Observing Tree Mortality from Above and Below: Long ‐ term Remote Sensing and Field Observations David M Bell dmbell@fs.fed.us USDA Forest Service, PNW Research Station February 27, 2018 1 Photo: Rupert Seidl
Presentation Roadmap • Tree Mortality, Forest Change, and the Permanent Sample Plot (PSP) Network • Wind River Experimental Forest: A Case Study in Dwarf Mistletoe impacts on Western Hemlock • Future Opportunities Leveraging Remote Sensing 2
Presentation Roadmap • Tree Mortality, Forest Change, and the Permanent Sample Plot (PSP) Network • Wind River Experimental Forest: A Case Study in Dwarf Mistletoe impacts on Western Hemlock • Future Opportunities Leveraging Remote Sensing 3
Tree Mortality in Western United States • Increases in mortality rates in the late 20 th century • Regional assessment of disturbance and mortality highlight trends in forest change van Mantgem et al. 2009. Science Cohen et al. 2016. Forest Ecology and Management 4
Diversity of Mortality Agents Photo: Emmingham Photo: USFS 5
Permanent Sample Plot (PSP) network 135 installations measured every 5-6 years > 100,000 tagged trees Distribution of 135 Active Permanent Plot Installations by Administrative Unit National Forest Land Research Natural Areas 23 23 Cascade Head, Wind River, 24 Pringle Falls Exp. For. 44 HJ Andrews Exp Forest 21 Mt. Rainier NP, Olympic NP, Mt. St. Helens NM http://pnwpsp.forestry.oregonstate.edu/ 6
Permanent Plot Duration by Seral Stage ( at the time of plot establishment ) 7
Inter-generational Science 2000 1900 1950 Thornton Munger (USFS) Leo Isaac (USFS) Jerry Franklin (USFS/UW) Walter Meyer (USFS) Mark Harmon (OSU) Tom Spies (PNW) 8
Presentation Roadmap • Tree Mortality, Forest Change, and the Permanent Sample Plot (PSP) Network • Wind River Experimental Forest: A Case Study in Dwarf Mistletoe impacts on Western Hemlock • Future Opportunities Leveraging Remote Sensing 9
T.T. Munger RNA
Western Hemlock Dwarf Mistletoe A. tsugense subsp. tsugense • Parasitic plant • Infects 10.8% of western hemlock trees in western Oregon • Infestation spread through projectiles Photo Ray Davis Hawksworth and Wiens 1996 11
Influence on Forest Structure: Witches Brooms and Crown die-back Heavily infected Tsuga heterophylla Un-infected Tsuga heterophylla with A. tsugense. 12
Dwarf Mistletoe Infestation at Wind River Experimental Forest S9: M106-M103 Thornton Taft Munger T. T. Munger 12-ha plot Research Natural Area Gifford Pinchot National Forest 300 S8: M102-M95 ~1 km DMR 6 5 S7: M94-M83 200 4 Mistletoe rating by plot Unifested 3 DMR = 0 DMR = 1 Low Infest. S6: M82-M71 2 DMR = 2 DMR = 3 1 100 DMR = 4 0 DMR = 5 S5: M70-M59 Severe Infest. DMR = 6 N/A S1: M3-M1 Strip number and plot range 0 S4: M58-M47 0 100 200 300 400 S3: M46-M21 WRCCRF 12 ha plot S2: M20-M4 S1: M3-M1 13 Shaw et al. 2008. Canadian J. of Forest Research
Spread and Intensification Uninfested vs. Initial/low Infestation Uninfested vs. Late/Severe Infestation 300 300 DMR DMR 0 0 1 200 6 200 100 100 0 0 0 100 200 300 400 0 100 200 300 400 Shaw et al. 2005. Canadian J. of Forest Research
Dwarf Mistletoe Reduced Growth • Tree growth was reduced for DMR 6 in all years and for intermediate infestations in later years Uninfested Low Infestation Severe Infestation 15
Dwarf Mistletoe and Tree Mortality • Reduced growth preceded increased mortality • Mortality rates were higher after 2004, regardless of growth • What are the implications for patches of mortality? 300 DM R 6 5 200 4 3 2 1 100 0 0 0 100 200 300 400 16
Dwarf Mistletoe and Tree Mortality • Reduced growth preceded increased mortality • Mortality rates were higher after 2004, regardless of growth • What are the implications for patches of mortality? 300 DM R 6 5 200 4 3 2 1 100 0 0 17 0 100 200 300 400
Dwarf Mistletoe and Tree Mortality • Reduced growth preceded increased mortality • Mortality rates were higher after 2004, regardless of growth • What are the implications for patches of mortality? 300 DM R 6 5 200 4 3 2 1 100 0 0 0 100 200 300 400 18
Verifying results in other landscapes • Do we see the same patterns in other parts of the Cascades? • H. J. Andrews Experimental Forest and LTER offers an opportunity HJA Permanent Plots: RS39 Observed presence of witches’ brooms 19
