Russian Academy of Sciences A.M. Obukhov Institute of atmospheric - - PowerPoint PPT Presentation

SLIDE 1

Modelling the biotic turnover in ecosystems of permafrost regions of the Northern EuroAsia under climate change

e-mail: nickolos@ifaran.ru

Nikolay N. Zavalishin Russian Academy of Sciences A.M. Obukhov Institute of atmospheric physics Laboratory of mathematical ecology

SLIDE 2

• living form diversity of plants constituting the basis for terrestrial ecosystems

serves a source for structural complexity of biotic turnover;

• small amount of exchanging processes controlled by similar mechanisms in

different ecosystems initiates universality in the cycle functioning. (Basilevich and Titlyanova, 2008) Similarity and differences of a biotic turnover in terrestrial ecosystems

• разнообразие жизненных форм растений, составляющих основу наземных

экосистем, является источником сложности структуры круговорота;

• небольшое число обменных процессов, тождественных по своим

механизмам в различных экосистемах, порождают универсальность функционирования круговорота. (Базилевич и Титлянова, 2008) Сходства и различия биотического круговорота в наземных экосистемах

SLIDE 3

Synthesis - ? Model classes for biotic turnover Detailed simulation models Qualitative “minimal” models

SLIDE 4

• aggregation of complicated static schemes of biological

turnover in terrestrial ecosystems to the simplest variants;

• design and calibration of dynamic models for carbon

and/or combined carbon-nitrogen cycling (turnover) in ecosystems on the basis of aggregated static schemes;

• verification of correspondence between model steady

states and real stable ecosystem types;

• calculation of stability boundaries for steady states

and oscillations of single and/or combined cycles, study

• f their evolution under input flows and parameter

variations;

• numerical estimations of changes in carbon and/or

nitrogen functioning which can be considered as a reaction of ecosystems to external natural and anthropogenic perturbations.

Main stages in mathematical modelling of biological turnover in ecosystems on the annual time scale

Problems and uncertainties

• methods of aggregation and

their verification ?

• how to select flow-storage

dependencies ?

• how to interpret multiple

equilibria ?

• how to avoid undesirable

stability boundaries ?

• how to reduce uncertainties in

the solution under uncertainties in initial data ?

SLIDE 5

Tundra and forest tundra ecosystems in permafrost regions of Northern Euroasia

Tundra peatland Larch sparse forest

Photos from (Bazilevich and Titlyanova, 2008) Distribution patterns of boreal forests (shaded area) and southern boundaries of the zones of continuous permafrost (thick broken line) and those

• f discontinuous permafrost (thin

broken line) in northern hemisphere. Location of Tura is indicated by an asterisk. Dots and numerals: Moscow (1), Krasnoyarsk (2), Ulaanbaatar (3), Yakutsk (4), Fairbanks (5), Edmonton (6), Winnipeg (7), and Montreal (8). (Forest ecosystems of the Enisey meridian, 2002)

SLIDE 6

?

Biotic turnover in tundra and forest tundra ecosystems of permafrost region

Local ecosystem of moss-shrub tundra at the Taymyr peninsula (Tareya): biotic cycles of carbon, nitrogen and mineral elements (Bazilevich et al., 1986; Bazilevich and Gilmanov, 1985) Photo from (Bazilevich and Titlyanova, 2008)

Local low-parametric dynamic model of coupled carbon- nitrogen cycles with climatic parameterization and steady states corresponding to tundra ecosystem types

SLIDE 7

Biotic turnover in larch forests in permafrost region of Northern Euroasia

?

Carbon budget of larch forest at age 105 in northern taiga

• f Eastern Siberia (Permafrost ecosystems, 2010)

Carbon and nitrogen budgets of larch forests (Larix gmelinii, Larix cajanderi) in forest tundra Enisey region in Eastern Siberia (Forest ecosystems of the Enisey meridian, 2002)

Local low-parametric dynamic model

• f coupled carbon-nitrogen cycles with

climatic parameterization and steady states corresponding to larch forest or forest tundra ecosystem types

Photo from (Bazilevich and Titlyanova, 2008)

SLIDE 8

Carbon and nitrogen flows in an ecosystem: photosynthesis, respiration, denitrification and nitrogen fixation consumption, litterfall, excretion accumulation in real increment, import and export, abiotic

• xidation,

translocation

Universal scheme of a biotic turnover in terrestrial ecosystems

Reservoirs: G – green phytomass, Pr – perennial phytomass, R – living roots, D+D’- dead standing phytomass, V+L+{Slh} – dead roots + litter + {humus}, Ph+Z – phyto- and zoophagues, Mo+F+Sph – microorganisms+fungi+saprophages, Sln – soil reserve nutrients.

