Microbial dormancy improves development and
experimental validation of ecosystem model
Gangsheng Wang1,2, Sindhu Jagadamma1,2, Melanie A. Mayes1,2, Christopher W. Schadt2,3,
J. Megan Steinweg2,3,4, Lianhong Gu1,2, Wilfred M. Post1,2
1
Environmental Sciences Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831, USA
2
Climate Change Science Institute, Oak Ridge National Laboratory, Oak Ridge, TN 37831, USA
3
Biosciences Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831, USA
4
Currently at Biological Sciences Department, University of Wisconsin-Baraboo/Sauk County, Baraboo, WI 53913,
USA
Supplementary Information
Correspondence: Gangsheng Wang
Bldg 4500N, F129-S, MS-6301
Oak Ridge National Laboratory
Oak Ridge, TN 37831-6301
Tel: (865)576-6685; Fax: (865)574-9501
Email: wangg@ornl.gov
1
MEND model
Fig. 1 in the main text
Diagram of the Microbial-ENzyme-mediated Decomposition (MEND) model. Soil organic
carbon pools include: particulate organic carbon (POC) (e.g., lignocellulose-like compounds and
starch-like compounds, denoted by P1 and P2, respectively), mineral-associated organic carbon
(MOC, denoted by M), dissolved organic carbon (DOC, D), adsorbed phase of DOC (QOC, Q),
active microbial biomass (BA), dormant microbial biomass (BD), POC degraded enzymes (e.g.,
EP1 and EP2 that break down P1 and P2, respectively), and MOC-degraded enzymes (EM). IP1,
IP2, and ID are external inputs to the pools of P1, P2, and D, respectively. Transformation fluxes
include: (1) POC1 (P1) decomposition (denoted by the flux F1 in equations in the Supplementary
Information), (2) POC2 (P2) decomposition (F2), (3) MOC (M) decomposition (F3), (4, 5)
adsorption (F4) and desorption (F5) between DOC (D) and QOC (Q), (6) DOC (D) uptake by BA
(F6), (7,8) dormancy (F7) and reactivation (F8) between BA and BD, (9, 10) BA growth
respiration (F9) and maintenance respiration (F10), (11) BD maintenance respiration (F11), (12)
BA mortality (F12), (13) synthesis of EP1 (F13,EP1), EP2 (F13,EP2), and EM (F13,EM), and (14)
turnover of enzymes (F14,EP1, F14,EP2, and F14,EM).
2
The dynamics of each soil carbon pool (see Fig. 1) in the Microbial-ENzyme-mediated
Decomposition (MEND) model can be described as
dP1
 I P1  (1  g D )  F12  F1
dt
(S1)
dP2
 I P 2  F2
dt
(S2)
dM
 (1  f D )  F1  F2   F3
dt
(S3)
dQ
 F4  F5
dt
(S4)
dD
 I D  f D  F1  F2   g D  F12  F3  F14, EP1  F14, EP 2  F14, EM   F6  ( F4  F5 )
dt
(S5)
dBA
 F6  ( F7  F8 )  F9  F10   F12  F13, EP1  F13, EP 2  F13, EM 
dt
(S6)
dBD
 ( F7  F8 )  F11
dt
(S7)
dEP1
 F13, EP1  F14, EP1
dt
(S8)
dEP2
 F13, EP 2  F14, EP 2
dt
(S9)
dEM
 F13, EM  F14, EM
dt
(S10)
dCO2
 F9  F10   F11
dt
(S11)
d
( P1  P2  M  Q  D  BA  BD  EP1  EP2  EM )  I P1  I P 2  I D  F9  F10  F11 
dt
(S12)
where the state variables (C pools) are described in Table S1 and Fig. 1; Eq. S11 indicates the
total heterotrophic respiration flux and Eq. S12 expresses the overall mass balance of the system.
The transformation fluxes (Fig. 1) are elucidated by Eqs. S13–S26 in Table S1.
