Regional drought and oak decline/mortality trends in the Ozark Highlands of Arkansas and Missouri
Zhaofei Fan, Department of Forestry, Mississippi State University
Xiuli Fan and Hong He, Department of Forestry, University of Missouri
Martin A. Spetich, U.S. Forest Service Southern Research Station, Hot Spring, AR
Stephen R. Shifley, U. S. Forest Service Northern Research Station, Columbia, MO
W. Keith Moser, US Forest Service Northern Research Station, St. Paul, MN
Mortality (density)
Since the late 1970’s oak decline and mortality have
plagued Midwestern upland oak-hickory forests,
particularly oak species in the red oak group (Quercus
Section Lobatae) across the Ozark Highlands of
Missouri, Arkansas and Okalahoma (Dwyer et al. 1995).
Advanced tree age and periodic drought, as well
as armillaria root fungi and red oak borer attack
are believed to contribute to oak decline and
mortality (Starkey et al. 2004). Declining trees
first show foliage wilt and browning followed by
progressive branch dieback in the middle and/or
upper crown. Trees eventually die if crown dieback
Continues. We analyzed oak mortality by species
(groups) and inventory year to illustrate the general
trend of oak decline and mortality by using the 19992006 FIA plots measured annually in the Ozark
Highlands of Arkansas and Missouri.
0.12
0.12
0.1
0.1
All oaks
0.08
Mortality (ba)
Introduction
0.06
All nonoaks
0.04
0.08
0.06
All nonoaks
0.04
0
0
1999
2000
2001
2003
2004
2005
1999
2006
0.18
0.18
0.16
0.16
Red oak group
0.12
0.1
0.08
White oak group
2004
2005
2006
Red oak group
0.06
0.02
0.02
White oak group
0
2000
2001
2002
2003
2004
2005
1999
2006
Inventory year
0.24
0.22
Blackjack oak
0.2
0.16
Scarlet oak
Black oak
0.14
Mortality (ba)
0.18
0.12
0.1
Northern red oak
0.08
0.06
0.04
Southern red oak
0.02
0
1999
2000
2001
2002
2003
2004
2005
2000
2002
2003
2004
2005
2006
Blackjack oak
Black oak
Scarlet oak
Northern red oak
Southern red oak
1999
2006
2001
Inventory year
0.28
0.26
0.24
0.22
0.2
0.18
0.16
0.14
0.12
0.1
0.08
0.06
0.04
0.02
0
2000
2001
2002
2003
2004
2005
Figure 1. Proportions of dead tree number and basal area (ba) by species and species groups in the Ozark Highlands
of Arkansas and Missouri.
2
2
1
0
1999
-1
-2
-3
2000
2001
2002
2003
2004
May
June
July
Aug
Sept
Oct
-4
AR Division 1
2005
2006
2007
PDSI
PDSI
1
0
1999
-1
3
May
June
July
Aug
Sept
Oct
2000
2001
5
May
June
July
Aug
Sept
Oct
2
1
2002
2003
2004
2005
2006
2007
0
1999
2000
2001
2002
2003
2004
-1
-2
-3
-2
-4
-3
2005
2006
4
3
2
2007
PDSI
3
PD SI
3
May
June
July
Aug
Sept
Oct
2000
2001
2002
2003
2004
-2
-3
AR Division 2
-4
MO Division 4
As shown, tree characteristics (crown class and dbh) and
competition condition were most important to oak decline and
mortality at stand or small scales (Fan et al. 2006, Shifley et
al. 2006). We will use a hierarchical approach to explore how
regional climate patterns like drought interact with local factors
to affect oak decline and mortality.
Major references
1
0
1999
-1
As shown by published dendrochronological studies, tree
growth (e.g., ring width) was moderately correlated with
drought. Oak decline and mortality trend coincided closely with
regional drought pattern (Figure 2). Drought pattern varied by
divisions in the Ozark Highlands. Contiguous divisions (MO
Division 5 and AR Division 2, MO Division 4 and AR Division
1) shared similar drought pattern. An coincidence analysis of
drought patterns and oak decline and mortality trends
indicated that drought was a major determinant of oak decline
and mortality regionally. Drought condition from May to
October (growing season) of both the current and previous
years contributed to oak decline and mortality. The 2001 and
2004 mortality peaks corresponded with the severe drought
from 1999 to 2000 and from 2002 to 2003.
2006
Inventory year
Inventory year
Compared with nonoak species, oak species overall
showed an escalating mortality trend in terms of proportion of
dead tree number and basal area (Figure 1 top). Red oak
group species decline and resultant mortality are the major
determinant of this trend, for white oak group species
maintained a relatively stable mortality rate in the sequence
(Figure 1 middle). Mortality varied significantly among major
red oak group species with blackjack oak having the highest
mortality rate and southern red oak the lowest mortality rate.
The mortality rank (from high to low) were blackjack oak, black
oak, scarlet oak, northern red oak and southern red oak
(Figure 1 bottom).
Dead tree number and basal area for nonoak and white
oak group species fluctuated around 4% during the sequence,
while for red oak group species as a whole the percentage
increased jaggedly from around 8% in 1999 up to 16-18 % by
2006. Based on the magnitude of regression coefficient
(slope), southern red oak had the greatest increase in
mortality during the sequence, most possibly indicating it was
most susceptible to decline. Next were black oak, blackjack
oak, scarlet oak and northern red oak. This species
susceptibility order conformed to the above mortality rate rank
to a certain extent.
0.1
0.04
0.26
Mortality (density)
2003
0.08
0.04
1999
• We compared oak decline and mortality trends and
palmer drought severity index (PDSI) curves derived
from NOAA data using coincidence analysis to illustrate
the impact of drought on oak decline and mortality.
2002
0.12
0
• We regressed the percentage (transformed) of dead
tree density and basal area against inventory year to
rank oak species by decline and mortality severity
based on the estimated regression coefficient (slope).
2001
Inventory year
0.14
0.14
0.06
2000
0.2
Mortality (ba)
Mortality (density)
2002
Inventory year
0.2
• We calculated the percentage of dead trees in terms
of density and basal area (ba) for different species
(groups) by inventory year based on the 1999-2006
annual FIA plots in the Ozark Highlands of Arkansas
and Missouri.
All oaks
0.02
0.02
Data and Methods
Results and Discussion
MO Division 5
Figure 2. Palmer drought severity index (PDSI) in the divisions of the Ozark Highlands of Arkansas and Missouri from May to October.
2005
2006
2007
Fan, Z., J. M. Kabrick, and S. R. Shifley. 2006. Classification
and regression tree based survival analysis in oakdominated forests of Missouri’s Ozark highlands. Can. J. For.
Res. 36: 1740-1748.
Shifley, S. R., Z. Fan, J. M. Kabrick, and R. G. Jensen. 2006.
Oak mortality risk factors and mortality estimation. For. Eco.
Manage. 229:16-26.