van Dijk ACEAS_stradbroke

229 views

Published on

Representing vegetation response in water and carbon models - using phenocams to scale from plant to satellite pixel to model grid cell

Published in: Education
0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total views
229
On SlideShare
0
From Embeds
0
Number of Embeds
68
Actions
Shares
0
Downloads
2
Comments
0
Likes
0
Embeds 0
No embeds

No notes for slide
  • Effect of assimilating (forcing) with MODIS actual data on ET in flux towers. Improvements in Tumbarumba and Howard Springs (forests) when LAI used. Not so much in grassland, open savanna and ash forest.
  • van Dijk ACEAS_stradbroke

    1. 1. Representing vegetation response in water and carbon models using phenocams to scale from plant to satellite pixel to model grid cell Albert van Dijk, Australian National University
    2. 2. Figure 2. Planet 10-4 m 107 m
    3. 3. http://www.bom.gov.au/water/ Australian Water Resources Assessment (AWRA) model
    4. 4. http://www-personal.umich.edu/~ivanov/HYDROWIT/Models.html Carbon cycle model
    5. 5. http://www.cgd.ucar.edu/tss/clm/pfts/igbp.gif
    6. 6. Richardson et al. (201) Global Change Biology 18, 566-584 Comparison of site estimated and biosphere model LAI
    7. 7. Guerschman et al. (in prep.)
    8. 8. Guerschman et al. (in prep.)
    9. 9. http://svs.gsfc.nasa.gov/vis/a000000/a003800/a003871/
    10. 10. Comparison between vegetation cover fraction from AWRA and from MODIS fractional cover product (Guerschman et al)
    11. 11. Performance test Guerschman et al. (in prep.) Table 1: Pearson correlation coefficients (r) between observed and modelled mean monthly evapotranspiration site n %Forest Default LAI EVI Alb LAI+EVI+Alb HoSp 56 0.731 0.800 0.794 0.738 0.830 Kyem 12 0.951 0.905 0.973 0.983 0.946 Tumb 93 0.802 0.858 0.810 0.798 0.850 ViPa 20 0.974 0.971 0.979 0.971 0.969 Wall 16 0.942 0.929 0.945 0.940 0.929
    12. 12. Model simulates seasonal vegetation dynamics Model forced with LAI, EVI and albedo climatology Weighting based on RMSE with MODIS EVI time series Best estimate ET and GPP AWRA-LC carbon (GPP/NPP) product version 0.5, data assimilation / blending Van Dijk, 2009
    13. 13. AWRA-LC v0.5 GPP (gC m-2 mo-1) Has been compared to GPP and NEE data from flux towers and other data cf. Roxburgh et al. (2007) (Van Dijk, 2009; Haverd, unpublished) Carbon exchange estimates
    14. 14. Andela et al. (2013) Biogeosciences 10, 6657-6676 biomass cover
    15. 15. Vegetation remote sensing: • Fuel and habitat structure • Canopy moisture content • Leaf area • Living & dry biomass cover Data acquisition: • LIDAR • Hyper-spectral • Ground-base measurements
    16. 16. ANU Phenomic & Environmental Sensor Array • 5 meteorological stations • triangulated time lapse cameras (1cm/pixel) • 30 wireless microclimate sensors (air temp & hum, light intensity, 2 x soil temp & moisture) • 20 tree diameter sensors
    17. 17. Figure 2. Planet 10-4 m 107 m Figure 3. Phenocam
    18. 18. www.ozewex.org
    19. 19. Biodiversity: finding groundwater dependent ecosystems

    ×