My notes from Matt’s seminar: Theory and field work addressing getting beyond binary decision making. Malaysia. Tropical forest becoming oil palm, rice, rubber, tea.
Reserve Selection vs Landscape Reserve Design
Optimality of Specialized vs Uniform Forest Management
Uniform (sustainable forest management) vs very protected areas, ignore what’s done outside. More intensive disturbance.
An important ecological nonlinearity: What’s the most ubiquitous spatial pattern? Aggregated, random, uniform? (50 hectares is .5 km by km, or 120 English acres, tag every tree taller than 1.3m CTFS, very nice spatial data). [Also an R package, though removed from cran]
((Aggregated? What about Jansen Connell? ))
For a fixed reserve area A, find the number of reserves (of size a) that maximizes the number m of species in the reserve: given that species are aggregated. (assume abundance distribution – log-normal from Hubble (lots of rare species). Neg Binomial model for aggregations. (Assume a minimum viable population, ~100 individuals).
Sanity check cases: under random model, regardless of A, one large reserve is optimal. Very aggregated, one large reserve is never optimal. Treats a two-step problem: economics tell you total area A, step 2: decide how to partition that. (Springborn Question: when should you be doing step 1 conditional on step 2?)
Conclusion: as biodiversity is more valuable, fewer larger reserves is better.
Field study on VJR areas, no pattern on biodiversity. hmm.
(comments on phylogeny & biodiversity?)
(indicator species diversity of impact.)
Biodiversity Conservation challenges under REDD+
Key-issues:
Additionality: No money to do what you already planned
Baseline: Additional compared to what
Leakage: Will this mean more logging elsewhere?
Non-permanence
Measurement
Monitoring
Temporal & Spatial Scales of management: Biodiversity outside the timescale of typical lumber or carbon concerns. How about just RED? Stack the biodiversity aspect, don’t bundle it?