Outline Problem R vs GIS Methods Further works Using R for time series analysis and spatial-temporal distribution of global burnt surface multi-year product Jedrzej Bojanowski* C´ esar Carmona-Moreno European Commission - Joint Research Centre Institute for Environment and Sustainability Global Environment Monitoring Unit, Ispra, Italy * Institute of Geodesy and Cartography, Remote Sensing Department, Warsaw, Poland The R User Conference 2008, August 12-14 Technische Universitt Dortmund, Germany Jedrzej Bojanowski*, C´ esar Carmona-Moreno EC Joint Research Centre Using R for time series analysis and spatial-temporal distribution of global burnt surface multi-year product
Outline Problem R vs GIS Methods Further works Problem R vs GIS Methods Further works Jedrzej Bojanowski*, C´ esar Carmona-Moreno EC Joint Research Centre Using R for time series analysis and spatial-temporal distribution of global burnt surface multi-year product
Outline Problem R vs GIS Methods Further works The need for global burnt area product Fires: a significant component of global ecosystem Influence on climate, carbon cycle, pollution... Climate change? Jedrzej Bojanowski*, C´ esar Carmona-Moreno EC Joint Research Centre Using R for time series analysis and spatial-temporal distribution of global burnt surface multi-year product
Outline Problem R vs GIS Methods Further works The need for global burnt area product Fires: a significant component of global ecosystem Influence on climate, carbon cycle, pollution... Climate change? PROBLEM Lack of an exhaustive base of past fires activities! Jedrzej Bojanowski*, C´ esar Carmona-Moreno EC Joint Research Centre Using R for time series analysis and spatial-temporal distribution of global burnt surface multi-year product
Outline Problem R vs GIS Methods Further works The need for global burnt area product Fires: a significant component of global ecosystem Influence on climate, carbon cycle, pollution... Climate change? PROBLEM Lack of an exhaustive base of past fires activities! TO DO Concatenation of two existing databases: GBS and L 3 JRC Jedrzej Bojanowski*, C´ esar Carmona-Moreno EC Joint Research Centre Using R for time series analysis and spatial-temporal distribution of global burnt surface multi-year product
Outline Problem R vs GIS Methods Further works GBS and L 3 JRC GBS L 3 JRC time range 1982–1999 2000–2007 input data NOAA/AVHRR SPOT VEGETATION temporal resolution 1 week 1 day spatial resolution approx 8 km approx 1 km advantages seasonality! area estimates! Jedrzej Bojanowski*, C´ esar Carmona-Moreno EC Joint Research Centre Using R for time series analysis and spatial-temporal distribution of global burnt surface multi-year product
Outline Problem R vs GIS Methods Further works Why R and not GIS? Wide functionality Import of all data formats Easy data manipulation Statistical and geostatistical analysis Graph plotting Map plotting Results into LaTeX code Jedrzej Bojanowski*, C´ esar Carmona-Moreno EC Joint Research Centre Using R for time series analysis and spatial-temporal distribution of global burnt surface multi-year product
Outline Problem R vs GIS Methods Further works Why R and not GIS? Wide functionality Import of all data formats Easy data manipulation Statistical and geostatistical analysis Graph plotting Map plotting Results into LaTeX code AUTOMATION! Jedrzej Bojanowski*, C´ esar Carmona-Moreno EC Joint Research Centre Using R for time series analysis and spatial-temporal distribution of global burnt surface multi-year product
Outline Problem R vs GIS Methods Further works Methods ◮ Data import ◮ Data manipulation ◮ Time series analysis ◮ Regression modeling ◮ Principal components analysis & 3D visualization ◮ Spatial temporal distribution visualization technique Jedrzej Bojanowski*, C´ esar Carmona-Moreno EC Joint Research Centre Using R for time series analysis and spatial-temporal distribution of global burnt surface multi-year product
Outline Problem R vs GIS Methods Further works Data import and storage GeoTiff rgdal rNetCDF data manipulation NetCDF Analyses GeoTiff Vizualization Jedrzej Bojanowski*, C´ esar Carmona-Moreno EC Joint Research Centre Using R for time series analysis and spatial-temporal distribution of global burnt surface multi-year product
