Biogeochemical fluxes in scenario simulations for the Baltic Sea in the period with Saint-Petersburg Baltic Eutrophication Model (SPBEM)

Size: px

Start display at page:

Download "Biogeochemical fluxes in scenario simulations for the Baltic Sea in the period with Saint-Petersburg Baltic Eutrophication Model (SPBEM)"

Mildred Quinn
5 years ago
Views:

Biogeochemical fluxes in scenario simulations for the Baltic Sea in the period 196-2 with Saint-Petersburg Baltic Eutrophication Model (SPBEM) Vladimir Ryabchenko1, Alexey Isaev2, Mikhail Molchanov2,

1 Biogeochemical fluxes in scenario simulations for the Baltic Sea in the period with Saint-Petersburg Baltic Eutrophication Model (SPBEM) Vladimir Ryabchenko1, Alexey Isaev2, Mikhail Molchanov2, Tatjyana Eremina2, Oleg Savchuck3, Roman Vankevich2 1St. Petersburg Branch of P.P. Shirshov Institute of Oceanology, RAS 2Russian State Hydrometeorological University, St.Petersburg 3Baltic Nest Institute, Baltic Sea Centre, Stockholm University Baltic Earth working group on scenario simulations 6 March 214, SMHI, Norrköping

2 SPBEM Hydrodynamic module Biogeochemistry module Cyanobacteria Summer spp. Diatoms Grazing Heterotrophs Autotrophs Mortality Si Nutrient uptake P Excretion N Detritus Mineralization Inorganic N, P & Si Bottom output Sinking Mineralization, P-retention, denitrification Burial z coordinate model is based on the based modelon of primitive equations of O.Savchuk, 22,and J.Mar.Systems, motion, heat salt 32,to23 transport, coupled a 28 dynamic thermodynamic model N, of Psea icesi cycling in the describes and coupled pelagic and sediment sub nm, 76 levels with dz=2msystems in the layer (,m) and dz=m hasbelow 12 pelagic m& 3 sediment state variables Neelov et al., 23. Proc. Estonian Acad. Sci. Biol. Ecol., 2(3): Myrberg K., Ryabchenko V., Isaev A. et al. 2. Boreal Env. Res. 1:

3 SCENARIOS OF CLIMATE & LOAD CHANGES Scenario of CO2 emission Global climate model Loads scenario name Land loads А1B ECHAM REF Observed mean Fixed as mean values averaged monthly values over А1B ECHAM BSAP same Linear Reduction to BSAP target А1B HadCM3 REF same Fixed as mean values averaged over A1B HadCM3 BSAP same Linear Reduction to BSAP target Fixed target values Fixed target values ECHAM/MPI OM from the Max Planck Institute Roeckner et al. 26; Jungclaus et al. 26 for Meteorology in Germany Gordon et al. 2 HadCM3 from the Hadley Centre in the UK

4 VERIFICATION Map of the location of 16 oceanographic monitoring stations from the Baltic Environment Database (BED) D Averaged Data from Gustafsson and Rodriguez-Medina, 211

5 Model-data comparison Mean observed (D) and (M) simulated values averaged over 16 stations (1971 2), their difference (Δ), and amount of stations in different ranges of cost function (C) Upper layer (-m) Scenario T ann T wint T sum С S ann NO3 wint PO4 wint mmol м-3 Near-bottom layer O2 Summ Chl summ ml/l mg m -3 T ann S ann С NO3 wint PO4 wint mmol м-3 O2 aut ml/l Dobs Δ ECHAM C< C< Δ HadCM C < C< СщСщы C M D S d C < 1 good agreement 1 C < 2 satisfactory bad. 2 C

6 RESULTS The difference between the future(271 2) and modern(1971 2) values of parameters Upper layer (-m) REF ECHAM HadCM3 T ann T wint T S summ ann Near-bottom layer NO PO O2 3 wint 4 wint sum Chl T summ ann S ann NO3 wint PO4 wint O2 aut ECHAM HadCM BSAP 1. Temperature increase in future climate especially in summer is higher in HadCM3 runs 2.. In REF runs, O2 decrease is greater with HadCM3 than with Echam forcing 3. Oxygen decrease in near bottom layer is less in BSAP runs than in REF runs. Unlike ECOSUPPORT simulations, there is an decrease rather than increase of near bottom oxygen.

7 RESULTS Oxygen concentration in the near bottom layer in the case of HadCM3 forcing with REF loads Modern(1971 2) Future(271 2)

8 RESULTS 8 Anoxic zones for REF runs 3 km Echam 3 Hadly Anoxic zones by the end of 21st century will be wider, compared to current conditions in reference ECHAM and HadCM3 runs.

9 RESULTS Comparison between REF and BSAP runs Anoxic zones(hadcm3) 3 km2 Anoxic zones(echam) 3 km BSAP 3 3 REF According to the ECHAM and HadCM3 driven BSAP scenario simulations, nutrient load reduction suggested in BSAP will not lead to any fundamental changes in the water quality in the end of this century. 2. Areas of anoxia and hypoxia will grow, but slower than in the reference runs.

10 Modern climate, Nitrogen fluxes (ktonsn/ year) 17 Ensemble means (ECOSUPPORT) ECHAM(SPBEM) HadCM3 (SPBEM) 12 7 Ne te xp or t Se di m en tb ur ia l D en itr ifi ca tio n at m.) + (la nd Lo ad s Ni tr og en fix at io n 2 Nitrogen fixation and denitrification in SPBEM runs are much higher than in ECOSUPPORT ensemble simulations Why?

