Fire Effects on Ecosystem Carbon and Nutrient Budgets

1
Fire Effects on Ecosystem Carbon and Nutrient
Budgets

• D.W. Johnson, J.D. Murphy, W.W. Miller, D.W.
  Glass, and R.F. Walker
• Natural Resources and Environmental Science
• University of Nevada, Reno

2
Fire and gaseous losses
Nutrient Volatilization temperature OM 100oC
(all volatilized) N 200oC (nearly all
volatilized S 375oC (significantly
volatilized P and K 777oC (only at high
temperatures) Ca 1484oC (left in
ash) (Raison et al., 1985)
3
Typical Initial Effects of Wildfire on Soil
Chemistry
Nutrient/ion Effect Total C or 0 Total
N or 0 Total S or 0 Total P, K, Ca,
Mg 0 NH4, NO3- Ortho-P
or SO42- Ca2, Mg2, K
4
Typical Effects of N-fixers (e.g., post-fire) on
Soil Chemistry
Nutrient/ion Effect Total C Total
N Total S 0 Total P,K,Ca,Mg 0 NH4,
NO3- Ortho-P , , or 0 SO42- or
0 Ca2, Mg2, K
5
The Global Carbon Cycle
Atmosphere 748 Annual Increase 3.9
Fossil Fuels 6
100-115
Fire 3-5?
Respiration 60
Detritus 60
Decomposition 60
Terrestrial Biota 500-800
Oceans 1000
Soil and Litter 1500-1600
6
Fire Effects on Litter and Soil Organic C
What burns completely volatilizes to
CO2, CH4
Very hot fires can cause soil organic C losses
Litter Layer Organic Carbon
Soil Organic Carbon
7
Fire Effects on Litter and Soil N
What burns completely volatilizes to
N2, N2O, NH3
Very hot fires can cause immediate volatilization
of organic N, NO3-, and NH4 from mineral soil
pH-induced NH4 --gt NH3 volatilization after the
fire
Litter Layer Organic Nitrogen
Soil Organic Nitrogen
Less intense fires cause NH4 increase because of
denaturing proteins, etc and has little effect
on soil organic N
Nitrification (later)
NH4
NO3-
Nitrate leaching (later)
Ammonium leaching (inhibited by strong soil
absorption)
8
Fire Effects on Litter and Soil P
Partially volatilized at temperatures above 777oC
only
Ortho-P in ash
Litter Layer Phosphorus
Oxidation to ortho-P
H2O
Soil Organic Phosphorus
Either increases or decreases in Ortho-P
pH-induced desorption
Soil Adsorbed Phosphorus
Leaching can be slightly elevated
Ca-phosphates (apatite)
Precipitation with Ca
9
Fire Effects on Litter and Soil S
Partially volatilized at temperatures above 375oC
Litter Layer Sulfur
Large increases in SO42-
Oxidation to SO42-
Soil Organic Sulfur
Soil Adsorbed Sulfate
pH-induced desorption
Increased leaching
10
Fire Effects on Litter and Soil K
Volatilization in hot fires only
K oxides
Oxidation
Litter Layer K
Soil Organic K
Soil Primary Mineral K
H2O
Increased Exchangeable K
Maybe slight increase in weathering with hot fire
Increased leaching, but most stays in soil
11
Fire Effects on Litter and Soil Ca and Mg
No volatilization
Ca, Mg oxides
Oxidation
Litter Layer Ca, Mg
H2O
Soil Organic Ca, Mg
Soil Primary Mineral Ca, Mg
Increased pH, exchangeable Ca2, Mg2
Maybe slight increase in weathering with hot fire
Increased leaching, but most stays in soil
12
Methods for Measuring Nutrient Losses with Fire

• Gaseous Losses
• Before and after sampling (not usually possible
  with wildfire)
• Gas sampling during fire (expensive)
• Erosional losses
• Stream discharge
• Runoff plots
• Point samples (gps, survey)
• Soil profiles
• See Hicks, D.L. 2001. A summary of techniques for
  measuring soil erosion. New Zealand Ministry for
  the Environment
• http//www.mfe.govt.nz/publications/ser/tech-repor
  t-69-land-nov01.html)

13
Pre-fire soil augering for chemistry samples
Pre-fire forest floor sampling
Pre-fire quantitative soil pit for accurate bulk
density, rock content, and soil weight
Post-fire soil sampling
14

• Quantitative Soil Pit
• For bulk density, rock content, and soil weight
• Approximately 4 hours per pit

15

• The New Way Quantitative Core
• Proportional sample (cuts through most rocks)
• Approximately 0.5 hours per pit