G Ph+Z Pr D+D’ R

V+L+{Slh} Mo+F+Sph

Sln

SLIDE 9

Aggregation of biotic turnover schemes in terrestrial ecosystems

G+Pr+R Ph+Z+F+Mo +Sph

D+D’+L+V+Sln

y1

C

f12

C

f13

C

q2

C

y2

C

f32

C

y31

C

q3

C

f23

C

q1

C

y1

N

q2

N

f12

N

f13

N

f31

N

f32

N

q3

N

f23

N

y32

C

y31

N

y2

N

G Ph+Z Pr D+D’ R

V+L+{Slh} Mo+F+Sph

Sln

G+Pr+R

D+D’+L+V+Sln

y11

C

f12

C

q2

C

y12

C

y21

C

q1

C

y12

N

q2

N

f12

N

f21

N

y32

C y33 C

y21

N

y32

N

SLIDE 10

Minimal aggregated compartment schemes of carbon and nitrogen cycles in tundra and forest tundra ecosystems

G+Pr+R C1, N1

Ph+Z+F+Mo+Sph

C2, N2 D+D’+L+V C3, N3

y1

C

f12

C

f13

C

y2

C

f32

C

y31

C

q3

C

f23

C

q1

C

y1

N

q2

N

f12

N

f13

N

f31

N

f32

N

q3

N

f23

N

y32

C

y31

N

y2

N

Storages of carbon and nitrogen: C1, N1 - phytomass; C2, N2 – phytophages and destructors (animals, fungi, bacteria); C3, N3 – dead organic matter of litter and root-based soil layer Flows : q1

C– carbon assimilation from the atmosphere, q1 N, q2 N– nitrogen input from adjacent

ecosystems with atmospheric nitrogen fixation by microorganisms, q3

C– dead organic

matter input from adjacent ecosystems, y1

C – autotrophic respiration, y2 C– heterotrophic

respiration, f12

C, f12 N – consumption of phytomass by phytophages, f32 C, f32 N – decay of

dead organic matter by destructors with denitrification, f23

C, f23 N – death of destructors

and phytophagues, y31

C, y31 N – export and run-off, y32 C – abiotic oxidation of dead

• rganic matter, f31

N – nitrogen uptake by vegetation from soil, f13 C, f13 N – litterfall.

NPP

SLIDE 11

1) any number of static schemes for various time moments can be a source for a dynamic model; 2) yi = mixi – output flows are linear; 3) qi = qi(xi), fki = fki (xk, xi); fik = fik(xi, xk) – flow functions are selected using biological information and expert knowledge, e.g.,

• donor type,
• recipient type,
• Lotka-Volterra type,

saturation types:

,

j ij j i ij

x L x x K + ,

i ij j i ij

x L x x K +

) )( (

i ij j ij j i ij

x N x L x x K + +

i ijx

α

j ijx

β

j i ij

x x γ

fki ( xk,xi )

x k ( t ) x i ( t ) x 1 ( t )

yk ( xk ) qk ( t ) qi ( t )

yi ( xi ) q1( t )

y1( x1 )

fik (xi , xk )

Two approaches to a dynamic model design by given «storage-flow» diagrams Mass-balance equations:

“Global” approach – started from (Svirezhev,1997) “Local” method

∑ − + − =

≠ = n i k k ik ki i i i

f f y q dt dx

, 1

) (

1) q* + f* = y* - at least one of the given diagrams is a dynamic equilibrium; 2) yi = mixi – output flows are linear; 3) fki = fki (xk, xi); fik = fik(xi, xk) – flow functions are represented asymptotically near the given equilibrium, :

