Climate risk on two vegetation axes—Tropical wet-to-dry and temperate arid-to-moist forests

John N. Williams, Raul Rivera, Hyeyeong Choe, Mark W. Schwartz, James H. Thorne

Research output: Contribution to journalArticlepeer-review

3 Scopus citations


© 2018 John Wiley & Sons Ltd Aim: Assess climatic risk to vegetation types associated with tropical and temperate ecosystems based on exposure analysis, which models future risk as a function of deviation from current climate variable distributions. Location: Oaxaca State, Mexico. Taxon: Broadly defined vegetation types used in state- and national-level vegetation classification systems. These types can be grouped into series representing dry-to-wet conditions for tropical and temperate vegetation. Methods: We used climate exposure analysis to compare current and future climate parameters for the major vegetation types of Oaxaca. This technique integrates a recent vegetation map with historical climate data (1981–2010) to produce a baseline climate layer that is compared to climate projections made with five different global circulation models for near-future (2015–2039) and end-century (2075–2099) periods using two emissions scenarios (RCP 4.5 and 8.5). We classified the frequency distribution of the baseline climate into five exposure classes where the closer values are to mean climate conditions, the lower the exposure. Future exposure was estimated by classifying the vegetation pixels into the same exposure classes, now based on future climate values. Increased exposure risk was assessed based on the fraction of pixels that moved into higher exposure classes from one period to another. Results and main conclusions: Our analysis showed four general trends: (a) the higher, current track emissions scenario produced much larger end-century climate exposure; (b) for the tropical vegetation series, tropical evergreen forests are projected as most exposed by end-century; (c) for the temperate vegetation series the matorral-shrubland and conifer forests are the most impacted; and (d) the five GCMs considered showed some convergence in their end-century climate exposure predictions, with coastal and low-elevation areas of the State projected to experience the greatest exposure, and the interior mountain slopes and central region projected to experience the least exposure and be the most climatically stable.
Original languageAmerican English
Pages (from-to)2361-2374
Number of pages2123
JournalJournal of Biogeography
StatePublished - 1 Oct 2018

Fingerprint Dive into the research topics of 'Climate risk on two vegetation axes—Tropical wet-to-dry and temperate arid-to-moist forests'. Together they form a unique fingerprint.

Cite this