Murray-Darling Basin Biodiversity
Carlos E. González-Orozco and Peter Unmack
Bernd Gruber, Arthur Georges and Duanne White
Monday, 1st June 2015
Institute for Applied Ecology
MDBfutures CRN Biodiversity Project
Contributed Staff Prof Arthur Georges Prof Max Finlayson (CSU) Dr Duanne White Dr Xiuwen Zhang Senior Postdoctoral Fellow Dr Bernd Gruber Postdoctoral Fellows Dr Peter Unmack Dr Carlos Gonzalez-Orozco Dr Mariagrazia Bellio (CSU) Dr Aaron Adamack (ARC) Postgraduate Students Angelica Lopez Kate Hodges Bruno Ferronato Research Assistant Matthew Young
Many people beyond the project also contributed by providing samples, locality information, advice, and other help including: Mark Adams (SA Museum), Michael Hammer (NT Museum), Tim Page, Jane Hughes & Mark Kennard (Griffith Uni), Jon Marshall (Qld DSITI), Bernie Cockayne Qld DNRM), Bruce Chessman (NSW EH), Gavin Butler & Dean Gilligan (NSW Fisheries), Tarmo Raadik (Vic DELWP), Luciano Beheregaray lab (Flinders Uni), James Shelley (Melb Uni) and many others. Kate Hodges, Tim Page, Phil Littlejohn, Alan Couch, Lara Upton, Elena Farnsworth, Margaret Sweeney, Ian Muir, Josh Malony all joined us to help with fieldwork
Acknowledgements
1. Biodiversity and
bioregionalization of the MDB
2. Species distribution modelling
3. Genetic population structure
Top-Down study patterns of
biodiversity
Bottom-Up study processes
that affect biodiversity
Management recommendations
Biodiversity and Conservation
Species Distribution Modelling (Bernd Gruber)
Bioregionalization of the MDB (Carlos E. González-Orozco)
Genetic population structure (Peter Unmack)
Ramsar Wetlands (Max Finlayson)
Sub-projects
One of the main pitfalls in Biodiversity
Biodiversity isn’t just species
• Biodiversity is the whole tree of life, not just the named species
• There are clades smaller and larger than the traditional species level
• Species are not comparable between lineages in any manner, just an arbitrary cut-off somewhere along a branch tree of life
Species richness for eucalyptus in the MDB
Number of species per grid cell
New methods applied to our project
1. Multiple-taxon combinations (Summary maps for raw diversity, clusters of fuzzy
logic, combine maps of significance)
2. Categorical Analyses of Neo- and Paleo- Endemism (CANAPE). Developed as part of software Biodiverse
3. CANAPE modelling under future climate change. Developed as part of software Biodiverse
Data Sources
1. Species distributions: Australia’s Virtual Herbarium, historical collections from museums
2. Phylogenies: Different universities, CSIRO
3. Climate change projections: Hadley global climate model (hadcm3) available from http://wallaceinitiative.org/
Multiple taxon combinations: Species richness for all taxon groups Acacias Eucalypts Plant genera Tree frogs Fishes
Categorical Analysis of Neo and Paleo Endemism (CANAPE)
© J.Plaza RBG Sydney
Paleo-endemism Neo-endemism
Wollemi Pine (200 Mya, Araucariaceae)
Darwin’s Finches (15-20 Mya, Galapagos)
What do we learn from the new methods? CANAPE
Raw values of species richness
CANAPE
Need for hypothesis tests: raw values do not tell us much, unless we understand how much to expect
Maps of CANAPE for five taxonomic groups in the MDB
CANAPE Acacia Eucalyptus Plant genera
Tree frogs Fishes No common phylogenetic
patterns across the studied taxon groups
González-Orozco, C.E et al. (2015) Manuscript2 in prep.
Phylo-bioregions for acacias and eucalyptus in the MDB
acacias
eucalyptus
“Phylo-Jaccard index express the
proportion of lineage diversity
that is shared between two set
of locations”
González-Orozco, C.E et al. (2015) Manuscript2 in prep.
Some conservation applications: acacias and eucalyptus in the MDBA as an example
González-Orozco, C.E et al. (2015) Manuscript3 in prep.