3
Table S1
Component fluxes in the MEND model (parameters are described in Table 1 in the main text)
Flux description
Particulate organic carbon (POC) pool 1 (P1)
decomposition (F1)
Equation
F1 
VP1  EP1  P1
K P1  P1
(S13)
POC pool 2 (P2) decomposition
F2 
VP 2  EP2  P2
K P 2  P2
(S14)
F3 
VM  EM  M
KM  M
(S15)
Mineral-associated organic carbon (MOC, M)
decomposition
Adsorption (F4) and desorption (F5) between
dissolved organic carbon (DOC, D) and adsorbed
DOC (QOC, Q)
DOC (D) uptake by microbes
Dormancy (F7) and reactivation (F8) between
active (MBA) and dormant (MBD) microbial
biomass (BA and BD)
MBA (BA) growth respiration (F9) and
maintenance respiration (F10)
F4  K ads  1  Q / Qmax   D
F5  K des  (Q / Qmax )
F6 
1
VD  mR  D  BA
YG
KD  D
F7  1  D /( K D  D) mR  BA
F8  D /( K D  D)  mR  BD
 1
 V  BA  D
F9    1 D
 YG  K D  D
 1
 m  BA  D
F10    1 R
 YG
 KD  D
Synthesis of enzymes for P1 (EP1, F13,EP1),
enzymes for P2 (EP2, F13,EP2), and enzymes for M
(EM, , F13,EM)
F11    mR  BD
F12  (1  pEP  pEM )  mR  BA
F13,EP1 = P1 / (P1 + P2 )× pEP × mR × BA
F13,EP2 = P2 / (P1 + P2 )× pEP × mR × BA
Turnover of enzymes (EP1, EP2, EM)
F14,EP1  rE  EP1
MBD (BD) maintenance respiration
MBA (BA) mortality
(S16)
(S17)
(S18)
(S19)
(S20)
(S21)
(S22)
(S23)
(S24)
(S25)
F13, EM  pEM  mR  BA
F14,EP 2  rE  EP2
(S26)
F14, EM  rE  EM
Notes: Italic symbols like Fi represent component fluxes in equations. Italic symbols P1, P2, M, Q, D, BA,
BD, EP1, EP2, and EM are state variables (soil carbon pools) in equations.
4
Table 1 in the main text
Microbial-ENzyme-mediated Decomposition (MEND) model parameters
Parameter
Description
Apriori range
Eq
Units
VP1
Maximum specific decomposition rate for P1
(0.1, 3)
S13
mg C mg−1 C h−1
VP2
Maximum specific decomposition rate for P2
(1, 50)
S14
mg C mg−1 C h−1
KP1
Half-saturation constant for P1 decomposition
(1, 100)
S13
mg C g−1 soil
KP2
Half-saturation constant for P2 decomposition
(1, 100)
S14
mg C g−1 soil
VM
Maximum specific decomposition rate for M
(0.05, 2)
S15
mg C mg−1 C h−1
KM
Half-saturation constant for M decomposition
(10, 1000)
S15
mg C g−1 soil
VD
Maximum specific uptake rate of D for growth
(0.0001, 0.5)
S18
mg C mg−1 C h−1
KD
Half-saturation constant for uptake of D
(0.05, 0.5)
S18
mg C g−1 soil
mR
Specific maintenance rate of BA = VD ∙ α /(1− α)
S18
mg C mg−1 C h−1
α
= mR /( VD + mR)
(0.01, 0.5)
β
Ratio of dormant maintenance rate to mR
0.001
S23
—
YG
True growth yield
(0.1, 0.9)
S18
—
fD
Fraction of decomposed P1 and P2 allocated to D
(0.1, 1)
S3
—
gD
Fraction of dead BA allocated to D
(0.01, 1)
S1
—
pEP
Fraction of mR for production of EP1 and EP2
(0.001, 0.1)
S25
—
pEM
Fraction of mR for production of EM
(0.001, 0.1)
S25
—
rE
Turnover rate of EP1, EP2, and EM
(0.0001, 0.01)
S26
mg C mg−1 C h−1
Qmax
Maximum DOC sorption capacity
(0.5, 5)
S16
mg C g−1 soil
KBA
Binding affinity
(1, 16)
Kdes
Desorption rate
(0.0001, 0.1)
Kads
Specific adsorption rate = Kdes ∙ KBA
LCF0
r0
—
(mg C g−1 soil) −1
S17
mg C g−1 soil h−1
S16
mg C mg−1 C h−1
Initial fraction of P1 = P1/(P1+P2)
(0.1, 0.95)
—
Initial active fraction = BA/(BA+BD)
(0.01, 1)
—
Notes: The column “Eq” lists the major equation # (see Supplementary Information) in which each
parameter is used.
5
Figure S1
Comparison between observed (Obs) and simulated (Sim) microbial biomass carbon (MBC) in
three soils (Gelisol, Andisol, Ultisol) with addition of Glucose or Starch. MEND and
MEND_wod denote the MEND model with and without dormancy, respectively. The error bars
are standard deviations estimated by the Critical Objective Function Index (COFI) method.
Sample sizes are 2559, 2672, and 2460 for the three soils, respectively.
6
7
8
Figure S2
Boxplots of MEND model parameter values derived by the Critical Objective Function Index
(COFI) method. Parameters are described in Table 1. The bottom and top of the box denote the
25th and 75th percentile (the lower and upper quartiles), respectively. The band near the middle of
the box is the 50th percentile (median). The small square in the box is the mean value. The ends
of the whiskers represent the lowest datum within 1.5 IQR (interquartile range) of the lower
quartile and the highest datum within 1.5 IQR of the upper quartile. The data beyond the two
ends of the whiskers might be considered outliers. Different letters above the boxes indicate the
parameter samples originate from significantly different distribution at a significance level of
0.05 according to the Kruskal-Wallis (KW) test. Sample sizes are 2346–2672 for the four soils.
9
10
11
12
13
14
15
16
17
18