Diamond-bit soil auger
Mounted on post-hole digger
Sample at 100cm in rocky soil 30 minutes from
start
16
Methods for Measuring Nutrient Losses with Fire
(cont)

• Soluble losses
• Stream discharge
• Runoff plots
• Lysimetry
• Water collectors (tension or tension-free) need
  water flux estimates to go with concentrations
• Resin collectors collect cumulative fluxes but
  no water quality data

17
Solution Collectors we use in the Sierra Nevada

• Snowmelt and ceramic cup lysimeter
• Cumulative flux and water quality with snowmelt
• Water quality only with lysimeter (need soil
  water flux data)

• Resin lysimeter
• Cumulative leaching flux
• No water quality information

18
Methods for Measuring Nutrient Transformations
with Fire

• Pre- and post-fire forest floor and soil analysis
• Post-fire sampling of ash (i.e., with dust
  buster)
• In the case of ephemeral nutrients (i.e., NH4,
  NO3-)
• one can miss the peaks and valleys
• Resin techniques capsules vs ion exchange
  membranes

19
Resin techniques for measuring soils nutrient
availability

• Soil nutrient availability
• Resin beads
• Resin bags (nylon stockings) originally (Binkley)
• Commercially available capsules now
• Membranes
• Membranes pieces originally
• Plastic stakes with imbedded membranes now

20
www.wecsa.com
21
Western Ag Innovations, Saskatoon
Canada www.westernag.ca
22
Resin techniques vs soil sampling for nutrient
availability

• Resins are more sensitive to soil moisture
• In dry soils, traditional extractions will
  measure relatively higher values than resins
• Which more accurately reflects nutrient
  availability?
• Resins integrate values over time whereas
  traditional
• soil extractions are snapshots
• Resins are more representative for ephemeral
  nutrients such as NO3- and NH4
• Need to keep well below saturation for resins

23
Case studies from Sierra Nevada Mountains Site
Locations
Reno
Truckee Prescribed fire in 2001
California
Nevada
North Lake Tahoe Prescribed fire in 2003
Little Valley Fire in 1981
Lake Tahoe
Gondola Wildfire in 2002
24
North Shore Prescription Fire

• Treatments
• Control (no harvest, no burn
• Harvest (thin, understory removal, chip residues)
• Burn
• Burn Harvest

25
North Shore Resin Stakes Results
26
Gondola Fire
A Wildfire on 3 July 2002 burned 9 of 16
previously established plots, allowing pre- and
post-fire sampling with unburned controls!
1 3 5 7 9
11 13 15
2 4 6 8 10 12
14 16

• Immediate post-fire
• Complete forest floor combustion
• Nearly complete foliage combustion
• Soil organic matter losses measured

Pre-fire and control plots Pinus jeffreyii,
Abies amabilis
27
Entisols and Inceptisols derived from decomposed
granite
28
Physical rock weathering during the fire
29
The Gondola fire almost completely removed the
forest floor and all associated nutrients
30

• N leaching fluxes (resin lysimeters) at Gondola
• Significant increases for three years post-fire
• Shift from NH4 to NO3- in years 1-2 post-fire
  (nitrification?)
• Total additional leaching is small (20 kg ha-1)
  relative to volatilization ( gt 400 kg ha-1)

Fire
2001-2002 2002-2003
2003-2004 2004-2005
31

• P leaching fluxes (resin lysimeters) at Gondola
• No increase in year one post-fire
• Significant increase in year two post-fire

Fire
2001-2002 2002-2003
2003-2004 2004-2005
32

• Soil solution NH4 and NO3- (ceramic cup
  lysimeters) at Gondola
• Initial increase in NH4, followed by NO3-, as in
  resin lysimeters
• Probably reflects fire-induced release of NH4
  followed by nitrification

33

• Soil solution P and S (ceramic cup lysimeters) at
  Gondola
• Small increase in ortho-P leaching
• Very large increase in sulfate leaching

Fire
34
Immediate post-fire changes in N Gondola site
35

• Effects of wildfire on soil chemistry at Gondola
  one year after the fire
• Losses of C and N in A horizon

36

• Effects of wildfire on soil chemistry at Gondola
  one year after the fire
• Slight increases in base cations and pH

37

• Effects of wildfire on soil chemistry at Gondola
• No significant effects on extractable P
• Slight increases in extractable SO42-

38

• A major post-fire erosion event 3 weeks after the
  fire
• Note hailstones on the ground
• Tan streaks are rills down to hydrophobic layer
  and were bone dry directly after this event
• Eroded ash collected in a riparian area below and
  was quantified, but the area from which it came
  remains unclear.