Example of the impact of climate change on phylogenetic diversity of eucalyptus
• Spatial data
• Phylogenetic tree
• Modelling species shifts under current and future scenarios
• Incorporation of phylogenetic approaches under current and future climate change scenarios
Eucalypt species are shifting under climate change by 2085
González-Orozco, C.E et al. (2015) Manuscript1 in prep.
Phylogenetic diversity of eucalypts on the move
González-Orozco, C.E et al. (2015) Manuscript1 in prep.
2014 2085
Disappearance of areas with paleo and neo-endemism of eucalypts by 2085 due to climate change
González-Orozco, C.E et al. (2015) Manuscript1 in prep.
2014 2085
Losses of unique branches from the eucalypts tree under climate change by 2085
González-Orozco, C.E et al. (2015) Manuscript1 in prep.
Eucalyptus (5 sp.) southwest Western Australia
Aquatic Biogeography MDB
M. adspersa
Goal
A. agassizii
Are there common phylogeographic patterns within MDB aquatic species?
Should we manage parts of MDB separately?
Goal
The MDB is surrounded by more basins (22) than any other in Australia
This provides a convoluted and dynamic biogeographic setting with different fishes having relationships with many surrounding basins
R. semoni
Outline
• Within basin biogeographic patterns
• Within basin genetic patterns
• Next generation sequence data
G. bispinosus
MDB contains
36 obligate freshwater fishes
3 diadromous fishes
at least 5 estuarine spp
N. australis
MDB has relatively few narrow range species (9/36)
Species # Basins Basin
Porochilus rendahli 1 Condamine Galaxias Tantangara 1 Murrumbidgee Galaxias fuscus 1 Goulburn Gadopsis bispinosus II 1 Goulburn Gadopsis marmoratus II 1 Wimmera Nannoperca australis B 1 Wimmera Nannoperca obscura 1 Lower Murray tribs Melanotaenia splendida 2 Warrego-Paroo Craterocephalus amniculus 3 Border, Namoi, Gwydir 15 species are essentially found across most of the MDB (>16/24 sub-basins)
MDB contains 14/36 (39%) endemic fishes
Maccullochella peelii (22) Craterocephalus fluviatilis (7)
Gadopsis marmoratus I (21) Galaxias Riffle (5)
Melanotaenia fluviatilis (20) Gadopsis bispinosus I (5)
Macquaria ambigua (20) Craterocephalus amniculus (3)
Galaxias rostratus (12) Galaxias fuscus (1)
Maccullochella macquariensis (12) Galaxias Tantangara (1)
Macquaria australasica (11) Gadopsis bispinosus II (1)
G. rostratus
MDB UPMGA faunal similarity based on Jaccard’s Index
0.0 0.2 0.4 0.6 0.8
UMU
KIE
OVE
BRO
CAM
LOD
MUR
GOU
MBI
LMU
LAC
MAC
LMT
BOR
GWY
NAM
CAS
MOO
DAR
CON
WAR
PAR
AVO
WIM
Murray
Darling
Warrego-Paroo
Wimmera-Avoca
Outline
• Within basin biogeographic patterns
• Within basin genetic patterns
• Next generation sequence data
H. klunzingeri
mtDNA Data
Most fishes in the basin have been examined
Most species show low haplotype numbers, diversity and structure (no multiple ESUs identified within MDB)
Some rivers are slightly distinct, especially Warrego-Paroo
Four species show higher within basin differences which should be considered as MUs
Craterocephalus amniculus C. fluviatilis
Nannoperca australis Gadopsis marmoratus
Microsatellite Data
Few species examined
Similar to mtDNA, but more resolution
M. peelii Lachlan, Macquarie, Gwydir (Rourke et al. 2011)
M. ambigua Macquarie, Paroo (but only slight) (Faulks et al. 2010)
M. australasica unclear given massive decline (Faulks et al. 2011)
N. obscura no diffs within MDB (Brauer et al. submitted)
N. australis most sub-basins distinct (Cole et al. in prep)
G. marmoratus most sub-basins distinct (Lean et al. in prep)
Next-gen SNP datasets are now being gathered for several species
Outline
• Within basin biogeographic patterns
• Within basin genetic patterns
• Next generation sequence data
H. klunzingeri P. grandiceps
Study Groups
Our genetic work is focused on four taxonomic groups:
Australian Smelt Retropinna semoni
River turtle Emydura macquarii
River shrimp Macrobrachium australiense
Yabby Cherax destructor
All are widespread, lack quirks such as hybridisation or cryptic species
Genetic Methods
Samples preserved in the field in liquid nitrogen
DNA extractions
Diversity Arrays Technology conducts the DNA sequencing using “double-digest restriction-site associated DNA sequencing” or ddRadSeq for short
Genetic Methods
ddRadSeq
Genomic DNA is “digested” using two enzymes which cut DNA at specific sites (e.g., AATTCC)
This creates lots of different length fragments
Fragments ~78 base pairs long get saved and sequenced
This produces approximately 2,500,000 reads per sample, or about 20 gigabases of sequence (seven human genomes)
Data is then cleaned up bioinformatically, but is time and computationally intensive
From DNA submission to data is around 3 months
Genetic Methods
DArT produces data file with around 30-40,000 variable loci (locus = unique place within the genome)
Because this data type is new, only a few methods are available
Complicated by the massive size of data files
Bernd is developing new pipeline using R package for analyses
Turtle Genetic Patterns
Turtle Genetic Patterns
RAxML, 400 reps
76 otus, 1246416 bp!