... ~ ~ ~ ~ ~ ~ ) ~ (

2 2

+ + + + + + = +

∗ ∗ j i i i j i ij j ij i ij ij ij

x x x x x x f x x f η ξ γ β α

∗

− =

i i i

x x x ~

∑ ∑ ∑ ∑

≠ ≠ ≠ ≠

+ + + − + + − + − =

i j i j j j j j i ij ji i j ij ji j i j ij ji i i

x x x x x dt x d ... ) ( ~ ~ ~ ) ( ) ( ~ ~ ~

2

η ξ γ γ β α α β

SLIDE 12

LOM C1, N1 DOM C2, N2

Litterfall Run-off NPP Consumption Input Run-off Decay Consumption

Carbon and nitrogen interaction is provided by two mechanisms (Logofet, Alexandrov, 1984): 1) intensity of litterfall (f12

C) is proportional to the C1/N1 ratio in the living phytomass that

reflects nitrogen starvation of plants; 2) decay rate for dead organic matter decreases with the increase of C3/N3 ratio.

Modelling a combined carbon-nitrogen turnover in tundra ecosystems: biological mechanisms

LOM – living organic matter without consumers, DOM – dead organic matter

SLIDE 13

Modelling a combined carbon-nitrogen turnover in ecosystems: mathematical form

LOM C1, N1 DOM C2, N2

Litterfall Run-off NPP Consumption Input Run-off Decay Consumption

Mathematical form for coupled N-C потоков (Alexandrov et al., 1994): 1) Litterfall :

• carbon flow: , nitrogen flow:

2) Decomposition of dead organic matter:

• carbon flow: , - nitrogen flow:

3) nitrogen uptake from soil by plants:

1 2 1 12 12

N C f

C C

α =

1 12 12

C f

N N

α =

2 2 2 2 21

C N d y

N N =

2 2 21

N d y

C C =

1 2 2 2 21 21

C C N f

N N

γ =

SLIDE 14

Asian polar biomes

0,00 100,00 200,00 300,00 400,00 500,00 600,00 700,00 0,00 5000,00 10000,00 15000,00 20000,00 Phytomass, gC/m2 NPP, gC/m2/year

Forest tundra Tundra Northern taiga

Phytomass and Net Primary Productivity of vegetation types

Data from (Bazilevich and Titlyanova, 2008)

1 01 1 1 01

C L N C K NPP

C C

+ =

Nitrogen in Asian polar biomes

0,00 100,00 200,00 300,00 400,00 500,00 600,00 700,00 0,00 10,00 20,00 30,00 40,00 50,00 60,00 70,00 Nitrogen, gN/m2 NPP, gC/m2/year Tundra Forest tundra Nothern taiga

SLIDE 15

The Net Primary Productivity (NPP) and phytomass values for two ecosystems adjacent in state and space are used: ⇒ − + =

1 1 1 01 1 1 01

C m C L N C K NPP

C C C

⎪ ⎩ ⎪ ⎨ ⎧ = + + = + +

• C
• C
• m

m C m C m m m

N C K C L C m NPP N C K C L C m NPP

1 1 01 1 01 1 1 1 1 01 1 01 1 1

) )( ( ) )( (

, ) ( ) (

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 01

• m

m

• m

m m m m m

• m
• m

C

N C N C C m NPP C m NPP C m NPP N N C m NPP C L + − + ⎥ ⎦ ⎤ ⎢ ⎣ ⎡ − − + =

m m m C C C

N C C L q K

1 1 1 01 1 01

) ( + =

• for turnover state to calculate
• for adjacent state
• solution of algebraic system

Ecosystem L01

C, gC/m2

K01

C, gC/gN/

year Moss-shrub tundra (Taimyr) 167.5 8.3 Larch in forest tundra (East. Siberia) 197.7 17.6

Parameter calibration methods for dynamic models of biological turnover

Coefficients of flow functions depending on the single storage are calculated from the given scheme. One-parametric flows: Two-parametric flows:

SLIDE 16

Dynamic equations of 3-component model (moss-shrub tundra): (1)

( )

⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎩ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎨ ⎧ − − + + − = − + + − = + + − − = + − + − + = + − − − = ⎟ ⎟ ⎠ ⎞ ⎜ ⎜ ⎝ ⎛ − − + =

3 2 3 2 32 1 3 2 3 31 2 23 1 13 3 3 3 3 3 2 32 2 23 1 2 1 13 3 3 3 3 2 3 2 3 32 1 2 12 2 23 2 2 2 2 2 3 32 23 1 12 2 2 2 2 2 3 2 3 1 31 1 13 1 2 12 1 1 1 1 1 13 2 12 1 01 1 01 1 1

