EN中文

Journal of Geographical Sciences >

2017 , Vol. 27 >Issue 4: 420 - 438

DOI: https://doi.org/10.1007/s11442-017-1385-5

Research Articles

Assessing the progress of desertification of the southern edge of Chihuahuan Desert:A case study of San Luis Potosi Plateau

  • NOYOLA-MEDRANO Cristina , 1 ,
  • MARTÍNEZ-SÍAS Valeria Abigaíl 2
Expand
  • 1. Facultad de Ingeniería / Area Civil / Geoinformática /Universidad Autónoma de San Luis Potosí, Manuel Nava # 8, Zona Universitaria Poniente, 78290, San Luis Potosí, SLP, Mexico
  • 2. Colegio de Posgraduados / Campus San Luis Potosí. Iturbide 73, Salinas de Hidalgo, SLP, Mexico

Author: Cristina Noyola-Medrano (1970-), specialized in natural resource and environmental remote sensing. E-mail:

Received date: 2016-03-18

  Accepted date: 2016-05-17

  Online published: 2017-04-20

Supported by

Program for Professional Development of Professors (PRODEP) through the convention PROMEP/103.5/ 13/6575

Copyright

Journal of Geographical Sciences, All Rights Reserved
Fold

Abstract

The aim of this study is to establish if the San Luis Potosi Plateau (SLPP), which is part of the southern edge of the Chihuahuan Desert, is generating desertification processes, indicating a progression of the desert toward the central part of Mexico. Therefore, we analyzed the temporal evolution of four environmental indicators of desertification: Normalized Difference Vegetation Index (NDVI), Normalized Difference Water Index (NDWI), Iron Oxides Index (IO) and Surface Temperature (ST). Landsat TM images are used to cover a period from 1990 to 2011. A new equation of total balance is proposed to generate an image of the overall evolution of each factor which is applied to get a probability map of desertification. The evolution of NDVI, NDWI and IO shows a behavior almost stable over the time. In contrast, the ST shows a slight increase. The outcomes of this study confirm periods of vegetation re-greening and 8.80% of the SLPP has the highest probability to develop desertification. The most affected area is the portion west of the region, and the east and south are the least affected areas. The results suggest a slight advance of the desert, although most of the area doesn’t have the necessary conditions to develop desertification.

Key words： Landsat; NDVI-NDWI; ST-IO; desertification; re-greening vegetation

Cite this article

NOYOLA-MEDRANO Cristina , MARTÍNEZ-SÍAS Valeria Abigaíl . Assessing the progress of desertification of the southern edge of Chihuahuan Desert:A case study of San Luis Potosi Plateau[J]. Journal of Geographical Sciences, 2017 , 27(4) : 420 -438 . DOI: 10.1007/s11442-017-1385-5

