Monitoring the Evolution of Agricultural Growth in an Arid Area in
Northwest Mexico Using Free Tools in the Cloud
Chihuahua State, in northwestern Mexico, has had strong growth in its
irrigated agriculture, going from 2010 – 2018 of 461,099 to 597,222 ha.
Although the total area planted has decreased due to often droughts.
Main growth has occurred in the Chihuahuan desert area, through
groundwater extraction, causing aquifers’ overexploitation. This
unsustainable growth will provoke a collapse in the management of water
resources. This work shows a methodology to determine the current
agricultural frontier and monitor the evolution of the irrigated area.
Methodology employs the Google Earth Engine (GEE) platform to obtain and
process satellite images, and was applied in the Laguna de Hormigas
aquifer, located in the Chihuahuan desert area. For this aquifer, 10.55
times its recharge is extracted. Methodology uses Sentinel 2, Landsat 5,
and Landsat 8 satellite images. To identify the agricultural frontier,
we use Sentinel 2 images (level 1C) with cloud cover less than 10% from
2015-2020. The agricultural frontier is obtained by classifying two
bands, one with maximum vegetative development and another with maximum
humidity in each pixel. The bands are generated from a statistical
historical analysis of the normalized difference vegetation index (NDVI)
and the normalized difference water index (NDWI). To monitor the
evolution of irrigated areas from 2000 – 2020, Landsat 5 and Landsat 8
images of level 2 collection 2 are processed. The agricultural frontier
defines the area where growth occurs. From Landsat images, for each year
one image with maximum vegetative development is obtained which is
processed to show if it has cultivated areas or not. Considering that
cultivated areas have an NDVI value greater than 0.2. Results shows that
agricultural development in the Laguna de Hormigas aquifer starts in
2010, but from 2016 is accelerated. Results are consistent with the
statistics published in the Agrifood and Fisheries Information Service.
The methodology developed is useful to analyze the spatio-temporal
evolution of agricultural areas in arid regions and can be used in
similar zones, without special computational requirements because it
uses databases and cloud tools.