Tendencia en el siglo XXI del Índice de Diferencias Normalizadas de Vegetación (NDVI) en la parte sur de la península de Baja California

Abstract

The use of remote sensing in scientific research has revolutionized our understanding by revealing the diversity and complexity of terrestrial systems, something difficult to appreciate in the past. One of the tools most extensively used in this field is the Normalized Difference Vegetation Index (NDVI), which is a system for the remote monitoring of vegetation in a synoptic temporal and spatial scale. In this particular case, we are interested in analyzing not only the average spatial or temporal behavior of the NDVI, but any trends it may display. It is a simple but nonetheless important method for the study of temporal series, since it allows to identify, in a general sense, a positive (increase) or negative (decrease) pattern of the elements in a particular environment during a given period.

This paper analyzes the average behavior and sustained trend of NDVI per pixel for the southern portion of the Baja California Peninsula, Mexico, during the period 2001-2015. MODIS images obtained from the public database Land Processes Distributed Active Archive Center (LPDAAC,https://lpdaac.usgs.gov/) of the National Aeronautics and Space Administration (NASA) were used. The information is in binary format with a 250-m spatial resolution (pixel unit), and a 16-day temporal resolution (MOD13Q1). NDVI data corresponding to the southern portion of the Baja California Peninsula are available in two tiles (h07v06 and H08v06); therefore, techniques for processing tiled images were applied to produce each average 16-day image and, subsequently, work on a monthly basis. MODIS products were cut and reprojected from the Universal Transverse Mercator (UTM) system to the WGS84 geographic coordinate system; MODIS data to obtain the average value and the trend of NDVI per pixel were processed using the programming software R through different libraries. A total of 1 377 985 pixels were analyzed, 67.04% of which showed average NDVI values between 0.1 and 0.3, consistent with arid zones. However, values > 0.5 were recorded in areas with a high vegetation cover in high altitudes, with peak values (> 0.8) within Sierra La Laguna Biosphere Reserve (RBSL), located to the south of the peninsula. The trend analysis revealed that in this same area within RBSL, the NDVI trend values are negative, indicating loss of vegetation vigor. Negative trend values matched closely the areas with the highest population density and tourism development around RBSL, indicating that the southern region of the peninsula is facing a strong pressure that is leading to alterations in the system. Its magnitude should have to be evaluated to determine if it causes structural changes only or if functional changes are also involved. Although the tourism sector has boosted economic growth in the southern portion of the Baja California peninsula, this activity has also involved the greatest alterations to the natural system, which is particularly evident in coastal subsystems. These tourism development and urbanization process, in addition to fragmentation, cause disturbance levels not yet measured in a synoptic scale. This first approximation sets the ground to continue searching for patterns of change that may be indicators of the condition of the systems analyzed. Tn turn, these could be used to establish useful indicators for decision-making, particularly with regard to the planning of tourist development initiatives. As discussed in this paper, the implementation of poorly designed projects may be the main stressors of ecological systems, which eventually may lead structural and functional changes in the system.