advances in land surface hydrology representation in inm
play

ADVANCES IN LAND SURFACE HYDROLOGY REPRESENTATION IN INM RAS EARTH - PowerPoint PPT Presentation

ADVANCES IN LAND SURFACE HYDROLOGY REPRESENTATION IN INM RAS EARTH SYSTEM MODEL Victor Stepanenko 1 1 Lomonosov Moscow State University, Research Computing Center Contributors: V.Yu.Bogomolov, V.N.Lykossov, E.M.Volodin, I.Mammarella,


  1. ADVANCES IN LAND SURFACE HYDROLOGY REPRESENTATION IN INM RAS EARTH SYSTEM MODEL Victor Stepanenko 1 1 Lomonosov Moscow State University, Research Computing Center Contributors: V.Yu.Bogomolov, V.N.Lykossov, E.M.Volodin, I.Mammarella, H.Miettinen, A.Ojala, T.Vesala, S.P.Guseva, A.Medvedev International Young Scientists School and Conference, Zvenigorod, 5 September 2017 V.M.Stepanenko (MSU) Advances in land surface hydrology 5 September 2017 1 / 54

  2. Пример: снегопады над Великими Американскими озерами (lake-effect snow) При холодных вторжениях континентального воздуха интенсивное испарение и конвекция приводят к образованию облачности и осадков. "Озерные снегопады" парализуют дорожную ситуацию, закрываются школы, отменяются полеты и т.д. В течение XX в. наблюдается тренд увеличение суммы снежных осадков в данном районе, +1.9 см год − 1 . V.M.Stepanenko (MSU) Advances in land surface hydrology 5 September 2017 2 / 54

  3. Example: convection over Great African Lakes Nocturnal convection over Victoria accounts for annual fishers death toll ∼ 5000 . Thiery et al. 2015, J. of Climate, DOI: 10.1175/JCLI-D-14-00565.1 V.M.Stepanenko (MSU) Advances in land surface hydrology 5 September 2017 3 / 54

  4. Example: cloudiness over the Ladoga Lake Ice-free lake evaporates, and resulting stratiform clouds are advected to Finland. Cloudiness increases the surface net radiation, and 2m-temperature rises by 15-20 ◦ C Eerola et al. Tellus A 2014, 66, 23929, http://dx.doi.org/10.3402/tellusa.v66.23929 V.M.Stepanenko (MSU) Advances in land surface hydrology 5 September 2017 4 / 54

  5. Freshwaters in global carbon cycle (Tranvik et al. 2009) (Bastviken et al. 2011) Total freshwater methane emission is 104 Tg yr − 1 , i.e. 50% of global wetland emission (177-284 Tg yr − 1 , IPCC, 2013) greenhouse warming potentials from freshwater-originating CO 2 and CH 4 are roughly equal V.M.Stepanenko (MSU) Advances in land surface hydrology 5 September 2017 5 / 54

  6. CO 2 emissions by lakes and rivers Raymond et al., 2013, Nature Водоемы Водотоки global emission of CO 2 by freshwaters is 2.1 Pg C yr − 1 lake emission is 0.3 Pg C yr − 1 , river emissions is 1.8 Pg C yr − 1 significant contribution of Volga hydropower reservoirs V.M.Stepanenko (MSU) Advances in land surface hydrology 5 September 2017 6 / 54

  7. Эмиссия парниковых газов из водохранилищ Затопленные экосистемы подвергаются длительному разложению в преимущественно анаэробных условиях В отличие от естественных водоемов, имеется дополнительный путь для эмиссии метана в атмосферу – через турбины V.M.Stepanenko (MSU) Advances in land surface hydrology 5 September 2017 7 / 54

  8. Global warming of lakes The majority of lakes are warming at a rate higher than T 2 m . O’Reilly et al., 2015, GRL, doi:10.1002/ 2015GL066235 V.M.Stepanenko (MSU) Advances in land surface hydrology 5 September 2017 8 / 54

  9. 1D lake model framework 1D equations result from boundary-layer approximation 1D heat and momentum equations k − ǫ turbulence closure Monin-Obukhov similarity for surface fluxes Beer-Lambert law for shortwave radiation attenuation Momentum flux partitioning between wave development and currents (Stepanenko et al., 2014) Soil heat and moisture transfer including phase transitions Multilayer snow and ice models 1D concept does not suffice the greenhouse gas modeling task, as it does not take into account differences between CH 4 & CO 2 emissions at deep and shallow sediments V.M.Stepanenko (MSU) Advances in land surface hydrology 5 September 2017 9 / 54

  10. k − ǫ turbulence closure V.M.Stepanenko (MSU) Advances in land surface hydrology 5 September 2017 10 / 54

  11. 1 D + framework Traditional 1D model concept 1 D + model concept 1 D + model includes friction, heat and mass exchange at the lateral boundaries Heat, moisture and gas transfer are solved for each soil column independently In 1 D + model horizontally averaged quantity f obeys the equation: ∂f 1 ∂ ∂f 1 dA ∂t = ∂z A k f ∂z + F ( z, t, f, A ) + H f dz . A A V.M.Stepanenko (MSU) Advances in land surface hydrology 5 September 2017 11 / 54

