (H mode). (6) All of Survey mode data is downlinked. Not all but some parts of Nominal and Burst mode data can also be dumped after the selection by operation scientists checking Survey data.

3 Data Connections in the MMO with the MDP The data connections in the MMO are summarized in Fig. 1. In the MMO, the Data Management Controller (DMC) treats all telecommands from and HK/scientiﬁc data to the ground. (In the cruise phase when the MMO is a part of the Mercury Cruise Composite, the MMO send them through the MPO.) For the integration of operations and data productions within limited resources, the MDP is inserted between the DMC and all payloads (ref. Section 2 (1)). The MDP has two Digital Processing Units (DPU). DPU1 connects to MPPE (MEA1, MEA2, MIA, MSA, HEP-ele, HEP- ion, ENA) and MGF-outboard. DPU2 connects to MGF-inboard, MDM, MSASI, 4 Y. Kasaba et al.

Fig. 1. The connections and the ﬂow of telecommand, housekeeping data, and science data of the MMO.

PWI (EWO-E/B, SORBET, MEFISTO, MAST-WPT-E). In some failure cases, some functions of failed DPU can be covered by another DPU via a link between two DPUs. Integrated controls and data reductions are managed by the software provided by payload teams installed in the MDP. The performance is maximized within the resource, CPU power (Clock: 24 MHz), work memory size (2 MB), program size (1 MB) and power consumption (∼14 W). The MDP puts all telecommands to and gets HK data from each component in every 1 sec (ref. Section 2 (2)(3)). Associated with this function, the MDP does not only sent telecommands from and HK data to the DMC, but also decode and produce them, triggered by some telecommands to the MDP or the activation by speciﬁcHK parameters from payloads. It enables us to do coordinated and synchronized operations of multiple instruments, automatic operations (ex. wire antenna extension, etc.), and emergent reactions (ex. latch ups, etc.). Following part of this paper is concentrated to scientiﬁc data output via the MDP.

4 Data Flow in the MMO with the MDP The MDP gets the raw data from all payloads in almost static formats (ref. Section 2 (4)) with the rate summarized in Table 1. The raw data is tentatively stored in the Data Storage (DS) in each MDP-DPU. The DPU-DS has 256 MB and forms multiple ring buffers. The software for each instrument in the DPU performs data reduction from these stored data, such as sorting along magnetic ﬁeld, FFT from waveform, Science operation concept of BepiColombo/MMO based on the MDP scheme 5

Table 1. The amount of scientiﬁc telemetry: Raw data from payloads (kbps); Storage in DPU-DS (min); averaged rate of Survey (L) and Nominal (M) to the DMC-DR after data compression.

time averaging, data compression, etc. Since the data in DPU-DS is also used for the Burst mode data, which is described in later, the DPU-DS keeps longer data for in-situ instruments than for others, 20 min for MEA-1/2, MIA, MSA, HEP-ele/ion, MGF-out/in and EWO-EFD, and 4 min for EWO-E/B (Table 1). After the data reduction, the MDP create the processed data telemetry to the Data Recorder (DR) in the DMC. The size of DMC-DR is 2 GB, which is divided into multiple partitions acting as ring buffers. The data production rate summarized in Table 1 follows the policy described in Section 2 (5)(6) in order to keep the telemetry limitation, 4 kbps or less in average: (a) All of the Survey mode data is potentially downlinked, with ∼1/3 of the averaged telemetry bandwidth. (b) A part of the Nom- inal mode data, ∼1/5 or less, can only be dumped to the ground. (c) The Burst mode data can only be dumped in the highest link phase (32 kbps) when Mercury is close to the Earth. In order to realize this concept, for each instrument group, the DMC-DR prepares 1 partition with the length of 1 week for Survey mode (L), same for Nominal mode (M), and 6 partitions with 40–20 min for Burst mode (H) (Table 2). The majority of the MMO science should be performed by the Survey and Nominal mode data sets. 6 Y. Kasaba et al.

Table 2. The length of the DMC-DR partitions for payload data sets.

