monitoring, barcoding network & recording

Report
Monitoring Barcoding Network and
Recording
Matthew Shepherd
Senior Specialist, Soil Biodiversity, Natural England
Why monitor soils?
• Soil science has concentrated on agricultural systems, physical and
chemical status.
• Learn from (semi) natural habitats
– Lessons for managed ecosystems
– Own interest
• A quarter of all biodiversity is found in the soil
Why monitor soils?
• Many monitoring efforts (eg. ECN, RSS) have focussed on chemical
or physical parameters
– yet soil biology does all the work!
• New advice from UK SIC, Defra SQuID project
• CS2007 – more soil and soil biological parameters than ever
before
• Try to be compatible, representative and affordable
Why monitor soils?
• CS survey in 1998 and 2007 measured soil biological parameters
• Measured tRFLP, soil mesofauna
Why monitor soils?
• ~12.8 quadrillion soil invertebrates present in the top 8 cm of GB
soils
• significant increase in total invertebrate catch in all Broad Habitats
• except for agricultural areas on mineral soils
• Due to increase in the catch of mites in 2007 samples
• small reduction in the number of soil invertebrate broad taxa (0-8cm)
recorded
• different seasonal conditions – more work needed
• Needs linking with habitat and chemistry work
• Oribatid data – Thanks to Aidan Keith – now have loose locaitons –
secret data!
13
LTMN Soils Method
13
11
• 1 habitat per NNR for
soil assessment
• 22 so far of ~43 total
• 8 broadleaved
woodlands
• 5 heathlands
• 6 calcareous
grasslands
• 6 neutral grasslands
• 2 dune grasslands
• 2 blanket bogs
• 4 raised bogs
• 5 fens
• 5 saltmarshes
11
13
13
11
12
12
12
11
13
12
13
13
11
11
11
12
11
11
12
LTMN Soils Method
• NE help contribute to
fieldwork and
Macaulay Scientific
Consulting do
fieldwork and
analysis.
• Use aerial photos
and veg survey data
to choose 5 ~similar
points.
• Survey from Sept
16th to Oct 16th
LTMN Soils Method
• Use GPS to locate
veg plot markers and
lay out 20m by 20m
soil plot to SW using
compass
• Each contains 100
2m by 2m sub-plots
• Same 4 sampled for
all plots – change
next time.
LTMN Soils Method
•
•
•
•
Take plot location photos
Sub-plot photos side and above
Vegetation survey
Soil auger assessment
LTMN Soils Method
• Cores taken – most bulked
• Wrapped, labelled chilled and
sent to Scotland.
• Different cores are letter coded:
• C for “curface” (0-15)
• A for “anderneath” (15-30)
• Physico-chemical properties
• Bulk density
• %C, %N
• Loss On Ignition
• pH
• CEC and cations
LTMN Soils Method
• B for beasties (0-8 cm
mesofauna)
• D for DNA (microbial
community) – tRFLP, PLFA
• E for eelworms (nematodes)
• F for fertiliser (N
mineralisation)
Baseline Results – Physico-chemical
Dry Bulk Density g cm-3
1.6
1.4
0 - 15 cm
1.2
15 - 30 cm
1
y = -0.349ln(x) + 1.6604
R² = 0.9749
0.8
0.6
0.4
0.2
0
0
20
40
60
80
% Organic matter (LOI)
100
120
Baseline Results – Physico-chemical
Cation Exchange Capactity (mEq
100g_1
Soil organic matter and cation exchange
capacity
y = 30.422ln(x) - 37.506
R² = 0.6967
140
120
100
80
60
40
20
0
0
20
40
60
80
% Soil organic matter (loss on ignition)
100
Baseline Results – Biochemical
Total Soil PLFA content nm g-1
Soil organic matter and total soil PLFAs
1800
y = 395.32ln(x) - 491.93
R² = 0.8434
1600
1400