  12. Coupling heat transport in water and soil z s 0 z s 0 T s 1 T s 1 F s 1 F s 1 Lake body z s 1 z s 1 T s 2 T s 2 Soil column 1 F s 2 F s 2 Soil column 1 z s 2 T s 3 z s 2 T s 3 F s 3 F s 3 z s 3 T s 4 z s 3 T s 4 Soil column 2 Soil column 2 z s 4 z s 4 F s 4 F s 4 Soil column 3 Soil column 3 Soil column 4 Soil column 4 Boundary conditions: at soil-water interface Soil column 5 Continuity of temperature (gas) Continuity of flux V.M.Stepanenko (MSU) Advances in land surface hydrology 5 September 2017 12 / 54

  13. 1D equations for enclosed basins Horizontally-averaged 3D equations for basic prognostic quantities: � � ∂t = · · · 1 c w ρ w ∂T ∂ A ( λ m + c w ρ w ν T ) ∂T − A ∂z ∂z − 1 ∂AS + 1 dA dz [ S b + F T,b ( z )] , – heat conservation equation (1) A ∂z A � � 1 + 1 ∂u ∂p ∂ A ( ν + ν m ) ∂u � � ∂t = · · · − + ρ w ∂x A ∂z ∂z + 1 dA dz F u,b ( z ) + fv, – momentum equation for x-speed component (2) A � � 1 + 1 ∂v ∂p ∂ A ( ν + ν m ) ∂v � � ∂t = · · · − + ρ w ∂y A ∂z ∂z + 1 dA dz F v,b ( z ) − fu – momentum equation for y-speed component (3) A V.M.Stepanenko (MSU) Advances in land surface hydrology 5 September 2017 13 / 54

  14. Barotropic pressure gradient and seiches y z Barotropic Lake surface Vertical cross-section, y = 0 (surface) seiches Wind are lake surface h x 2 and related L x, 0 v h x 1 u L y, 0 velocity oscillations u after strong wind events. x x Turbulent kinetic energy profile � dh N dt A 0 ( t ) = 2 � 1 dt A 0 ( t ) = − dh S (modeled), June 2013, Kuivajarvi 0 vL W − E hdξ, Lake, seiches produce TKE near Mass conservation dt A 0 ( t ) = 2 � 1 dh E dt A 0 ( t ) = − dh W bottom 0 uL S − N hdξ, ∂x ≈ gπ 2 � g ∂h s h E − h W L W − E, 0 , 4 Barotropic pressure gradient force ∂y ≈ gπ 2 g ∂h s h N − h S L S − N, 0 . 4 Surface oscillations in the model V.M.Stepanenko (MSU) Advances in land surface hydrology 5 September 2017 14 / 54

  15. Biogeochemical processes in the model Photosynthesis, Sinks and sources of gases in a lake respiration and BOD are Biochemical empirical functions of oxygen temperature and Chl-a O 2 demand (Stefan and Fang, 1994) (BOD) Oxygen uptake by sediments (SOD) is Photosynthesis Sedimentary controlled by O 2 oxygen concentration and Methane Respiration demand temperature (Walker and production (SOD) Snodrgass, 1986) Methane Methane production ∝ P 0 q T − T 0 oxidation , P 0 is 10 CO 2 CH 4 calibrated (Stepanenko et al., 2011) Turbulent diffusion Methane oxidation follows Michaelis-Menthen Bubble transport equation V.M.Stepanenko (MSU) Advances in land surface hydrology 5 September 2017 15 / 54

  16. Model validation for Seida Lake Guseva et al., Geogr. Env. Sust., 2016 Seida lake location Bubble flux (starting from 01.07.2007) V.M.Stepanenko (MSU) Advances in land surface hydrology 5 September 2017 16 / 54

  17. Kuivaj¨ arvi Lake (Finland) Mesotrophic, dimictic lake Area 0.62 km 2 (length 2.6 km, modal fetch 410 m) Altitude 142 m a.s.l. Maximal depth 13.2 m, average depth 6.4 m, depth at the point of measurements 12.5 m Catchment area 9.4 km 2 Secchi depth 1.2 – 1.5 m Point of measurements V.M.Stepanenko (MSU) Advances in land surface hydrology 5 September 2017 17 / 54

  18. Measurements Measurement raft Conducted since 2009 by University of Helsinki Ultrasonic anemometer USA-1, Metek GmbH Enclosed-path infrared gas analyzers, LI-7200, LI-COR Inc. Footprint of the Four-way net radiometer (CNR-1) raft measurements relative humidity at the height of 1.5 m (MP102H-530300, Rotronic AG) thermistor string of 16 Pt100 resistance thermometers (depths 0.2, 0.5, 1.0, 1.5, 2.0, 2.5, 3.0, 3.5, 4.0, 4.5, 5.0, 6.0, 7.0, 8.0, 10.0 and 12.0 m) Turbulent fluxes were calculated from 10 Hz raw data by EddyUH software V.M.Stepanenko (MSU) Advances in land surface hydrology 5 September 2017 18 / 54

  19. Water temperature Measurements Model Mixed layer depth and surface temperature (RMSE=1.54 ◦ C) are well reproduced Stratification strength in the thermocline is overestimated Model results lack frequent temperature oscillations in the thermocline V.M.Stepanenko (MSU) Advances in land surface hydrology 5 September 2017 19 / 54

  20. Oxygen Stepanenko et al., Geosci. Mod. Dev., 2016 Measurements Model Seasonal pattern is well captured: oxygen is produced in the mixed layer and consumed below Oxygen concentration in the mixed layer is underestimated by 1-1.5 mg/l , and more significantly during autumn overturn V.M.Stepanenko (MSU) Advances in land surface hydrology 5 September 2017 20 / 54

Recommend


More recommend