39
Eroded ash collected in a riparian area below and
was quantified, but the area from which it came
remains unclear.
40
Eroded ash collected in a riparian area below and
was quantified, but the area from which it came
remains unclear.
41
(No Transcript)
42
Little Valley, Nevada Reconstructing nutrient
budgets from a wildfire in 1981
Pre-fire Pinus jeffreyii
Pre-fire Pinus contorta

• Immediate post-fire
• Foliage and forest floor are totally combusted
• Soil organic matter losses unknown

• 20 years post-fire
• 80 N-fixing Ceanothus velutinus
• 20 non-fixing shrubs

43
Little Valley Wildfire in 1981No pre-fire
dataSamples from former fire and nearby
forestReconstructed nutrient budgets

• Immediate post-fire
• Foliage and forest floor are totally combusted
• Soil organic matter losses unknown

• 20 years post-fire
• 80 N-fixing Ceanothus velutinus
• 20 non-fixing shrubs

44

• Effects of wildfire and post-fire vegetation on
  soil chemistry at Little Valley 20 years after
  the fire
• Greater C and N in former fire site

45

• Effects of wildfire and post-fire vegetation on
  soil chemistry at Little Valley 20 years after
  the fire
• Much greater base cations and higher pH in former
  fire site

46

• Effects of wildfire and post-fire vegetation on
  soil chemistry at Little Valley 20 years after
  the fire
• No significant effects on extractable P or SO42-

47
Pinus jeffreyii
Ceanothus velutinus
Previous observations in mature (not recently
burned) systems in upper Little Valley (Johnson,
1995) Comparisons of soils in beneath 5 paired
adjacent, mature stands of Ceanothus velutinus
and Pinus jeffreyii in upper Little Valley
48
Red alder, a post-fire N fixing species in the
Pacific Northwest, causes substantial nitrate
leaching (Van Miegroet and Cole, 1984)
Coats (1976) has found elevated nitrate in soil
solutions beneath riparian mountain alder in the
Tahoe basin. Does snowbrush cause increased
nitrate leaching?
49
Soil solutions from snowbrush-dominated former
fire had slightly elevated nitrate and ortho-P
concentrations (Stein, 2006)
50
Runoff solutions from snowbrush-dominated former
fire had substantially elevated nitrate and
ortho-P concentrations at certain peak
times (Stein, 2006)
51
Soils chemistry in adjacent Ceanothus velutinus
and Pinus jeffreyii in Little Valley, Nevada
(Johnson, 1995)
52
Soils chemistry in adjacent Ceanothus velutinus
and Pinus jeffreyii in Little Valley, Nevada
(Johnson, 1995)

• Conclusions
• Like red alder in the Northwestern US, Ceanothus
  in the Sierra enriches the soil in C and N
• Unlike red alder, Ceanothus does not result in
  soil acidification or reduction in extractable P

53
Conclusions

• Wildfire at Gondola caused elevated NH4, NO3-,
  ortho-P, and (especially) SO42- concentrations
  for two years.
• Post-fire erosion was substantial, but seemed to
  be largely limited to one major storm event three
  weeks after the fire.
• Occupation of the Little Valley fire (1981) by
  N-fixing snowbrush for two decades caused only
  slight elevations in NH4 and NO3-.

54
Ecosystem Contents Two Decades after the Fire at
Little Valley
55

• Reconstructed C and N Budgets at Little Valley
• Losses by fire and salvage logging
• Post-fire changes after 20 years

Deposition 10
Leaching 2
56
Ecosystem Contents Two Decades after the Fire at
Little Valley
57

• Reconstructed Ca and Mg Budgets at Little Valley
• Losses by salvage logging (assumed no fire loss)
• Conversion to ash
• Post-fire changes over 20 years

Deposition 40
Deposition 260
Leaching 90
Leaching 600
58

• Little Valley reconstructions suggest large
  release of soil mineral and/or organic Ca and Mg
  during fire.
• What do results from Gondola say?
• Fire caused no significant effects on Ca or Mg!

