Trends in 15-year MODIS NDVI time series for Mexico

The number of studies exploring long data records have increased in recent years and trend analysis has become a frequently-used analysis approach. This study tested parametric and non-parametric regression methods for trend estimation in MODIS NDVI data. The full record of 15 years of MODIS data from the Terra satellite was filtered to exclude clouds, shadow or other data of inferior quality from this analysis. Yearly averages and dry and wet season means were analyzed using linear least-squares and Theil-Sen regression methods. Corresponding statistical tests (F-test and Mann-Kendall test) indicated the significance of each regression model. Consistent spatial patterns were found in MODIS NDVI trend analysis which match with local knowledge and studies exploring socioeconomic, environmental and demographic factors of vegetation and land cover change in Mexico. Significant (p<;0.05) positive trends in NDVI were found in the states of Chihuahua, northern Durango and Nuevo Leon due to increased woody coverage or Oaxaca due to vegetation densification. Negative trends occur in Sonora and Coahuila due to climate variability and around larger cities. Seasonal trend analysis helped interpreting and discerning anthropogenic from natural drivers.

[1]  Rasim Latifovic,et al.  Landsat-7 ETM+ radiometric normalization comparison for northern mapping applications , 2005 .

[2]  T. Mitchell Aide,et al.  Land change for all municipalities in Latin America and the Caribbean assessed from 250-m MODIS imagery (2001–2010) , 2012 .

[3]  Hans von Storch,et al.  Misuses of Statistical Analysis in Climate Research , 1995 .

[4]  Per Jönsson,et al.  Seasonality extraction by function fitting to time-series of satellite sensor data , 2002, IEEE Trans. Geosci. Remote. Sens..

[5]  Rasim Latifovic,et al.  Monitoring Cumulative Long-Term Vegetation Changes Over the Athabasca Oil Sands Region , 2014, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[6]  Zhiqiang Yang,et al.  Detecting trends in forest disturbance and recovery using yearly Landsat time series: 1. LandTrendr — Temporal segmentation algorithms , 2010 .

[7]  Christopher Conrad,et al.  TiSeG: A Flexible Software Tool for Time-Series Generation of MODIS Data Utilizing the Quality Assessment Science Data Set , 2008, IEEE Transactions on Geoscience and Remote Sensing.

[8]  René R. Colditz,et al.  Detecting Change Areas in Mexico Between 2005 and 2010 Using 250 m MODIS Images , 2014, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[9]  M. Clark,et al.  The influence of socioeconomic, environmental, and demographic factors on municipality-scale land-cover change in Mexico , 2012, Regional Environmental Change.

[10]  R. Fernandes,et al.  Parametric (modified least squares) and non-parametric (Theil–Sen) linear regressions for predicting biophysical parameters in the presence of measurement errors , 2005 .

[11]  Rob J Hyndman,et al.  Detecting trend and seasonal changes in satellite image time series , 2010 .

[12]  C. Justice,et al.  High-Resolution Global Maps of 21st-Century Forest Cover Change , 2013, Science.

[13]  M. Clark,et al.  Vegetation change and land tenure in Mexico: A country-wide analysis , 2013 .

[14]  M. Clark,et al.  Deforestation and Reforestation of Latin America and the Caribbean (2001–2010) , 2013 .

[15]  Sheng Yue,et al.  The influence of autocorrelation on the ability to detect trend in hydrological series , 2002 .

[16]  Sheng Yue,et al.  Applicability of prewhitening to eliminate the influence of serial correlation on the Mann‐Kendall test , 2002 .

[17]  C. Woodcock,et al.  Continuous monitoring of forest disturbance using all available Landsat imagery , 2012 .