Owl Computer Games

Today we explore the methods I purpose to use in my research. I've attached two papers by Nathan Schumaker of the EPA in Corvallis.  He is one of my advisors on my committee.  I will be using the program he developed, called HexSim, to answer my research question.  This is the same program that was recently used to update the Spotted Owl Recovery Plan (release 2 weeks ago).

The Premise

Basically, Nathan programs really cool owl computer games, other than nobody gets to control the owl.  The good news is you can play as long as you want (simulations can go on for 1000 years), and you get as many owl lives as you specify (say 1000 in a population).  These games (actually simulations) are used to explore habitat-species interactions at landscape scales--the scale at which the BLM's WOPR changes land-use.

In terms of the owl, we basically want to figure out if the population will go up or go down.  This is a summed function of births minus deaths over time.  A population whose births = deaths is said to be 'regulated', and is described as having a FITNESS of one (a fitness less than one means the population will go down, more than one and it will go up).  As we all know, the range-wide population of owls is probably still going down at a rate of 3.9% a year (or with a fitness of 0.961).  The question is why.  The traditional answer has been habitat loss, but inclusion of barred owl and climate change add additional uncertainity.  While habitat loss may have been effectively limited by North West Forest Plan that we talked about last time, the fact that population continues to go down doesn't mean that habitat then isn't important (indeed, the rate of decrease has been halved or more, effectively doubling or tripling the amount of time we have before extinction).

Habitat is extremely important for more species, and most demographic characteristics (births and deaths) are fundamentally a result of the quality of habitat.  HABITAT QUALITY is important because it arguably provides a better mechanism for balancing conflicting land-use needs (it is very hard to actually find a pile of dead owls anywhere).

Yet, habitat quality is only one peice of the puzzle.  Just because a particular forest is of a high quality doesn't mean that the benefits it bestoys will be passed to an individual owl.  Thus, OCCUPANCY becomes a second important consideration when understanding how a landscape will affect a particular species.  Occupancy may be affected by fragmentation and connectivity (these are opposites). 

Consequently, population FITNESS = (HABITAT QUALITY) x (OCCUPANCY)

BLM land, with its checker-board pattern landscape, is an archtypical example of fragmentation from human activities.  Thus, when it comes to assessing impacts of proposed land-use changes, it is logical that connectivity / fragmentation needs to be assessed.  Unfortunately, this has not been considered in any of the recent 3 big owl changes: the RECOVERY PLAN, the update to CRITICAL HABITAT, nor [in particular] the WOPR.

This is where the owl computer game comes in...

The Papers

PROJECTING WILDLIFE RESPONSES TO ALTERNATIVE FUTURE LANDSCAPES IN OREGON’S WILLAMETTE BASIN

Schumaker, N. et al. 2004. Ecological Applications, 14(2): 381-396

Schumaker_2004.pdf

**NOTE: good overview of habitat-species models in a cool project that Schumaker collaborated on with David Hulse at UO.  The study looks at habitat quality changes that have happened in the Willamette Valley and as predicted over three different scenarios into the future (much like the IPCC on Climate Change with A1[development], A2[business-as-usual] and B[conservation] scenarios).

USING HABITAT INDICES TO PREDICT HABITAT CONNECTIVITY

Schumaker, N. 1996. Ecology, 77(4): 1210-1225

Schumaker_1996.pdf

**NOTE: only read INTRO and DISCUSSION and scan pictures as they give some idea of how the model works, and what the results look like.**