1 Introduction

Argue that a region is developing desertification processes has a significant impact on the planning of land management policies. For that reason, it’s important to analyze if the region is showing signs of desertification. The clear identification of the causes and processes of desertification can be hindered by the interaction between natural and anthropic phenomena and, by confusion between the environmental cyclical oscillations of short term, with those representing long-term degradation (Xu et al., 2011; Granados-Sánchez et al., 2013).
The soil degradation occurs in whatever portion of the earth’s surface, but it is fixed as “desertification” when it occurs in arid, semi-arid and sub-humid dry areas, derived from the human impact (Helldén, 1992; Reynolds et al., 2007). Desertification is not just a problem of soil loss, but it touches on the stability and development of human populations. Moreover, the effects of desertification affect not only the source but also very remote spots from the inception of the phenomenon (Tao, 2004).
The United Nations (UN) contends that desertification threatens the livelihoods of a billion people in over 100 countries spread over one-fourth of the earth’s land surface. Africa is the most damaged continent by desertification, followed by Latin America, the Caribbean and Southern Europe (Ezcurra, 2006). In Mexico, the arid and semi-arid lands are chiefly in the deserts of Sonora, Chihuahua and the central regions influenced by the rain shadow effect caused by the Sierra Madre Occidental and Oriental (Pando-Moreno et al., 2002). Previous studies have reported 1,240,000 km2 of arid and semi-arid regions, which is equivalent to approximately 60% of the entire territory of Mexico (Díaz-Padilla et al., 2011). Despite the problem posed by desertification mechanisms, De Waroux and Lambin (2012) explain that the processes of degradation of arid and semi-arid lands have received little attention.
Based on this scenario, it is advisable to set up mechanisms to help evaluate whether the arid and semi-arid soils of Mexico have a mechanism of irreversible desertification. For this reason, it is equally important to determine the concept of cyclical swing vegetation which is largely linked to climatic fluctuations. The first work report of cyclical swing took place in the Sahara between 1980 and 1990 (Tucker et al., 1991). The results of this work show that there was an increase in the vegetation when it increased precipitation and that this period followed a prolonged dry spell between 1980 and 1984. Other studies were supported by the analysis of photographs, field work and implementation normalized index (NDVI), to demonstrate the reversibility of the coating degradation vegetation (Helldén, 1992).
Because of the importance of determining if a region is under a phenomenon of desertification or cyclic oscillation, several works supported by Geoinformatics techniques have been generated (Santini et al., 2010; Gamo et al., 2013). These studies include the role of satellite images and the implementation of Geographic Information Systems (GIS). In this way, Collado and others (2002) examined the phenomenon of desertification for the Province of San Luis (Central Argentina). They have applied the techniques as Spectral Mixture Analysis (SMA) and analysis of change from Landsat images. They concluded the area showed severe signs of degradation, although, in the zone there is vegetation recovering. Haboudane and others (2002) applied the analysis of indices generated from the spectrum shape to record that the optical properties of the soil surface can be utilized as indicators of degradation in semi-arid environments. Also, they set indices which integrate short-wave infrared bands to detect pedogenetic indicators that serve to characterize the degradation of soils with sparse vegetation. Another technique for detecting desertification applies standardized values of NDVI and precipitation (z-score) from NOAA AVHRR images. The outcomes of this survey concluded that the behavior of the vegetation seems to be at odds with the common view that the desertified areas are increasing worldwide (Helldén and Tottrup, 2008). However, these results have the problem of spatial resolution, which can mask the behavior-ground vegetation locally. Some sources have reported about the potential of using high-resolution satellite data such as Landsat TM to evaluate the desertification process and reversibility in degradation (Helldén and Tottrup, 2008; Ouedraogo et al., 2014).
Several authors have treated the evaluation of desertification in Mexico, and each point indicates different factors that contribute to desertification processes, particularly in the northern and central part of Mexico. Among the studies produced along this topic, one of them has reported moisturizing as a critical ingredient for the assessment of desertification (Huber-Sannwald et al., 2006). The oxidation, resulting from desert varnish, is another important feature for assessing land degradation because some elements such as wind abrasion and human disturbance can completely remove the varnish, or disrupt its appearance greatly (Dorn, 1986). On the other hand, Becerril-Piña and others (2015) analyzed with Landsat imagery the phenomenon of desertification for the Central Mexican Plateau; pointing out that one of the most significant factors is the vegetation-soil relationship because it permits to show that desertification is closely linked to deforestation and land use change. Nevertheless, a study released in the Sonoran Desert, highlights the importance of the relationship among albedo, surface temperature, and NDVI, because these factors present a cyclical pattern, and it is observed an increase in apparent desertification surfaces in years with longer periods of radiation (Tereshchenko et al., 2012).
Considering the above, the aim of this paper is to identify if the SLPP, in the southern edge of the Chihuahuan Desert, is developing processes of desertification. For this purpose, it has been analyzed and integrated into a balance equation, four parameters considered as potential indicators of desertification processes: Vegetation Index (NDVI), Moisture Index (NDWI), Iron Oxides Index (IO) and Surface Temperature (ST). All the data to implement the indices is extracted directly from Landsat TM images and supplemented with information on rainfall and temperature from weather stations.

2 Materials and methods

2.1 Study area

SLPP is located in the Province of the Central Plateau, in the southern edge of the Chihuahuan Desert in Mexico. The area studied is located between 22°20'-24°30'N and 100°25'- 102°18'W (Figure 1) with a total area of 29,100 km² (INEGI, 2012). Altitude values in this region vary from 1570 to 2680 m a.s.l.
Figure 1 Location map of the region surveyed
The climate of the region is classified as dry steppe (BS) in the Köppen system, with the variants BS0 (the driest) and BS1 (the least dry). Precipitation is scarce and irregular, with an annual average from 240 to 770 mm and, the average annual temperature varies from 13.5°C to 20.5°C (INEGI, 2012). Common plant communities in the study area are the microphyllus desert scrub, rosetophilous scrub, crasicaule scrub, and thorny scrub patches (Hernández et al., 2001). Between urban and suburban vegetation, there are native species (Prosopis laevigata, Opuntia spp., Agave lechuguilla, Yuca filifera, Sabal mexicana and Quercus spp.), and introduced species (Eucalyptus camaldulensis, Casuarina equisetifolia, Ficus spp., Populus fremontii, Salix babylonica, Pinus spp., Schinus molle and Ligustrum japonicum), as well as vegetation related to irrigation and rain-fed crop areas (Galindo and García, 1986).

