Environmental Remote Sensing Analysis

Monitoring land cover changes using Landsat imagery

This set of works applies optical remote sensing techniques to analyze three critical environmental phenomena in Argentina: deforestation in the northwest, fires in Patagonia, and snow/glacier dynamics in Cuyo. Each case uses specific spectral indices processed with ESA SNAP from Landsat imagery, following standardized methodologies in scientific literature.

Remote Sensing Landsat ESA SNAP Environmental Analysis NDVI
← Back to Main Portfolio
English Español

Project Summary

The following projects analyze different environmental phenomena using remote sensing techniques with Landsat imagery:

Project Location Period Index used Satellite Product
Deforestation Joaquín V. González, Salta 1986–2017 NDVI Landsat 5 / Landsat 8 Land cover change map
Fires Lago Cholila, Chubut January–April 2015 NBR / ΔNBR Landsat 8 Burn severity map
Glaciers/Snow Cerro de la Majadita, San Juan 2018 (seasonal) Snow Index (SI) Landsat 8 Seasonal dynamics map

1. Deforestation in northwest Argentina

The Chaco Salteño region has experienced significant transformation due to agricultural expansion. This analysis quantifies the loss of native forest around Joaquín V. González between 1986 and 2017.

Methodology

  • Download of Landsat 5 (1986) and Landsat 8 (2017) images, path 230, row 77.
  • Calculation of the Normalized Difference Vegetation Index (NDVI) for each date:

NDVI Formula: NDVI = (NIR - Red) / (NIR + Red)

  • Unsupervised classification (k-means, 3 classes) of stacked NDVI.
  • Category assignment:
    • Green: Preserved vegetation (high NDVI without change).
    • Orange: Areas with decreased NDVI due to forest loss.
    • Yellow: Low vegetation zones without significant changes.
Deforestation map in Salta (1986-2017)

Deforestation map in Salta (1986-2017) - NDVI change analysis

Result

The map shows a clear pattern of reduction in green areas, corresponding to the expanding agricultural frontier. The technique allows quantification of the transformed area and localization of deforestation hotspots.

2. Fires in Patagonia

In 2015, a large-scale fire affected the Lago Cholila region (Chubut). This work evaluates damage severity using the Normalized Burn Ratio.

Methodology

  • Landsat 8 images before (21/01/2015) and after (11/04/2015) the fire, path 232, row 89.
  • Calculation of the Normalized Burn Ratio (NBR) for each scene:

NBR Formula: NBR = (NIR - SWIR2) / (NIR + SWIR2)

  • Obtaining the temporal difference (ΔNBR):

ΔNBR Formula: ΔNBR = NBRpre - NBRpost

  • Threshold classification into seven severity categories, from "vegetation regrowth" to "high fire severity".
Fire severity map in Lago Cholila (2015)

Fire severity map in Lago Cholila (2015) - ΔNBR analysis

Result

The fire core zone shows high ΔNBR values (red tones), indicating moderate-high severity. Peripheral areas show lower severity, while some green sectors suggest post-fire regrowth.

3. Seasonal snow and glacier dynamics in Cuyo

The Andean cryosphere is a strategic water reservoir. This analysis identifies permanent and seasonal snow cover in Cerro de la Majadita (San Juan) during 2018.

Methodology

  • Landsat 8 images from summer (04/01/2018) and winter (15/07/2018), path 233, row 81.
  • Calculation of the Snow Index (SI) for each date:

Snow Index Formula: SI = Red / SWIR1

  • Determination of a spectral threshold to discriminate snow-covered areas.
  • Binary combination of summer and winter layers to obtain four classes:
    • No snow
    • Snow only in summer
    • Snow only in winter
    • Permanent snow (year-round)
Seasonal snow/ice dynamics map in San Juan (2018)

Seasonal snow/ice dynamics map in San Juan (2018) - Snow Index analysis

Result

Permanent snow/ice bodies (dark blue) are clearly identified, crucial for seasonal water supply. The distribution of seasonal snow reflects the altitudinal gradient and slope exposure.

Highlighted technical aspects

Software

Processing performed with ESA SNAP, standard tool in optical remote sensing.

Data

Landsat Collection 2 Level 2 images, with atmospheric corrections already applied.

Spectral indices

Appropriate selection according to the studied phenomenon (NDVI, NBR, SI).

Temporal analysis

Multitemporal comparison to capture change dynamics.

Final reflection

These works demonstrate the capability of remote sensing to monitor environmental changes at regional scale with quantitative rigor. The selection of appropriate indices, consistent processing in SNAP, and contextual interpretation of results allow transformation of satellite data into useful information for territorial and environmental management.