Spatiotemporal variability and predictability of Normalized Difference Vegetation Index (NDVI) in Alberta, Canada

Area of Science:

• Environmental Science
• Remote Sensing
• Climate Change Studies

Background:

• The Normalized Difference Vegetation Index (NDVI) is crucial for monitoring global vegetation productivity and its response to climate change.
• Alberta, a key agricultural and forestry region in Canada, lacks comprehensive studies on climate change impacts on vegetation productivity.
• Understanding these impacts is vital for policy and stakeholder decision-making.

Purpose of the Study:

• To analyze variations in smoothed NDVI (sNDVI) and climate variables (precipitation, temperature) in Alberta.
• To investigate the relationships between sNDVI and climate variables, assessing predictability.
• To compare linear regression and artificial neural network-genetic algorithm (ANN-GA) models for sNDVI estimation.

Main Methods:

• Non-parametric Mann-Kendall monotonic test and Thiel-Sen's slope for trend analysis.
• Development and comparison of linear regression and ANN-GA models.
• Utilized smoothed NDVI (sNDVI) data alongside precipitation and temperature records from 1982-2011.

Main Results:

• Alberta's sNDVI showed an increasing trend over 30 years, indicating a longer growing season.
• Northern Alberta's vegetation productivity is primarily temperature-driven, while southern Alberta's is precipitation-driven.
• The ANN-GA model demonstrated superior performance in predicting sNDVI compared to the linear regression model.

Conclusions:

• Climate change is influencing vegetation productivity in Alberta, with distinct regional drivers.
• Non-linear modeling approaches like ANN-GA are more effective for capturing complex vegetation-climate relationships.
• Findings provide valuable insights for ecological management and policy in Alberta.