2.2 Methodology

The methodology to extract the areas with the greatest possibility of developing desertification is based on the analysis of the spatial-temporal variability of four factors: Normalized Difference Vegetation Index (NDVI), Normalized Difference Water Index (NDWI), Iron Oxides Index (IO) and Surface Temperature (ST) (Figure 2).
Figure 2 Methodology implemented in this work
2.2.1 Obtaining satellite images
For this work, Landsat TM5 images for the years 1990, 1995, 2000, 2005 and 2011 were used (Table 1). All the images were selected for the dry season (January to April). The images have been obtained by the GLOVIS server, managed by the USGS (United States Geological Survey). ENVI 4.7 is used for the treatment of images, and the final integration of the maps is made in ArcGIS 10.2.
Table 1 List of images used in this study
IMAGE DATE IMAGE DATE IMAGE DATE
LT5028043 03-16-1990 LT5028044 03-16-1990 LT5029044 03-07-1990
LT5028043 04-15-1995 LT5028044 04-15-1995 LT5029044 04-22-1995
LT5028043 03-11-2000 LT5028044 03-11-2000 LT5029044 04-19-2000
LT5028043 03-09-2005 LT5028044 03-09-2005 LT5029044 02-28-2005
LT5028043 04-27-2011 LT5028044 04-27-2011 LT5029044 04-02-2011
2.2.2 Radiometric correction
The first step consists of converting the digital number values to radiance values for all bands, including the band 6 (thermal). For this purpose, it uses ENVI 4.7 software, which applies the equation suggested by Chander and others (2009) (Equation 1):
${{L}_{\lambda }}=\text{ }Grescale\times Qcal+Brescale$ (1)
where Lλ is spectral radiance at the sensor’s aperture [W/(m2·sr·μm)], Grescale is band-specific rescaling gain factor [(W/(m2 sr μm))/DN], and Brescale is band-specific rescaling bias factor [W/(m2 sr μm)]
The computation of the reflectance is generated for the bands 1, 2, 3, 4, 5, and 7. Likewise, for this process, it uses ENVI 4.7 based on the equation suggested by Chander and others (2009) (Equation 2):
$\rho \lambda =\frac{\pi {{L}_{\lambda }}{{d}^{2}}}{ESU{{N}_{\lambda }}\sin \theta }$ (2)
where ρλ is radiance in units of W/(m2·sr·µm), d is earth-sun distance in astronomical units, ESUNλ is solar irradiance in units of W/(m2·µm), and θ is sun elevation in degrees.
2.2.3 Surface temperature (ST)
To calculate the surface temperature (ST), the values of the radiance of band 6 were used to work the algorithm of Emissivity Normalization of ENVI 4.7. This algorithm generates values of ST in Kelvin (K). Immediately, K values were converted into Celsius Degrees (°C). The algorithm for the ST (Eq. 3) is based on the proposal of Chander and others (2009).
$ST=\frac{{{K}_{2}}}{ln\left( {{K}_{1}}/{{L}_{\lambda }}+1 \right)}$ (3)
where ST is brightness temperature of sensor (Kelvin), K2 is calibration constant 2 in Kelvin (1260.56), K1 is calibration constant 1 in [W / (m2 sr μm)] (607.76), ln is natural logarithm.
2.2.4 Normalized Difference Vegetation Index (NDVI) calculation
NDVI permits to find out if the vegetation is healthy, with some grade of stressor simply there is no vegetation cover (Huete et al., 2002). The algorithm of the vegetation index is generated based on the values of reflectivity. This index works with data located in the wavelengths from 0.66 μm (Red) to 0.86 μm (NIR), because in the red region, it is observed a strong absorption by the presence of pigments such as chlorophyll, whereas the NIR region presents high reflectances related to the structure of the plant.In order to obtain the NDVI index, it uses Equation 4:
$NDVI=\frac{{{\rho }_{NIR}}-{{\rho }_{RED}}}{{{\rho }_{NIR}}+{{\rho }_{RED}}}$ (4)
where ρNIR is 0.77-0.90 µm, which corresponds to band 4 of Landsat TM, and ρRED is 0.63-0.69 µm, which corresponds to band 3 of Landsat TM
The results of calculations of NDVI for a given pixel have always ranged from -1 to +1. Zero means no vegetation and the values close to +1 (0.8-0.9) indicate the presence of green and abundant foliage.
2.2.5 Iron Oxides Index (IO)
This band ratio highlights hydrothermally altered rocks or weathering soils that have been subjected to oxidation of iron (Drury, 1987). In this study, the IO is applied (Eq.5) because in the semi-desert areas it is common that soils developed desert varnish, which it can be detected with high values of IO. This rock rust or desert patina forms only on physically stable soil surfaces that are no longer subject to frequent precipitation, fracturing, wind abrasion or some other removal process (Spatz et al., 1989; Ustin et al., 2009). Consequently, zones with low values of IO are considered areas with soils degraded.