Presentation Roadmap • Tree Mortality, Forest Change, and the Permanent Sample Plot (PSP) Network • Wind River Experimental Forest: A Case Study in Dwarf Mistletoe impacts on Western Hemlock • Future Opportunities Leveraging Remote Sensing 20
Source: USGS Key Features of Landsat Missions 5-8 (1985-present) • Spatial Resolution = 30 m • Multiple observations per year (8-day returns) • History of Forest Change Detection 21
TimeSync Observations of Change Manual delineation of segments Landsat image chips for all years Google Earth High res 22 Cohen et al. 2010. Remote Sensing of Environment
Remote Sensing of Forest Change • Some forest change can be subtle. • Subtle changes develop over several years Forest Change Example (Aerial Photos) 1999 2005 2007 2016
Remote Sensing of Forest Change • Some forest change can be subtle. • Subtle changes develop over several years • Slow, subtle changes in forest canopies (e.g., declines) are now ubiquitous in the western USA. Mountain Western USA Lowland Western USA Cohen et al. 2016. Forest Ecology and Management
Forest “decline” in western US Forests • Across western US forests, multiyear drought preceded forest decline events (Bell et al. in review)
Forest “decline” in western US Forests • Across western US forests, multiyear drought preceded forest decline events (Bell et al. in review) • Forest declines represented in tree mortality (Cohen et al. 2016) • Mortality depended on tree species, density, and size (Bell et al. in prep )
From Trees to Landscapes with Remote Sensing • Linking long-term (or inventory) data with remote sensing for scaling and mapping • Integrating understanding of process with observation of pattern • New cloud computing (e.g., Google Earth Engine) offers technology transfer opportunities 300 DM R 6 5 200 4 3 2 1 100 0 0 0 100 200 300 400
Conclusions and Implications • For western hemlock, dwarf mistletoe is impacting forest dynamics at several scales. • Dwarf mistletoe reduced growth and increased mortality in western hemlock individuals. • Increased tree sensitivity to dwarf mistletoe in recent years implies changes in stressors. • Shifting tree sensitivity could increase the forest area and size of growth reduction and mortality patches. • Jointly, field and remote sensing data can provide a basis for multi-scale monitoring of forest health. • PSP data supports understanding of mechanisms • Remote sensing provides a basis for broader application
Thank You! • Thanks to my collaborators: Dave Shaw, Rob Pabst, Warren Cohen, Andy Gray, Matt Reilly, Tom Spies, and Zhiqiang Yang • Thanks to USFS PNW Research Station, NASA, and HJA LTER for support • Thanks to Wind River for access to the canopy crane • Thanks to a long list of field crews who collected the field data (1926 to present) 29
USFS designations and PSP establishment 1908 – National forests 1930’s: Metolius, Quinault, established in PNW TT Munger, Pringle Falls RNA 1997-98: Hagan, 1913: Wind River EF 1968: Wildcat Mtn RNA 1941: Neskowin RNA Torrey-Charlton RNA 1924: PNW Forest & Range Exp.Stn 1948: HJ Andrews EF 1970’s RNAs: Bagby, Steamboat, Flynn, Goat Marsh, M. Santiam 1931: Pringle Falls EF 1958: Twin Cr RNA 1982: Mt St Helens NM 1934: Cascade Head EF 1900 1950 2000 1947: Douglas-fir 1979-81: Mount St. Helens, (TT Munger RNA) 1910: Douglas-fir Metolius, Neskowin, Hagan (Willamette NF) 1946: Alder-conifer (Cascade Head EF) 1978: Mt. Rainier, Hoh River 2010: High-elev 1938: Ponderosa pine plots (HJ Andrews) (Pringle Falls EF) 1977-80: Noble fir 1914: Douglas-fir (OR, WA) 1935: Hemlock-spruce (Wind River EF) (Cascade Head EF) 2006: Watershed plots 1971-79: HJA reference 1926-30: Douglas-fir (HJ Andrews) stands + watersheds (Olympic NF, Gifford Pinchot NF, Mt. Hood NF) 30
Examples of PSP Measurement Effort 31
Addressing Long-term Forest Dynamics • Examine forest and individual tree change of decades (and even centuries) 32
Dwarf Mistletoe Dispersal 33
Remote Sensing of Forest Change Cohen et al. 2010. Example (Aerial Photos) 1999 2005 2007 2016
Recommend
More recommend