1200
1000
800
600
400
200
0
0
20
40
60
80
% Soil Organic Matter (Loss on Ignition)
100
Baseline Results – Soil Function: C storage
140
Carbon storage tonnes ha-1
120
100
80
0-15 cm
15-30 cm
Poly. (0-15 cm)
Poly. (15-30 cm)
60
40
y = -0.0418x2 + 4.6273x - 23.402
R² = 0.6242
20
y = -0.0386x2 + 4.3262x - 21.342
R² = 0.6193
0
0
20
40
60
80
Field water content %
100
Baseline Results - Soil function: decomposition
40
35
Soil C:N ratio
30
25
20
15
y = 0.2703x + 14.995
R² = 0.5885
10
5
0
-10
0
10
20
30
40
50
% Cover of woody species
60
70
80
Baseline results - Soil organism communities: tRFLP
Baseline Results - Soil communities: tRFLP
1
Baseline
Results:
Overall soil patterns (2011 data)
Baseline
results
– interactions
0.8
C:N ratio
ericaceae
0.6
Gram +ve:-ve ratio
PLFA Fun:Bac
ratio
bryophytes
and lichens
0.4
0.2
PCA axis 2
Exch_Acidity
NH4-N_min
trees
ferns
litter Olsen P
Gram +ve
Fe
PLFAact
LOI
Fungal PLFA
%C
% water PLFA_total
Bac PLFA
Al
Na
NO3-N min
%N
sedges
Gram -ve
and rushes
K
bare ground
Mn
Total P
Mg
CEC
shrubs
and climbers
Bulk Density
tRFLP richness
0
-0.2
-0.4
herbs
VAM PLFAs
tRFLP Shannon
Diversity
-0.6
grasses
-0.8
Ca
tRFLP evenness
pH
-1
-1
-0.5
0
PCA axis 1
0.5
1
1.5
Baseline Results: implications for future work
• Size of change detectable varies site to site...
– pH – ~0.4 pH units
– ~20% change in bulk density
– tRFLP - 7% change in evenness, 12% change in richness
• Soil physico-chemical properties change slowly
• Soil biological properties may be more sensitive indicator...
• Different habitats have distinct soil communities
• Soil function – indicators and proxies – more measures needed?
Future analyses and plans
• Continue with baseline – comparison over time
• Include new analyses - earthworms, root biomass, genetic analysis
• Develop new approaches
– Metabarcoding project – CEH & NHM – mesofauna
– Earthworm DNA?
– More multivariate analyses
• Write up – plan to present site by site data and full baseline report
after 5 years
• Comparison with CS2007, CS1998 data – compare agricultural soils
• Apply same methodology in experimental work, other monitoring
Answering the big questions...
• Soil resistance and resilience to perturbations
• Disturbance/fire at Thursley
• What are the soil communities in our priority habitats?
• Clear microbial (and other?) communities
• How do these compare with other habitats?
• Extend this method to other sites/experiments & compare
CS2007
• Do soil characters/function lag or lead changes?
• What will happen to soil carbon in seminatural habitats?
• Trends in soil biodiversity – where are changes seen and why?
– Wait and see!
But...
• Most mesofauna samples are still not sorted and identified
• Anyone interested – can borrow NE microscope
• 5 samples - probably around 500-1000 beasts in total!
We need a way to
identify very large
numbers of
invertebrate
specimens quickly
and cheaply
DNA metabarcoding?
Barcoding and Metabarcoding
• Alternative approach is metabarcoding.
• Mitochondrial DNA passed down female line only – no
recombination during meiosis
• Gradual change by mutations at “regular” rate
• Differences and similarities should indicate timings of
divergence of species.
• Similar story for ribosomal RNA
• Sections of these are used as “barcodes” to characterise
spp.