$OI=\frac{{{\rho }_{RED}}}{{{\rho }_{BLUE}}}$ (5)
where OI is iron oxide index, ρRED is 0.63-0.69 µm, corresponds to band 3 of Landsat TM, and ρBLUE is 0.45-0.52 µm, which corresponds to band 1 of Landsat TM
2.2.6 Normalized Difference Water Index (NDWI)
The moisture content is an indicator of the relationship between precipitation and potential evapotranspiration as a manifestation of water demand exerted by the surroundings. According to Gao (1996) NDWI is defined as:
$NDWI=\frac{{{\rho }_{\text{NIR}}}-{{\rho }_{\text{SWIR}}}}{{{\rho }_{\text{NIR}}}+{{\rho }_{\text{SWIR}}}}$ (6)
where ρNIR is 0.77-0.90 µm (Band 4 of Landsat TM), and ρSWIR is 1.56-1.78 µm (Band 5 of Landsat TM)
This band ratio is capable of detecting moisture because, in wetlands, the NIR region shows the maximum reflectivity and the SWIR zone shows the maximum absorption (Gao, 1996).
2.2.7 Temporal differences
From every factor obtained, there have been created maps of change to know temporary differences. To get this product, a simple subtraction is executed between a pair of images from different dates (Equation 7).
$Di{{f}_{xf}}={{f}_{t1}}-{{f}_{t2}}$ (7)
where Difxf is differences of x factor (NDVI, NDWI, ST, IO), ft1 is value of x factor in time 1 (oldest date), and ft2 is value of x factor in time 2 (newest date).
At this stage, four images were obtained for every factor, called as difference images. In all of the cases, the analysis is made pixel-to-pixel, and it is considered a positive result when there is a decrease in value between the date 1 (oldest date) and the date 2 (newest date). Otherwise, a negative result meant that there is an increase in value between the two epochs.
2.2.8 Computing balance equation
It is common to analyze temporal changes based on comparisons of two different dates to estimate the rate of change. In this way, for analysis that takes more than two periods, it is necessary to obtain several rates of change for every pair of images. This approach can only take a partial survey of the phenomenon. Nevertheless, to get a perspective more balanced changes generated in the region, this work proposes a new balance equation from four images of changes of every factor obtaining in the previous step (Equation 8).
${{B}_{xf}}=\sum\limits_{i=1}^{n}{\left( {{f}_{t1}}-{{f}_{t2}} \right)}$ (8)
where Bxf is image of final balance of “x” factor (NDVI, NDWI, ST, IO), and $\left( {{f}_{t1}}-{{f}_{t2}} \right)$ denotes difference of “x” factor for each couple of dates analyzed.
Positive or negative values were obtained for every pixel. In the case of NDVI, NDWI, and IO factors, a positive value indicates a greater risk of degradation of soils, because it involves loss of vegetation, moisture or desert varnish. In contrast, for ST, a negative value indicates an increase in temperature, which would raise the risk of development of desertification processes.
2.2.9 Desertification vulnerability map
The analysis of the probability to desertification was performed using the images of the final balance of every factor to develop a classification by a decision tree. For this classification, the NDVI is established as a most important factor to explain the desertification, followed by NDWI, ST, and OI (Table 2). The level of significance is set based on previous works on desertification (Pando- Moreno et al., 2002; Oyama and Nobre, 2004; Ustin et al., 2009).
Table 2 Thresholds considered establishing the map of probability to desertification
Factor Threshold
NDVI > 0.3
NDWI > 0.3
ST <-3°C
IO >0.5
2.2.10 Validation of the map of desertification probability
The verification of the final map of desertification probabilities was done based on accurate observations on 20 sites visited between the months of March and April of 2011.
The observation schedule was from 8 a.m. to 12 p.m. of a day. The covering of green vegetation, desiccation cracks, and desert varnish layer was obtained using a proportional value from the total area of a square of 1 m * 1 m. The surface temperature was measured with a thermometer, and the value of the amount of moisture was estimated from the parameters established by USDA-NRCS (1998). The level of probability of desertification was established according to the criteria of Table 3.
Table 3 Parameters used to perform field verification
Green
vegetation (%)		 Moisture (%) Surface
temperature (°C)		 Desert
varnish (%)		 Desiccation
cracks (%)		 Desertification
probability
>80 >40 <20 >50 0 Very low
60-80 30-40 20-22.5 40-50 0 Low
40-60 20-30 22.5-25 30-40 0-10 Moderate
20-40 10-20 25-27.5 20-30 10-20 High
<20 <10 >27.5 <20 >20 Very high

