Evaluating the use of digital repeat photography to assess seasonal sensitivity of vegetation to climate in the tropics
Abstract
Plant phenology is the study of life cycle responses of vegetation to seasonal variations in environmental factors. Phenological timing is sensitive to small fluctuations of climate and is easy to observe, thus it is considered to be one of the best biological indicators of variability and change. The relatively low cost and ease of use of ground-based repeat digital photography (DRP) makes this an ideal tool for deployment in remote locations, Indonesia for example. Indonesia is a tropical biodiversity hotspot and an important component of the global carbon cycle. In recent decades, ecosystems found here have been modified as a result anthropogenic activities, such as forestry and mining, as well as changes in climate patterns. Understanding phenological changes among different forest types is important in assessing which forest type appears to be most sensitive to change. To compare phenology among different forest types in Indonesia and to determine whether DRP can pick up seasonal vegetative signals, several landcover types (i.e., plantations, mangrove, primary and secondary forests) were monitored using DRP. Images were analyzed using a Matlab script that compares the brightness of Red, Green and Blue channels to interpret the Green Excess Index (GEI) in response to seasonal climatic fluctuations (i.e. temperature and precipitation). Seasonal trends in green up were compared among the different land cover types and correlated with climate data. Results indicate that the use of DRP can in fact illustrate phenological dynamics in response to seasonal climatic variability. Most sites showed a decrease in GEI during the rainy season, with increasing green up months thereafter. Responses to such changes in environmental factors were most noticeable (sensitive) in heavily impacted landcover types, secondary forests for example, which showed higher variability and sensitivity to environmental changes. Within one secondary forest a significant, positive relationship between greening and temperature (R^2 = 0.082), and greening and precipitation (R^2 = 0.013) (p value <0.01) exists. Continued monitoring of these sites is needed to verify these trends and to help better understand the phenological dynamics of these diverse landcover types. Overall, preliminary results show that this near-surface remote sensing of forest phenology allows for the long-term monitoring of a highly diverse and potentially threatened ecosystem at a relatively low cost. This will not only provide long-standing assessment of forest progression, but will also help communicate the long-term importance of protecting forest ecosystems, and provide data that may aid in policy making and forest conservation.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2013
- Bibcode:
- 2013AGUFM.B51A0263R
- Keywords:
-
- 1632 GLOBAL CHANGE Land cover change;
- 1640 GLOBAL CHANGE Remote sensing;
- 0480 BIOGEOSCIENCES Remote sensing;
- 0439 BIOGEOSCIENCES Ecosystems;
- structure and dynamics