Calcium
Magnesium
59

• Little Valley reconstructions assume no soil
  organic matter combustion. What do results from
  Gondola say?
• Fire caused significant losses of both C and N in
  A horizon

Carbon
Nitrogen
60
However, total soil C losses were minimal most C
was lost from forest floor and vegetation. After
20 years, soil C in former fire in Little Valley
is only slightly higher than in the forest and
ecosystem C content is much lower.
61
Similarly, total soil N losses were minimal in
the Gondola fire. Most N was lost from forest
floor and vegetation. After 20 years, both soil
and ecosystem N contents in former fire in Little
Valley is significantly higher than in the forest

62
Soil Ca increases after fire at Gondola were not
as inexplicably large as at Little Valley. Thus,
we hypothesize that post-fire increases in soil
Ca in Little Valley were due largely to uptake
and cycling by post-fire shrub vegetation.
Pre Post (1 yr)
Forest Fire (20 yr)
Gondola Little Valley
63
In contrast to Little Valley, soil magnesium
showed a net loss at Gondola reasons not yet
clear. Again, we hypothesize that the increases
in soil Mg at Little Valley are due to uptake and
cycling by post-fire shrub vegetation
Pre Post (1 yr)
Forest Fire (20 yr)
Gondola Little Valley
64
How good were the assumptions for fire N budget
reconstruction in Little Valley?

• Results
• Good for foliage and forest floor loss
• Low for soil loss
• Leaching is probably negligible in comparison to
  other losses
• Very uncertain for erosion

65
Conclusions from Gondola and previous studies on
older fires

• Wildfire caused
• Large losses of C and N by volatilization
• Significant increases in N leaching, but
  volatilization
• dominates N loss (gt90)
• Comparisons with older burned sites shows that
• C losses not made up until the site is reforested
• N losses can be made up quickly if N fixers are
  present

66
Prescription Fire
LOW TO MODERATE BURN SOIL TEMPERATURE lt 50o C
67
N losses from Prescribed Fire Many studies
68
N losses from Prescribed Fire Many studies
Loss from 0-10 cm soil only
69
Truckee Harvest x Prescribed Fire Study Pinus
jeffreyii forest
70
Ultic Haploxeralfs derived from andesite
71
(No Transcript)
72
(No Transcript)
73
(No Transcript)
74

• The prescription fire at Truckee showed no
  significant effects of any treatment on
• Soil solution nutrient concentrations
• Cumulative soil leaching (resin lysimeters)
• Soil chemical composition
• The only effects treatments were on nutrient
  budgets
• C and nutrient export via harvested material
• C, N, and S losses from the forest floor and
• understory during burning

75
Approximately half the forest floor was consumed
(typical of prescribed fire in the Sierra Nevada)
76
(No Transcript)
77
(No Transcript)
78
(No Transcript)
79
(No Transcript)
80
(No Transcript)
81
Soil Solution at Truckee
82
Truckee C and N Budgets Treatment No harvest,
burn only
83
Truckee C and N Budgets Treatment Bole only
harvest and then burn

• Following harvest
• Decrease in
• vegetation
• Increase in forest
• floor due to residue
• inputs
• Following burn
• Largest decrease in
• forest floor pools
• Conclusion
• Fire dominates losses

Total loss -35.2
Total loss -318
84
Truckee C and N Budgets Treatment Whole-tree
harvest followed by burn

• Following harvest
• Largest decrease
• in vegetation
• Minimal change in
• forest floor
• Following burn
• Decrease in
• forest floor pools
• Conclusion
• Harvest dominates losses

Total loss -36.7
Total loss -282
85
Summary of Results for C Loss (Mg ha-1) Truckee
Harvest x Prescribed Fire Study
86
Summary of Results for N Loss (kg ha-1) Truckee
Harvest x Prescribed Fire Study
87
Truckee Site - Conclusions

• Little or no effect of fire on soil solution,
  soil leaching, or soil nutrients
• Substantial effect of harvest and fire on forest
  floor C and N loss
• Harvest effects gt fire in whole-tree harvest
• Harvest effects lt fire in bole only harvest
• Total removals approximately the same in both
  cases

88
Although prescription fire is often less severe
than wildfire, it can sum up to greater N losses
than wildfire over time and may prevent
replacement by N fixers
Simulated N losses with regular prescription fire
using a spreadsheet model (after Johnson et al.,
1998)
89
Nitrogen Fluxes - Little Valley, NV
kg ha-1 yr-1
Deposition (mostly snow)
Lv nutrient cycle
Wildfire 3 (300-400 kg ha-1 100 yr-1)
0.3
Post-fire N-fixation 35-70
Prescription Fire 7-15
12
Snowmelt
0.7
Litterfall
0.03
Leaching
90
Overall Conclusions

• Fire can have substantial short-term effects on
  soil nutrient availability which can in turn
  affect
• Water quality
• Initial revegetation, invasives, etc.
• Over the longer term, the major effects of fire
  on site nutrient status are dominated by volatile
  N losses during the fire and post-fire N fixation
  (if any)