Format

Send to

Choose Destination
Int J Biometeorol. 2016 Sep;60(9):1389-403. doi: 10.1007/s00484-015-1132-5. Epub 2016 Jan 14.

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

Author information

1
State Key Laboratory Base of Eco-hydraulic Engineering in Arid Area, Xi'an University of Technology, Xi'an, 710048, China. jrengui@163.com.
2
State Key Laboratory Base of Eco-hydraulic Engineering in Arid Area, Xi'an University of Technology, Xi'an, 710048, China. jcxie@mail.xaut.edu.cn.
3
College of Natural Resources and Environment, Northwest A&F University, Yangling, 712100, China. hailong.he@hotmail.com.
4
Department of Civil & Environmental Engineering, University of Alberta, Edmonton, T6G 1H9, Canada.
5
State Key Laboratory Base of Eco-hydraulic Engineering in Arid Area, Xi'an University of Technology, Xi'an, 710048, China.
6
State Key Laboratory of Simulation and Regulation of Water Cycle in River Basin, China Institute of Water Resources and Hydropower Research, Beijing, 100038, China.

Abstract

As one of the most popular vegetation indices to monitor terrestrial vegetation productivity, Normalized Difference Vegetation Index (NDVI) has been widely used to study the plant growth and vegetation productivity around the world, especially the dynamic response of vegetation to climate change in terms of precipitation and temperature. Alberta is the most important agricultural and forestry province and with the best climatic observation systems in Canada. However, few studies pertaining to climate change and vegetation productivity are found. The objectives of this paper therefore were to better understand impacts of climate change on vegetation productivity in Alberta using the NDVI and provide reference for policy makers and stakeholders. We investigated the following: (1) the variations of Alberta's smoothed NDVI (sNDVI, eliminated noise compared to NDVI) and two climatic variables (precipitation and temperature) using non-parametric Mann-Kendall monotonic test and Thiel-Sen's slope; (2) the relationships between sNDVI and climatic variables, and the potential predictability of sNDVI using climatic variables as predictors based on two predicted models; and (3) the use of a linear regression model and an artificial neural network calibrated by the genetic algorithm (ANN-GA) to estimate Alberta's sNDVI using precipitation and temperature as predictors. The results showed that (1) the monthly sNDVI has increased during the past 30 years and a lengthened growing season was detected; (2) vegetation productivity in northern Alberta was mainly temperature driven and the vegetation in southern Alberta was predominantly precipitation driven for the period of 1982-2011; and (3) better performances of the sNDVI-climate relationships were obtained by nonlinear model (ANN-GA) than using linear (regression) model. Similar results detected in both monthly and summer sNDVI prediction using climatic variables as predictors revealed the applicability of two models for different period of year ecologists might focus on.

KEYWORDS:

Predictive model; Smoothed NDVI (sNDVI); Variability; sNDVI-climate relationships

PMID:
26768143
DOI:
10.1007/s00484-015-1132-5
[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Springer
Loading ...
Support Center