1230818 characters are constant
1896 variable characters are pars-uninformative
Pars-informative characters = 13702
SEQ Pine Sandgate SEQ Wivenhoe
SEQ Pine Bunya SEQ Albert
NEN Tweed Uki NEN Georges NEN Jackadery NEN Yates
NEN Casino NEN Hunter Clarencetown NEN Hunter Isis NEN StClaire
MDB Narromine MDB Mollee
MDB Eulo MDB Toonborough
MDB Boothulla MDB Dartmouth
MDB Ambathala MDB Sandford
MDB Binya MDB Bourke
MDB Gunnedah MDB Pindari MDB Coonabarabran MDB Gin Gin
MDB Bingara MDB Wetherell MDB Lock 10
MDB Barren MDB Bowman
MDB Burrinjuck MDB Lake Forbes
MDB Tarrawingee MDB Tintaldra MDB Laanecoorie MDB Tahbilk MDB Gurra Gurra MDB Mungabareena MDB Morgan Cadel MDB Murray Bridge
MDB Cudgegong MDB Bonshaw MDB Kurmalla MDB Goondiwindi MDB Leslie
MDB Archers MDB Kupunda
MDB Mitchell SEQ Fraser Mackenzie
SEQ Fraser Bowarrady SEQ Mary Petrie SEQ Mary Borumba
SEQ Barambah SEQ Mary Pioneers
SEQ Kolan CEQ Fitz Fairbairn CEQ Fitz Alligator
CEQ Fitz Widewater CEQ Fitz Korcha
CEQ Fitz Warrinilla CEQ Styx
CEQ Laglan CEQ Ross
NEQ Normanby Jacks NEQ Johnstone Warina
NEQ Barron Gardens NEQ Eubenangee NEQ Barron Tinaroo
BULL Como BULL Gumbo BULL Quilpie LEB Dunn LEB Cullyamurra LEB Eulbertie
CEQ Fitz Hoods
LEB
NEN
SEQ
CEQ
NEQ
MDB
SEQ
MDB Turtle Genetic Patterns d = 5
MDB-Ambathala
MDB-Archers
MDB-Barren
MDB-Bingara
MDB-Binya
MDB-Bonshaw MDB-Bourke
MDB-Bowman
MDB-Burrinjuck
MDB-Coonabarabran MDB-Cudgegong
MDB-Dartmouth
MDB-Gin-Gin MDB-Goondiwindi
MDB-Gunnedah
MDB-Gurra_Gurra
MDB-Kupunda MDB-Kurmalla
MDB-Laanecoorie
MDB-Lake_Forbes
MDB-Leslie
MDB-Lock_10
MDB-Mitchell
MDB-Mollee
MDB-Morgan-Cadel MDB-Mungabareena
MDB-Murray_Bridge
MDB-Narromine
MDB-Pindari
MDB-Sandford MDB-Tahbilk MDB-Tarrawingee MDB-Tintaldra
MDB-Wetherell
Subset MDB Turtle Genetic Patterns
Turtle Genetic Diversity NEN
SEQ
CEQ
NEQ
MDB
LEB / BULL
NEN
-Hu
nte
r C
lare
nce
tow
n
NEN
-St
Cla
ire
NEN
-Hu
nte
r Is
is
NEQ
-No
rman
by
Jack
s LE
B-E
ulb
erti
e
LEB
-Cu
llyam
urr
a
MD
B-N
arro
