A GENERIC REMOTE SENSING APPROACH FOR LARGE-SCALELAND COVER AND LAND USE SYSTEMS MAPPING

Fifth International Conference on Remote Sensing and Geoinformation of the Environment 2017, 20-23 March, 2017, Cyprus

A GENERIC REMOTE SENSING APPROACH FOR LARGE-SCALE LAND COVER AND LAND USE SYSTEMS MAPPING

Beatriz Bellón

*, Agnès Bégué

, Danny Lo Seen

, Valentine Lebourgeois

, Margareth Simões

a

CIRAD, UMR TETIS, Maison de la Télédétection, 500 rue Jean-François Breton 34090 Montpellier, France; (0033) 4 67 55 86 15; E-mail: [email protected], [email protected],

[email protected], [email protected]

b

Embrapa Solos, Rua Jardim Botânico, 1024. CEP 22460-000 Rio de Janeiro, Brazil; (0055) 21- 21794607; E-mail: [email protected]

KEY WORDS: Remote Sensing, agricultural systems, vegetation index, time series, MODIS, land use

ABSTRACT:

In response to the need of spatial information for supporting agricultural monitoring, we present a new remote sensing object-based approach for objective and repeatable large-scale agricultural systems mapping. This approach can be broken down into two steps: 1. A stratifi- cation of land units, based on a normalized difference vegetation index (NDVI) time series at regional level; 2. A semi-automatic land cover and land use classification, performed in each land unit at field level.

To produce the land units, a principal component transformation was first applied to an an- nual dataset of MODIS (MODerate Imaging Spectroradiometer) NDVI images. A series of segmentations were then performed on the principal component images that contain the es- sential information on the physiognomy and phenology of the vegetation cover. An unsuper- vised evaluation method was used to identify the optimum segmentation which successfully delineates homogeneous units in terms of land use. Then, for each land unit, land cover and land use classifications were carried out at field level through segmentation of a Landsat 8 high resolution mosaic image and unsupervised classification of the MODIS NDVI time se- ries. Finally, a Dynamic Time Warping clustering method was applied to combine the differ- ent classes by their temporal behaviour to produce a final land use systems classification which has been validated with in situ data.

A map of the main cropping systems of the Brazilian state of Tocantins, an agricultural ex- pansion region, has been successfully produced for the year 2015 following this approach.

This study shows the potential of unsupervised, repeatable remote sensing approaches to pro-

vide valuable large-scale baseline spatial information for supporting agricultural monitoring.