The Burst mode is special. Base concept is as follows: (a) [All] Majority of Survey mode data should be downlinked, as fast as possible. (b) [MPPE except ENA, MGF, PWI, and MSASI] A scientist team waits and checks the Survey mode data based on the specific science motivations. Based on the decision, part(s) of the data should be downlinked as Nominal mode. (This decision should be at least once per week, before the Nominal mode data in the DMC-DR is overwritten.) (c) [MPPE except ENA, MGF, and PWI-EWO] When the telemetry has capacity, one of Burst mode partition is dumped, by the selection based on Survey and Nominal mode data. For these in-situ instruments, we will fulfill the MMO scientific objectives (see Section 1) by Survey-mode only for (1), and by Survey/Nominal-mode data for (2)/(3)/(4). Burst- mode can expend them for (2)/(3)/(4) with the highest quality and time-resolution data. MDP also provides a lossless compression function based on the arithmetic coding algorithm. Two algorithms are also implemented by PI teams. One is the Rice algorithm (cf. Rice et al., 1993) by MPPE, lossless compression for the data that main values are close to zero. Another is a waveform compression algorithm by PWI, lossy compression with sub-band coding with high compression ratio (more than 60%), based on the one adopted to the LRS (Lunar Radar Sounder)-WFC instrument aboard the Kaguya spacecraft (Kasahara et al., 2008).

5 Data Dumped to the Ground Figure 2 shows an operation example of the Survey (L), Nominal (M), and Burst (H) modes. At the aphelion with the dayside periherm (the top of Fig. 2), the MMO stays in the nightside magnetosphere and along the magnetopause. Survey mode continuously takes the data. Nominal mode also takes continuously, but is partially dumped for several hours covering the key regions selected by scientiﬁc motivations. Science operation concept of BepiColombo/MMO based on the MDP scheme 7

Fig. 2. Operation image of the Burst (H), Nominal (M), and Survey (L) modes on the Mercury orbit: (Top) at Aphelion, with the apoherm at nightside. (Bottom) at Perihelion, with the apoherm at dayside. Time after the apoherm is also shown (orbital period: 9.3 hours) with an expected Magnetosphere.

At the perihelion with the nightside periherm (the bottom of Fig. 2), the MMO stays long in the solar wind. Survey mode also takes all. Selection of the Nominal mode covers shocks, magnetopause, cusp, and/or the nightside region for several hours. For all cases, Burst mode sometimes takes the data in 4–20 min during the part when the Nominal mode dump can be planned. Since the Hermean environment is small and fast, we expected the time length of Nominal and Burst modes are limited but can catch key science issues. 5.1 Survey (L) mode data Survey (L) mode data is for the continuous monitoring of the target, in order to get averaged and/or macroscopic characteristics of Hermean environment, which request the continuity as the most important issue. It is also used for the decision of M/H-mode data dump plan. MPPE-ENA (remote sensing of energetic neutral atoms) and MDM (dust counter) have relatively lower data rate and only provide this data. Others provide monitor data with lower resolutions in time, energy, spectrum, etc. 8 Y. Kasaba et al.

(ex. MPPE: density, velocity, temperature, and energy spectrum in 1–4 spins; MGF: 4 Hz magnetic waveform; PWI: 4 Hz electric waveform, electron density, and low resolution spectrum in 4–8 spins; MSASI: low resolution images) 5.2 Nominal (M) mode data Nominal (M) mode data is expected to have qualities enough for standard in- tensive analyses. Payloads providing this mode create higher resolutions in time, energy, spectrum, etc. It is also used for the decision of H-mode data dump plan. (ex. MPPE: moment, energy spectrum, and 3D distributions in 0.5–4 spins; MGF: 8 Hz magnetic waveform; PWI: 8 Hz electric waveform, electron density, and higher resolution spectrum in 0.5–4 spins; MSASI: higher resolution images) This data will be the most key part for the achievement of our mission objectives. Since the telemetry bandwidth to the ground is not enough to send all of them, the quasi real-time selection by a scientist team looking the Survey mode data becomes the most critical part of the mission. Although only ∼1/5 of the stored data can be dumped in average, we can flexibly get continuous data set on our demand (ex. Fig. 2), i.e., 1/5 (2 hours) of one orbit, a specific orbit (9.3 hours) per 5 orbits, etc. Even under these restrictions, we convince to be able to investigate key issues which requires more resolution and quality than Survey mode data, like the special kinematics and non-MHD features expected around Mercury, the structures and processes of various regions (shocks, magnetopause, auroral regions, cusps, plasma sheets, etc.), the time variations in short–medium scales, etc. 5.3 Burst (H) mode data As the description in Section 4, DPU-DS tentatively stores raw data for the onboard data reductions. These data sets can also be copied to the Burst mode partitions prepared in the DMC-DR. Some of them will be partially dumped to the ground based on the quasi real-time selection by a Scientist team looking Survey/Nominal data. By this scheme, we can potentially get the highest quality data, short (less than 4–20 min) but continuous enough for covering major boundaries and events expected in the Hermean environment. Since the MMO spacecraft should be ‘an integrated environment surveyor’, a data set covering all observational parameters essential for integrated analysis are simultaneously latched into the DMC-DR. Although the continuous Burst mode data cannot be stored in the DMC-DR, we can keep 6 sets in the DMC-DR in maximum, and will be dumped a part from one of them. Usually Burst data sets will be latched at the specific time planned on the ground defined by scientific objectives depends on operation phases. At the timing, a preset Burst-trigger command is sent from the DMC to the MDP. The MDP keeps the data stored inside the DPU-DS with the length of 4–20 min in maximum, and sends them to one of the Bust mode partitions in the DMC-DR. (ex. If we try to get the data inside the plasma sheet, we will set trigger commands around the time when the MMO will pass the magnetic equator region.) After the operation, DMC-DR will store 5–6 data sets. After the check by Survey or Nominal mode data covering the timings, we define which part of the data should have a value to consume our telemetry resource. We expect that the maximum data length restricted by the size of DPU-DS, 20 min, is long enough to hit aimed targets by this scheme. (One exception Science operation concept of BepiColombo/MMO based on the MDP scheme 9