/ / / / / / C N C C C N N C N m q dt dN N C C N C C m q dt dC C C N N C N N m C K dt dN N C m C C K dt dC C N C C N C N m dt dN N C C C L N K C dt dC

N N N N N N C C C C C N N N N N C C C C C N N N N C C C C

γ γ α α γ α α γ γ α γ α γ γ α γ α γ

Dynamic equations of the combined carbon-nitrogen cycle in the ecosystem of moss-shrub tundra

Storages - gN/m2, gC/m2, flows - gN/m2·year, gC/m2·year.

C1 = 664.0, N1=12 C2=24.5, N2=4.0 C3=5348, N3=466.2

y1

C=213.4

f12

C=13.1

f13

C=137.2

y2

C=118.8

f32

C= 199.4

y3

C=14.3

q3

C=2.2

f23

C=93.7

q1

C=363.7

5 f12

N=0.18

f13

N= 3.82

f31

N= 4.0

f23

N=3.34

q3

N=0.2

y3

N=0.6

q2

N=0.2

f32

N=3.9

y2

N=3.8

SLIDE 17

0 – arctic desert; 1 – moss-shrub tundra; 2 – sphagnum-shrub wetland; 3 – birch sparse forest; 4 – mixed forest with spruce; 5 – shrub tundra; 6 – larch forest. Interpretation of equilibria by (Abaimov, 2005), (Karelin, Zamolodchikov, 2008).

• oscillatory domain

Stability boundaries for steady states in models of a biotic turnover in tundra and forest tundra ecosystems in permafrost region

Stability boundaries of stationary dynamic regime of the biological turnover:

NPP intensity, gС/m2/ year Decay intensity, 1/gN/year moss-shrub tundra larch forest

SLIDE 18

Климатические сценарии и климатозависимые параметры

Climate change scenarios from the global climate model IPSL (CMIP5): RCP-2.6 (softly warm) – +0.9 … 2.3 ºC up to 2100 globally, +1,0 … 3.0 ºC locally RCP-8.5 (extremely warm) - +3.2 … 5.4 ºC up to 2100 globally, +3.8 … 6.0 ºC locally Temperature dependent model parameters: NPP = NPP(Ca)- NPP of vegetation increases under atmospheric CO2 content Ca; m2

С =m2 С(T,H) – heterotrophic respiration intensity;

γ32

С =γ32 С(T,H) – intensity of decay for dead organic matter depend on the annual air temperature and

total precipitation in a polynomial form.

SLIDE 19

Fires with climate change scenarios:

• RCP-8.5
• RCP-2.6

0 – arctic desert; 1 – moss-shrub tundra; 2 – sphagnum-shrub wetland; 3 – birch sparse forest; 4 – mixed forest with spruce; 5 – shrub tundra; 6 – larch forest. Interpretation of equilibria by (Abaimov, 2005), (Karelin, Zamolodchikov, 2008).

• oscillatory domain

Climate change scenarios :

• RCP-8.5
• RCP-2.6

NPP intensity, gС/m2/ year Decay intensity, 1/gN/year

Response of a biotic turnover in tundra and forest tundra ecosystems in permafrost region to climate change

Stability boundaries of stationary dynamic regime of the biological turnover:

SLIDE 20

Conclusions

3). 1) «Global» approach to dynamic compartment model design by static “storage- flow” schemes helps in constructing coupled models of carbon-nitrogen cycle functioning in tundra and forest tundra ecosystems of the permafrost region; 2) Softly warm climate change scenario RCP-2.6 leads the moss-shrub tundra to the oscillatory dynamic regime while RCP-8.5 results in transformation into shrub tundra. The leading climatic factor is total annual precipitation; 3) RCP-2.6 remains the northern Larch forest and raise all components of its biological turnover while harder RCP-8.5 scenario tends to transform it into a birch forest; 4) Joint impact of climate change and fires can transform state of both ecosystems in a catastrophic way: under RCP-8.5 forest and tundra can be removed into wetland state while RCP-2.6 can transform larch forest into shrub tundra.

SLIDE 21

Thank you for attention !

Supported by the RFBR projects 12-05-01092a, 14-05-00193a and the Program 12 of the RAS Earth Sciences Department