min
e
NEN
-Geo
rges
N
EN-J
acka
der
y
SEQ
-Alb
ert
NEQ
-Jo
hn
sto
ne
War
ina
NEN
-Yat
es
CEQ
-Ro
ss
CEQ
-Lag
lan
N
EQ-E
ub
enan
gee
SEQ
-Mar
y P
ion
eers
SEQ
-Fra
ser
Mac
ken
zie
SE
Q-F
rase
r-B
ow
arra
dy
SEQ
-Pin
e B
un
ya
MD
B-A
mb
ath
ala
NEQ
-Bar
ron
Tin
aro
o
LEB
-Du
nn
NEQ
-Bar
ron
Gar
den
s N
EN-C
asin
o
CEQ
-Sty
x
MD
B-M
olle
e
SEQ
-Wiv
enh
oe
MD
B-G
un
ned
ah
MD
B-M
itch
ell
MD
B-C
on
nab
arab
ran
MD
B-G
in-G
in
MD
B-P
ind
ari
MD
B-L
ake
Forb
es
CEQ
-Fit
z-K
orc
ha
MD
B-K
up
un
da
MD
B-D
artm
ou
th
MD
B-B
urr
inju
ck
CEQ
-Fit
z W
idew
ater
0.20
0.15
0.10
0.05
0.00
BU
LL-Q
uilp
ie
NEN
-Tw
eed
Uki
M
DB
-Les
lie
SEQ
-Mar
y P
etr
ie
MD
B-S
and
ford
B
ULL
-Gu
mb
o
CEQ
-Fit
z W
arri
nill
a
CEQ
-Fit
z A
lliga
tor
MD
B-B
ow
man
M
DB
-Bin
ya
MD
B-B
oo
thu
lla
CEQ
-Fit
z Fa
irb
urn
MD
B-T
arra
win
gee
MD
B-C
ud
gego
na
MD
B-B
arre
n
SEQ
-Ko
lan
MD
B-L
ock
1.0
MD
B-L
aan
eco
ori
e M
DB
-Gu
rra
Gu
rra
MD
B-M
urr
ay B
rid
ge
MD
B-T
ahb
ilk
MD
B-M
org
an-C
adel
M
DB
-Bin
gara
MD
B-G
oo
nd
iwin
di
MD
B-M
un
gab
aree
na
MD
B-K
urm
alla
MD
B-B
on
shaw
MD
B-T
inta
ldra
M
DB
-Bo
urk
e
MD
B-A
rch
ers
SEQ
-Mar
y B
oru
mb
a
MD
B-W
eth
erel
l
SEQ
-Bar
amb
ah
SEQ
-Pin
e Sa
nd
gate
CEQ
-Fit
z H
oo
ds
MD
B-T
oo
nb
oro
ugh
BU
LL-C
om
o
MD
B-E
ulo
Shrimp Genetic Patterns
d = 5
MDB-Ashf
MDB-Bour
MDB-Goon
MDB-Goul
MDB-Laan
MDB-Oven
MDB-Pall
MDB-Toon
MDB-Umur
MDB-Wagg
MDB-Wilc
Shrimp Genetic Diversity
Yabby Genetic Diversity d = 20
MDB-Ashf MDB-Bour MDB-Goon
MDB-Laan MDB-Naga MDB-Oven
MDB-Pall
MDB-Toon
MDB-Umur MDB-Wagg
MDB-Wilc
Yabby Genetic Diversity
Paroo Sth MDB Nth MDB
ARC Linkage Grant
Drivers of fine scale
genetic spatial
structuring in
aquatic organisms
Conclusions
MDB, at least for turtles has high levels of genetic diversity
Paroo River consistently sticks out, although possibly for different reasons
Some north-south separation in shrimp and yabby
Severn River tends to stand out as being a little different
C. fluviatilis
Thank You
M. splendida