Title          : Description of L4C (NRv4.1 and Vv3040) Sensitivity Test Runs

Author         : Lucas A. Jones, lucas@ntsg.umt.edu, Univ. of Montana

Date           : 02/05/2018

Project        : NASA IDS Project 2018, Nick Parazoo (PI), Nicholas.C.Parazoo@jpl.nasa.gov, JPL/Cal Tech

Temporal Extent: L4C Nature Run, daily Jan 1, 2000 - Sept 30, 2017
	 	 L4C Vv3040, daily Mar 31, 2015 - Dec 31, 2017

Spatial Format : Subsets for Alaska study locations in 9-km and 9-km x PFT timeseries format.
		 All datasets provided in HDF-5 file format.

Examples       : Example figures provided in the /browse directory.
	         
Notes          : This distribution provides L4C GPP sensitivty (i.e. constraint exclusion) tests
	       	 using the Vv3040 L4C calibration. Each test represents GPP results following  the 
		 exclusion of a constraint or group of constraints. Test treatments described in
		 h5 group-level metadata and listed here for convenience:

		 NoFt - GPP model with freeze-thaw excluded.
		 NoSmrz - GPP model with root zone soil moisture excluded.
		 NoVpd - GPP model with vapor pressure deficit excluded.
		 NoTmin - GPP model with minimum daily air temperature excluded.
		 AparOnly - GPP model with all meteorological constraints excluded (i.e. APAR*LUEmax).
		 Full - Full GPP model with APAR and all meteorological contraints included (same as standard L4C NRv4.1 subset)

Caveats        : As can be seen from the browse images, the freeze-thaw and root zone constraints have little
	       	 impact on the GPP results for the Alaska locations. This is influenced by model configuration and 
		 calibration against global flux tower locations. Boreal/Arctic plant functional-type parameters 
		 (grasslands, shrublands, deciduous broadleaf and evergreen needleleaf forests) are influenced by lower latitude 
		 and more arid locations in the calibration dataset.
		 
		 Browse images also show that the minimum air temperature (Tmin) constraint remains active throughout the summer. 
		 This indicates that Tmin exerts control over the spatial effective maximum light use efficiency, which
		 compensates for latitudinal differences in the calibration dataset.

		 FPAR during the winter months and seasonal transitions can be unreasonably high for short periods (spikes).
		 This results from the MOD15A2H FPAR 'backup' algorithm and is present in Collection 6, but not in Collection 5.
		 L4C transitioned from using Collection 5 and now uses Collection 6 in versions Vv3040 and later.  These spikes
		 are mostly mitigated in GPP by the Tmin and freeze-thaw constraints, but low non-zero winter-time GPP values 
		 can persist in some locations.