Report of St. Petersburg Team

O.G. Chertov, M.A. Nadporozhskaya E.V. Abakumov

Biological Research Institute St. Petersburg State University

2005

INTAS 01-0633 SILVICS

Introduction A close cooperation with the Pushchino

and Fraunhofer teams Development of a theoretical

background for the SOM model Incorporation of a new experimental

data into the models Formulation a new version of ROMUL

model Test the models for different spatial

scales

The laboratory experiments Impact of biochemical parameters of

plant debris on the rate of their decomposition

Impact of the disposition of decomposing matter (pure or in mixture with different soil material) to specify difference of above-ground and below-ground litter decomposition and patterns of decomposition in organic layers

Specification of nitrogen mineralisation in dependence on SOM and soil properties

The field works include:

Experiments on decomposition of forest litter fall of different quality in the forest

A study of SOM accumulationin a process of primary soil formation

Theoretical analysis of the decomposition process

0

20

40

60

80

100

1 t

%

Sugars and proteins

Celluloses

Lignin

Biochemical concept

0

20

40

60

80

100

0 t

Re

sid

ua

l m

ass

, %

Litter

Humified litter

Humus

Successional discrete concept

Continuum of litter quality loss and humification

40

60

80

100

1

Litter Humus t

Re

sid

ua

l m

ass

, %

Ågren Bosatta concept

Model of SOM and N dynamics ROMUL

The model is based on a classical concept of “humus type” (Humusform)

Experimental base for the model compilation is published and author’s data on organic debris decomposition in controlled conditions

The rate of litter and SOM humification and mineralisation is dependent on quality of litter, soil temperature and moisture, and some soil physical and chemical parameters

There is a specification of rate variables for above and below ground litter cohorts

The model calculates the dynamics of organic matter and nitrogen during the decomposition with gross CO2 and available N evaluation

The model was evaluated against the long-term experimental data

The model is in use as a soil compartment in three forest ecosystem models

Flow chart of ROMUL model

Li - Undecomposedlitter on soil surface

F.i - complex of humussubstances with

undecomposed debris(humified organic layer)

Lju -

undecomposedlitter in mineral

topsoil

F ju - complex of humus

substances withundecomposed debris in

mineral topsoil (“labil humus”)

H - humusbonded withclay minerals

Lju0 -

Belowgroundlitter fall

ki1L

ki2L

ki3L

K j4L K j

5S

K6

K j1S

K j3S

K j2S K j

4S

K j5S

Soil surface

Li0 -

Abovegroundlitter fall

Elaboration of a new ROMUL version

A large set of experimental data for SOM decomposition allows for a revision of ROMUL model

The kinetic coefficients of litter and SOM mineralisation were re-calculated using Bleasdale function and a special program (A.S. Komarov and M.A. Nadporozhskaya)

This allowed to specify the mineralisation rate in two sub horizons of forest floor (F and H) and a peat

A structure and test program of a new version of ROMUL model was compiled and preliminary tested

Calculation of kinetic coefficients of organic debris mineralisation and humification

Stage of fast decomposition reflects a mineralisation of fresh organic debris

Stage of slow decomposition represents a mineralisation of humified organic debris - not the material with increased concentration of lignin only

The function of Bleasdale was used for approximation of experimental curves:y = (a + bt) - 1/c or y = (a + bt)1/c

Flow chart of a new version of ROMUL model

Green – previous version, red – new components

L in A0 or poor decomposed peat

F in A0 or mean decomposed peat

New:Sub-horizon H in forest floor (A0)or well decomposed peat

Mineral topsoil

L_above

F_aboveCHS above

L_belowLabile

humus1

F_belowCHSbelowLabile

humus2

K1

K2

K3

H_below

0.3024*K2

K4*(1-f(CN))

K4*f(CN)

K1bK2b

K4b

L0b

K5a

K5bK3b

H_above

K6

K8

K8

L0a

The use of forest ecosystem model EFIMOD for research and practical implementation at forest stand, local and regional levels

Recently, the idea on the necessity to have a cascade of

forest ecosystem models with a different spatial resolution

was dominated in the terrestrial ecosystem modelling

Now there are technical opportunities allowing for a use

of one basic model type at any spatial levels without the

loss of information obtained at the lower levels

Some results of and prospects for the implementation of

one basic model type to cover different spatial scales in

forest ecosystem modelling were investigated

Methods and Material

Standard EFIMOD simulations of a single stand growth and soil changes were performed for the model use at different scales:

Individual tree growth Stand level: effects

of environmental changes; thinning regimes

Local (landscape) level: silvicultural regimes in forest enterprise (case studies)

Regional level: soil carbon dynamics for a large forest area

Individual tree growth

0

2

4

6

8

10

12

14

16

18

20

0 5 10 15 20 25 30 35 40 45

Time, years

Tre

e h

eig

ht,

m

Trajectories of individual tree growth on 25-m transect in a modelled Norway spruce stand

Map of individual trees’disposition on the modelled plot

Hierarchy of spatial scales for the application of a stand level

model Stand level: Parameters of individual trees’ growthStand/soil parameters in detailNo generalised parameters for forest area Local/landscape level:

Optionally parameters of individual tree growthStand/soil parameters in detailsGeneralised parameters of any format for forest area

Regional level:No parameters of individual tree growth

Optionally stand/soil parameters in full detailsGeneralised parameters of any format for forest area

The results of EFIMOD runs at different scales shows that

The application of one basic stand-level forest model for different spatial scales has positive prospects for its further development

At local and regional levels, this approach was used by Chumachenko et al. (2003: ForRus), Ho et al. (1999: LANDIS), Garman (2004), Kurz & Apps (1999: CBM-CFS2) and Nabuurs et al. (2003: EFISCEN)

The approach can be an additional methodological option that will be more effective for the practical implementation of the forest modelling for the realisation of the concept of Sustainable Forest Management

Case study I and II: Application of the EFIMOD-Pro for the

analysis of carbon balance at different silvicultural regimes in forests of Central

European Russia

Collaboration with Project’s Teams

Close co-operation with Pushchino and Fraunhofer teams

Participation in the Case Study Participation in the interpretation

and presentation of the results of geovisualisation and Exploratory Spatial Data Analysis (ESDA) for Case Study

Links to other projects

EU INTAS Project 01 512 Podzol St. Petersburg State University

Project ‘Changes of Soils and Soil Cover under Anthropogenic Factors’

Russian Federal Science and Technology Program “Global Climate Changes and Carbon Cycle”, part 14 “Soil as a source of greenhouse gases”

Publications for the period 2002-2005

International journalsPublished 4Submitted 3

Proceedings and national journalsPublished 7Submitted 3

Abstracts to conferences 19

21 presentations at international and national scientific meetings for the period May 2002 - February 2005

Pushkin, SPB (3) Gent(1)DSS Vienna (3), Uni Hohenheim (1) Trippstadt Forest Station (1) Pushchino (3),Kazan (2),Quebec (1)ForMod Vienna (3)ECEM 04 (2)

Acknowledgements

The participants of SPBU team

acknowledge

colleagues from other teams of the

Project,

the Administration of the Biological

Institute,

the Department of Soil Science and Soil

Ecology of St. Petersburg State

University and

the Dokuchaev Soil Museum for their

active

collaboration and valuable help