New version of PnET-Succession with important bug fix

[Brian Miranda]

Version 2.1 of the PnET-Succession extension has just been released on the LANDIS-II website (http://www.landis-ii.org/extensions/pnet-succession).  This update includes an important bug fix that relates to dispersal and the determination of seed sources (mature cohorts).  The prior version was using the youngest cohort of a species to determine whether a mature cohort was present that could serve as a source of seed.  The code has been corrected to check whether the oldest cohort is at or above the maturity age to determine seed sources.

This bug has implications for dispersal by seeding, but only in cases where multiple cohorts of a species are present on a site, and at least one cohort is below the age of maturity.  The condition of multiple cohorts of disparate ages is most likely to occur for shade tolerant species that can establish underneath other mature cohorts.  We encourage users to update to the latest version of the extension as soon as possible.  The PnET-Output extension has also been updated to v2.1 with only minor changes (http://www.landis-ii.org/extensions/pnet-output).

The full list of changes in v2.1 is listed below (also in the Users Guide):

• An important bug related to dispersal was fixed in this version.  Previously, the age of the youngest cohort of a species was used to determine if a mature cohort was present on a site for seeding purposes.  The test should use the age of the oldest cohort of a species to check for maturity and determine sources of seed for dispersal.  This error has been corrected.
  
• Rename [SurfaceRunoff] to [PrecLoss] in the Site Output file to distinguish between water lost due to soil saturation (SurfaceRunOff) and water lost due to other factors (PrecLoss; e.g., slope, impervious surface).  Add tracking of PrecLoss variable.
  
• Rename [Layer] to [TopLayer] in the Cohort Output file to denote that the value reported is the highest layer in which the cohort appears.  The top canopy layer has the highest layer value.
  
• The allocation of precipitation events to subcanopy layers has been adjusted so that the precip events are randomly assigned to layers, but not constrained to a single event per layer.  This can result in multiple precip events occurring (with their associated runoff, interception, leakage, etc.) for a given layer, especially when the number of precipitation events is greater than the number of subcanopy layers on a site.  This resolves a discontinuity in the water cycle when the number of cohorts was low relative to the number of precip events.
  

Other Bug Fixes:

• ·         Fixed the DiscreteUniformRandom function to be inclusive of the maximum value.  Previous implementation may have been slightly biased in the shuffling of subcanopy layers.

Cheers,

-Brian Miranda