• COi – cytochrome oxidase 1 gene
• Also 18SRNA
• Prokaryotes- 16SRNA
Barcode Region for Animals
The Mitochondrial Genome
COI
Target
Region
An actual mosquito barcode a 650 letter word:
mosquito-COI:
CGCGACAATGATTATTTTCAACTAACCATAAGGATATTGGAACATTATAT
TTTATTTTTGGAGCTTGAGCAGGAATAGTAGGAACTTCTCTAAGTATTTT
AATTCGAGCAGAATTAGGACACCCTGGAGCCTTTATTGGTGATGATCAAA
TTTATAATGTTATTGTAACAGCTCATGCTTTTATTATAATTTTTTTTATA
GTTATACCTATTATAATTGGAGGATTTGGAAATTGACTAGTCCCTCTAAT
ACTAGGGGCCCCAGATATGGCTTTCCCTCGAATAAATAATATAAGATTTT
GAATATTACCCCCCTCTTTAACTCTTCTAATTTCTAGAAGTATAGTAGAA
AATGGAGCTGGAACAGGGTGAACTGTATATCCTCCTCTATCCTCAGGAAT
TGCTCATGCAGGAGCTTCAGTAGATTTAGCTATTTTTTCATTACATTTAG
CAGGAATTTCTTCAATTTTAGGAGCAGTTAATTTTATTACAACAGTTATT
AATATACGAGCACCAGGAATTACTCTTGACCGAATACCGTTATTCGTTTG
ATCTGTAGTAATTACAGCAGTATTATTATTACTTTCTTTACCAGTATTAG
CTGGAGCTATTACTATACTTTTAACAGATCGAAACTTAAATACATCATTC
• If you can extract, and amplify barcodes from a community – cross
ref with barcodes for known species
• Generate spp. List
• Not quantitative – differential amplification
• Problem – not enough spp. barcoded
• Problem – extraction, amplification methods not well developed
Metabarcoding
Mass sequencing reduces time and cost
Uses CO1 barcode
gene
Next-generation sequencers
e.g. ‘454’ / Illumina
 Produce many parallel
sequences
 Limited in the length of
sequence reads
 Error can occur at amplification
and sequencing stages – leads
to noisy results
~120,000 reads in two 1/8 plates
(~345 bp/sequence)
Step 1: Denoise
~2000 reads
Step 2: Cluster similar sequences
These are our molecular OTUs
~2000 reads
Step 3: Assign taxonomy
Anopheles Mosquito
Spider
Fruit fly
Emerald tree python
Output: species x site table
Which treatments
work?
Control
Agricultural
plough
Swipe
Forestry
plough
Forage
harvest
Disc plough
Control
Turf stripping
Pitfall trap data
Standard (Spiders, carabids, ants)
Standard
Metabarcoding (All arthropods)
R2=0.76
The best treatments are the most aggressive ones
Pitfall trap data
Standard (Spiders, carabids, ants)
Standard
Metabarcoding (All arthropods)
Conclusion: metabarcoding produces useful
information for restoration ecology.
• NE project to develop method – Dave Spurgeon, Rob Griffiths,
Daniel Read at CEH
• 3 sites sampled along transects at differing proximites
– Old spp. Rich chalk grassland
– Improved grassland
– Grassland managed to “revert” to chalk grassland
• 2 sets of mesofauna extracted
– metabarcoding
– morpho ID & spp. barcoding
• Metabarcoding – problems with primers – will COi work?
• 18S RNA better – but good enough for spp?
• Morpho ID shows some differences
• Communities (PCA) similarities are
– old <–> improved <–> reverting
– I’m ID’ing samples for NHM – Alfried Vogler – lots of photos
– Use for this project and put on BOLD
– Challenge for NHM in small size
– Must retain voucher specimen!
• Barcoding a specimen leaves a “permanent” legacy of an ID, and
enables comparison to others (checking or defining)
• Link with location – a genetic NBN
• Many specimens on current databases wrongly ID’ed
• Some barcodes are of foreign material
• Correct group-specific primers should help...
• Better photos of lots of features.
Issues and Questions
•
•
•
•
•
•
What’s stopping you barcoding things you ID?
Reagents?
Costs?
Lack of knowing where to go?
Would coordination help
Is there a role for NE? Museums? Universities? Biological Record
Centres?
• Biodiversity groups and networks?
Soil Biodiversity Support Groups
•
•
•
•
•
•
No soil biodiversity society for UK – SES in USA/Canada
Help is out there!
Facebook page
Blogs
Us lot!
What else?
– Record centres?
And Recording?

similar documents