11 22 2 Nitrogen fluxes (ktonsn/ year) 17 Ensemble means (ECOSUPPORT) ECHAM(SPBEM) HadCM3 (SPBEM) 12 Modern climate, Ne te xp or t Se di m en tb ur ia l De ni tri fic at io n at m.) + (la nd Lo ad s Ni tr og en fix at io n Ensemble means (ECOSUPPORT) ECHAM(SPBEM) HadCM3 (SPBEM) Future climate, Ne te xp or t Se di m en tb ur ia l De ni tri fic at io n at m.) + (la nd Lo ad s fix at io n 2 Ni tr og en Nitrogen fluxes (ktonsn/ year) 3

12 4 Cyanobacteria, mgn/m3 Diatoms, mgn/m3 Summer species, mgn/m3 Heterotroph community, mgn/m3 Baltic Sea 3 SPBEM ECHAM run 3 2 Seasonal succession of diatoms, summer species, cyanobacteria and zooplankton in the upper layer, averaged over Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 4 3 Cyanobacteria, mgn/m3 Diatoms, mgn/m3 Summer species, mgn/m3 Heterotroph community, mgn/m3 Baltic Proper Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

13 SPBEM ECHAM run Table 3. Calculated (M) and observed (D) values of nitrate, phosphate, ammonium, dissolved oxygen and chlorophyll a in the sea upper layer (-m) at 2 monitoring stations in the Baltic Proper (averages over the period 1971 to 2). Station and forcing Mean winter nitrate, mmol m-3 Mean winter phosphate, mmol m-3 Mean summer Mean summer oxygen, ml l-1 chlorophyll a conc., mg m-3 M D M D M D M D SE Gotland Basin,E Gotland Deep BY1, E Entire Baltic Sea

14 Baltic Proper 4 3 Cyanobacteria, mgn/m3 Diatoms, mgn/m3 Summer species, mgn/m3 Heterotroph community, mgn/m3 cyano summer sp. 3 N:P ratio Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct N-fixation start PP diatoms PP cyano N -fixation Temperature, C excretion Dec T 19 PP ss Nov Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Salinity, Jan 1 2 N:P Ratio 1 2 DIP, mg/m 3 zoopl. diatoms DIN, mg/m3 2

15 1. By this moment, N:P ratio reaches its minimum after N-limited spring bloom creating favorable conditions for N-fixation 2. Modeled summer T is higher than observed by 1. C and reaches prescribed NF-threshold (13 C) earlier (in the beginning of June) 3. The Redfield DIP excess is also two-three times higher than observed because of the overestimated winter values 4. Taken together, all these conditions lead to a massive nitrogen fixation and strong cyanobacteria bloom already in June. Zooplankton biomass increases accordingly, accompanied by increased nutrient excretion, thus introducing fixed nitrogen into biotic cycles 6. Intensification of nutrient regeneration favors the growth of summer species. As a result, their PP is much higher than PP of diatoms and cyanobacteria.

So, we have in SPBEM much more intensive recycling within water column

Initial (in 197) mean volume averaged pools of dissolved inorganic

and phosphorus in the sediments (in kton) for the entire Baltic

3 9 SPBEM 24 7 13 24 The main reasons of that are: 1) higher initial P

16 So, we have in SPBEM much more intensive recycling within water column under the same external loads as in ECOSUPPORT. Initial (in 197) mean volume averaged pools of dissolved inorganic nitrogen (DIN) and phosphorus (DIP) and area averaged pools of nitrogen and phosphorus in the sediments (in kton) for the entire Baltic ECOSUPPORT Ensemble mean Water DIN Water DIP Sediment N Sediment P SPBEM The main reasons of that are: 1) higher initial P and N sediment content in the BS (in the comparison with ECOSUPPORT simulations); 2) higher summer temperatures.

17 CONCLUSIONS 1. Changes in eutrophication indicators in reference HadCM3 driven run is greater than in similar ECHAM driven run. The main reason is higher water temperature increase, T, in HadCM3 run which will (1) reduce oxygen concentrations due to its lower solubility in warmer water and (2) accelerate organic matter mineralization and oxygen consumption. Expanding hypoxia will increase phosphorus release from the sediments resulting in reduced removal of phosphorus from the ecosystem, while simultaneous increase of denitrification leads to augmented removal of nitrogen from the ecosystem. Higher T together with increased nitrogen fixation will intensify primary production and increase phytoplankton biomass.

18 CONCLUSIONS 2. According to the ECHAM and HadCM3 driven BSAP scenario simulations, nutrient load reduction suggested in BSAP will not lead to any fundamental changes in eutrophication indicators in the end of this century. In particular, areas of anoxia and hypoxia will grow, but slower than in the reference runs. 3.The estimates are qualitatively consistent with the estimates of ECOSUPPORT project, but impact of climate change on eutrophication was much stronger. 4. As the analysis of biogeochemical fluxes showed, nitrogen fixation and denitrification in SPBEM runs are several times higher than in ECOSUPPORT ensemble simulations. The main reasons of that are: 1) higher initial P and N sediment content; 2) higher summer temperatures in modern climate.. These simulations should be viewed as sensitivity analysis of the model solution to the initial conditions, which, as was shown, are a significant source of uncertainty of the final result.

Development of indicators of eutrophication for the Eastern Gulf of Finland

Development of indicators of eutrophication for the Eastern Gulf of Finland Dr. Tatjana Eremina Russian State Hydrometeorological University BSAP targets for the Gulf of Finland Indicator of eutrophication