Table 3. Parameters used in the onboard automatic trigger system MATRIX.

is EWO-E and B data for electric and magnetic waveforms, which can only cover last 4 min of this interval. Therefore, the MDP design allows to do another scheme, ‘4 min × 5 sets × 5–6 partitions’.) The automatic onboard data triggering system ‘MATRIX’ (MDP Automatic TRIg- ger compleX) is also implemented, partially referred to the THEMIS design (Taylor et al., 2008). Table 3 summarizes the parameters for the triggering i.e., the magnetic ﬁeld (MGF), particles (MPPE), electric ﬁeld, plasma waves, and radio waves (PWI), which are created in the MDP from the raw data in each spin (4–5 sec). In the trigger decision logic, four parameters from this list can be used in maximum. If the data ex- ceeds the threshold level preset by the parameter table uploaded by the telecommand, the MATRIX system activates automatic triggering, as a Burst trigger command similar to the one from the DMC. In this scheme, we can get the data not only after the trigger but also before it, 4–20 min in maximum. We should note that the parameter tunings of this system is hard for any important event triggers. Although the automatic triggering case can also store 6 events in maximum, try and error actions should be needed during the 1st Mercury year observation. We will decide whether it can be used or not after this result.

6 Summary MESSEMGER conveyed new views of Mercury, which should have impacts to the observation plan of BepiColombo. The scheme with the MDP described in this 10 Y. Kasaba et al. paper provides the ﬂexibility for such demands to the MMO. We need to wait addi- tional 10 years from now, but convince our success in future.

Acknowledgments. The authors would like to thank the colleagues of Mitsubishi Heavy In- dustries who developed hardware, OS, middleware of the MDP. The authors also wish to thank S. Yokota, K. Asamura, A. Matsuoka, Y. Kasahara, T. Imachi, K. Ishisaka, A. Kumamoto, and all members contributing to the PI parts of the MDP software. The authors are also grateful to all BepiColombo project members in the world.

