Open Data Cube for Natural Resources Mapping in Mexico

Abstract

In recent years, the efforts to enhance the analysis of Earth’s surface with satellite imagery have forced the scientific community to develop different techniques and methodologies. The Open Data Cube aims to provide tools to execute multi-temporal analysis and get accurate products, excluding low-quality pixels in small or large areas of study with an accuracy subject to the resolution of the data used for the analysis. This means that we can make use of the full potential of Earth observation data available from satellite data providers, in this document we take a closer look at Landsat Imagery and its applications. The beginning of the implementation of the Open Data Cube platform began in 2018, positioning itself as a valuable source of spatial data for Natural Resources projects in INEGI and seeks to support the decision-making process based on territorial analyzes with great certainty. The use of this technological solution represents a great leap between the traditional visual interpretation of raster data and the automated processing of data in time series. 1 Theoretical and technical basis 1.1 Background The Geographic Division in the National Institute of Statistics and Geography (INEGI) is the National Mapping Agency in Mexico. For more than 50 years it has been producing maps in multiples themes; from framework mapping (topographic maps, relief – DEMS, orthoimages, census area mapping, and others, as well as Natural Resources Maps on Geology, Water; ground and surface, Soils, Climate and Vegetation maps for all the country. Being a country that is still going through rapid changes in land use – land cover, there is the challenge of producing Vegetation, Soil and Water maps, among others in a more frequent way. Although all the Natural Resources mapping has always been based on Remote Sensing Imagery (aerial photographs and images from several sensors and satellites: SPOT, LANDSAT, etc), we need now to be able to manage, process and analyze massive amounts of Imagery data in an effective way, and combine the knowledge of experts with state of the art analysis methods for Remote Sensing Data. We expect the Open Geospatial Datacube to give us that capability. Kalpa Publications in Computing Volume 13, 2019, Pages 70–78 Proceedings of the 1st International Conference on Geospatial Information Sciences O. S. Siordia, J.L. Silván Cárdenas, A. Molina-Villegas, G. Hernandez, P. Lopez-Ramirez, R. Tapia-McClung, K. González Zuccolotto and M. Chirinos Colunga (eds.), iGISc 2019 (Kalpa Publications in Computing, vol. 13), pp. 70–78 1.2 Geospatial Data Cube Due to many factors that observations from satellite are subject, for example, clouds, topographic shading, cloud shadows, and instruments failure, it has been necessary to improve the quality of these observations taking into consideration the pixel saturation for each spectral band, band contiguity, slope, cloud or cloud shadow and terrain shadow. Another factor involved is the time extent, most of the satellite products for land-cover mapping are restricted on single dates, even though the accuracy of the observations can be affected by atmospheric phenomena, seasonal changes, anthropogenic interventions, etc. In order to give a reasonable response to these problems, diverse approaches and methods have been presented, being the multi-temporal classifications and proceedings the ones that show better performances. This is the basis of the Data Cube, in which it is possible to execute different algorithms, perform supervised and unsupervised classifications, extract training samples efficiently, identify and fill contaminated pixels (cloud, cloud shadow) with optimal pixels from the dataset archive, and consequently, provide accurate hypothesis in the following studies [1]. Fig. 1. Open Data Cube concept [1] Different satellite imagery datasets have been used to build the core of other data cubes, this depends on the objectives and purposes of the developers, in this case, the core dataset is the available archive of Landsat-4, Landsat-5, Landsat-7, and Landsat-8 from the year 1984 to the present (at September 1, 2019) with 113, 928 scenes that cover the 100 percent of the national territory. This archive is provided by the United States Geological Survey USGS with a pixel resolution of 30 meters. Mexico has a unique geographic distribution that made necessary to create a special grid with tiles of 150 x 150 kilometers of size, not only to fit the Landsat imagery accurately, but also to include in the future Sentinel imagery that provides better spatial resolution (10 meters per pixel), with this setting it is possible to display a time-series observations of every pixel and thus provide every observations for analysis [2]. Open Data Cube for Natural Resources Mapping in Mexico J. L. Ornelas De Anda et al.