Outline Problem R vs GIS Methods Further works Generalization GBS L3JRC 8 km 1 km weekly daily GENERALIZATION 0.5 degree monthly Jedrzej Bojanowski*, C´ esar Carmona-Moreno EC Joint Research Centre Using R for time series analysis and spatial-temporal distribution of global burnt surface multi-year product
Outline Problem R vs GIS Methods Further works Time series Jedrzej Bojanowski*, C´ esar Carmona-Moreno EC Joint Research Centre Using R for time series analysis and spatial-temporal distribution of global burnt surface multi-year product
Outline Problem R vs GIS Methods Further works Seasonality shift Time series for Kazahstan 30000 Burnt area in [ km 2 ] 20000 10000 0 1985 1990 1995 2000 2005 Time Area burnt in months in GBS Area burnt in months in L3JRC 150000 150000 Burnt area in [ km 2 ] 50000 50000 1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 6 7 8 9 10 11 12 Months Months Jedrzej Bojanowski*, C´ esar Carmona-Moreno EC Joint Research Centre Using R for time series analysis and spatial-temporal distribution of global burnt surface multi-year product
Outline Problem R vs GIS Methods Further works Area estimation 90 1982−1999 (GBS) 2000−2006 (L3JRC) 60 30 latitude 0 −30 −60 −90 0 1000 2000 3000 4000 Yearly mean burnt area [ km 2 ] Jedrzej Bojanowski*, C´ esar Carmona-Moreno EC Joint Research Centre Using R for time series analysis and spatial-temporal distribution of global burnt surface multi-year product
Outline Problem R vs GIS Methods Further works Probability map Jedrzej Bojanowski*, C´ esar Carmona-Moreno EC Joint Research Centre Using R for time series analysis and spatial-temporal distribution of global burnt surface multi-year product
Outline Problem R vs GIS Methods Further works Probability extension algorithm Jedrzej Bojanowski*, C´ esar Carmona-Moreno EC Joint Research Centre Using R for time series analysis and spatial-temporal distribution of global burnt surface multi-year product
Outline Problem R vs GIS Methods Further works Probability extension algorithm Jedrzej Bojanowski*, C´ esar Carmona-Moreno EC Joint Research Centre Using R for time series analysis and spatial-temporal distribution of global burnt surface multi-year product
Outline Problem R vs GIS Methods Further works Probability extension algorithm Jedrzej Bojanowski*, C´ esar Carmona-Moreno EC Joint Research Centre Using R for time series analysis and spatial-temporal distribution of global burnt surface multi-year product
Outline Problem R vs GIS Methods Further works Probability extension algorithm Jedrzej Bojanowski*, C´ esar Carmona-Moreno EC Joint Research Centre Using R for time series analysis and spatial-temporal distribution of global burnt surface multi-year product
Outline Problem R vs GIS Methods Further works Probability extension algorithm Jedrzej Bojanowski*, C´ esar Carmona-Moreno EC Joint Research Centre Using R for time series analysis and spatial-temporal distribution of global burnt surface multi-year product
Outline Problem R vs GIS Methods Further works Probability extension algorithm Jedrzej Bojanowski*, C´ esar Carmona-Moreno EC Joint Research Centre Using R for time series analysis and spatial-temporal distribution of global burnt surface multi-year product
Outline Problem R vs GIS Methods Further works Probability extension algorithm Jedrzej Bojanowski*, C´ esar Carmona-Moreno EC Joint Research Centre Using R for time series analysis and spatial-temporal distribution of global burnt surface multi-year product
Outline Problem R vs GIS Methods Further works Probability map after extension Jedrzej Bojanowski*, C´ esar Carmona-Moreno EC Joint Research Centre Using R for time series analysis and spatial-temporal distribution of global burnt surface multi-year product
Outline Problem R vs GIS Methods Further works Principal Components & 3D interactive visualization Jedrzej Bojanowski*, C´ esar Carmona-Moreno EC Joint Research Centre Using R for time series analysis and spatial-temporal distribution of global burnt surface multi-year product
Outline Problem R vs GIS Methods Further works Principal Components & 3D interactive visualization Jedrzej Bojanowski*, C´ esar Carmona-Moreno EC Joint Research Centre Using R for time series analysis and spatial-temporal distribution of global burnt surface multi-year product
Outline Problem R vs GIS Methods Further works Spatial-temporal distribution Jedrzej Bojanowski*, C´ esar Carmona-Moreno EC Joint Research Centre Using R for time series analysis and spatial-temporal distribution of global burnt surface multi-year product
Recommend
More recommend