References Balogh, A., R. Grard, S. Solomon, R. Schulz, Y. Langevin, Y. Kasaba, and M. Fujimoto, Missions to Mercury, Space Sci. Rev., 132,2–4, 611–645, 2007. Baumjohann, W., A. Matsuoka, W. Magnes, K.-H. Glassmeier, R. Nakamura, H. Biernat, M. Delva, K. Schwingenschuh, T. Zhang, H.-U. Auster, K.-H. Fornacon, I. Richter, A. Balogh, P. Cargill, C. Carr, M. Dougherty, T. S. Horbury, E. A. Lucek, and 17 authors, Magnetic field investigation of Mercury’s magnetosphere and the inner heliosphere by MMO/MGF, Planet. Space Sci., 56, 279–286, 2010. Benkhoff, J., J. van Casteren, H. Hayakawa, M. Fujimoto, H. Laakso, M. Novara, P. Ferri, H. R. Middleton, and R. Ziethe, BepiColombo—Comprehensive exploration of Mercury: Mission overview and science goals, Planet. Space Sci., 58,2–20, 2010. Christon, S. P., J. Feynman, and J. A. Slavin, Dynamic substorm injections—Similar magnetospheric phenomena at earth and Mercury, in Magnetotail Physics, p. 393–400, Johns Hopkins University Press, 1987. Connerney, J. E. P. and N. F. Ness, Mercury’s magnetic field and interior, in Mercury, edited by F. Vilas, C. R. Chapman, and M. S. Matthews, pp. 494, The University of Arizona Press, Tucson, 1988. Fujimoto, M., W. Baumjohann, K. Kabin, R. Nakamura, J. A. Slavin, N. Terada, and L. Zelenyi, Hermean magnetosphere-solar wind interaction, Space Sci. Rev., 132(2–4), 529–550, 2007. Hayakawa, H., Y. Kasaba, H. Yamakawa, H. Ogawa, and T. Mukai, The BepiColombo/MMO model payload and operation plan, Adv. Space Res., 33(12), 2142–2146, 2004. Kasaba, Y., J.-L. Bougeret, L. G. Blomberg, H. Kojima, S. Yagitani, M. Moncuquet, J.-G. Trotignon, G. Chanteur, A. Kumamoto, Y. Kasahara, J. Lichtenberger, Y. Omura, K. Ishisaka, and H. Matsumoto, The Plasma Wave Investigation (PWI) onboard the BepiColombo / MMO: First measurements of electric fields, electromagnetic waves, and radio waves around Mercury, Planet. Space Sci., 56, 238–278, 2010. Kasaba, Y. et al., The Mission Data Processor (MDP) onboard the BepiColombo / MMO, to be submitted. Kasahara, Y., Y. Goto, K. Hashimoto, T. Imachi, A. Kumamoto, T. Ono, and H. Matsumoto, Plasma wave observation using WaveForm Capture in the Lunar Radar Sounder on board the SELENE Spacecraft, Earth Planets Space, 60, 341–351, 2008. Milillo, A. et al., The BepiColombo mission: An outstanding tool for investigating the Hermean environment, Planet. Space Sci., 58,40–60, 2010. Mukai, T., H. Yamakawa, H. Hayakawa, Y. Kasaba, and H. Ogawa, Present status of the BepiColombo / Mercury Magnetospheric Orbiter, Adv. Space Res., 38(4), 578–582, 2006. Nogami, K., M. Fujii, H. Ohashi, T. Miyachi, S. Sasaki, S. Hasegawa, H. Yano, H. Shibata, T. Iwai, S. Minami, S. Takechi, E. Grun, and R. Srama, Development of the Mercury Dust Monitor (MDM) onboard BepiColombo mission, Planet. Space Sci., 56, 108–115, 2010. Rice, R. F., P.-S. Yeh, and W. H. Miller, Algorithms for high speed universal noiseless coding, Proc. AIAA Comp. Aerospace 9 Conf., San Diego, U.S.A., Oct. 19–21, 1993. Saito, Y., J. A. Sauvaud, M. Hirahara, S. Barabash, D. Delcourt, T. Takashima, K. Asamura, and the BepiColombo MMO MPPE team, Scientific objectives and instrumentation of Mercury Plasma Particle Experiment (MPPE) onboard MMO, Planet. Space Sci., 56, 182–200, 2010. Solomon, S. C., R. L. McNutt, Jr., R. E. Gold, and D. L. Domingue, MESSENGER mission overview, Space Sci. Rev., 131,3–39, 2007. Science operation concept of BepiColombo/MMO based on the MDP scheme 11

Taylor, E., P. Harvey, M. Ludlam, P. Berg, R. Abiad, and D. Gordon, Instrument data processing unit for THEMIS, Space Sci. Rev., 141,1–4, 153–169, 2008. Yamakawa, H., H. Ogawa, Y. Kasaba, H. Hayakawa, T. Mukai, and M. Adachi, Current status of the BepiColombo/MMO spacecraft design, Adv. Space Res., 33(12), 2133–2141, 2004. Yoshikawa, I., O. Korablev, S. Kameda, D. Rees, H. Nozawa, S. Okano, V. Gnedykh, V. Kottsov, K. Yosh- ioka, G. Murakami, F. Ezawa, and G. Cremonese, The Mercury sodium atmospheric spectral imager for the MMO spacecraft of BepiColombo, Planet. Space Sci., 56, 224–237, 2010.