3 Results and discussion

3.1 Observed changes in NDVI, ST, IO and NDWI

This segment indicates the changes noted for every factor among the different dates analyzed. In the case of the NDVI, it is remarkable the great decrease between 2000 and 2005 (Figure 3c). Nevertheless, for the range of 2005-2011, it is observed stability and slightly recovering in the values of NDVI in the bigger portion of the region (Figure 3d). In the rest of the analyzed dates, moderate decreases of NDVI are presented throughout the area, and some areas show high and moderate increases, especially, in the eastern and southern portions (Figures 3a and 3b).
Figure 3 Maps showing changes in NDVI for a couple of dates: a) 1990-1995; b) 1995-2000; c) 2000-2005; d) 2005-2011
Figure 4 Maps showing the changes of NDWI for a couple of years: a) 1990-1995; b) 1995-2000; c) 2000- 2005; d) 2005-2011
The outcomes of the NDWI analysis allow observing decreases in moisture between the ranges 1990-1995 and 2000-2005 (Figures 4a and 4c). In contrast, the ranges 1995-2000 and 2005-2011 showed increases of moisture in most of the region (Figures 4b and 4d).
The dynamics of ST show for the first couple of dates (1990-1995) a high increase in the greater part of the region (Figure 5a). Between 1995 and 2000 a low decrease of ST appears in the entire study area, except for small areas which present high and moderate increase (Figure 5b). Between the years 2000 and 2005 there is a modest decrease in most of the region and to the Central and Southern portion, the reduction of ST is low (Figure 5c). The high increases in ST for the whole region returns to occur between 2005 and 2011 (Figure 5d).
Figure 5 Maps showing changes of ST for a couple of dates: a) 1990-1995; b) 1995-2000; c) 2000-2005; d) 2005-2011
The analysis of change for IO produced the following results: For the first time range (1990-1995), a lower increase of IO is observed in the entire study area, and a stable state of IO is in the south of the study area (Figure 6a). Between 1995 and 2000, the whole area remains stable values of IO (Figure 6b). However, between 2000 and 2005 there are small zones with little decrements of IO (Figure 6c) and, for the last time range (2005-2011), high and low increments of IO are again observed in the entire study area (Figure 6d).
Figure 6 Maps showing the change of IO for a couple of years: a) 1990-1995; b) 1995-2000; c) 2000-2005; d) 2005-2011
The results do not allow establish a clear trend of steady decrease for NDVI, NDWI, and IO, or a constant increase of ST. This kind of behavior would be expected for a zone that is developing processes of desertification (Dorn, 1986; Collado et al., 2002; Ustin, 2009; Díaz-Padilla et al., 2011). Rather, the behavior in all factors is fluctuating, so there are dates with significant increases, followed by periods of time where the values of the factors decrease. For this reason, it is necessary to obtain the final balance of the behavior of each parameter, to get an overview more realistic of the changes generated in the study area.
The final balance of NDVI permits to verify decreases greater than 0.3 in the West-Central and North of the study zone (blue color; Figure 7a), involving mainly the municipalities of Santo Domingo, Vanegas and some offices of Real de Catorce, Charcas, and Guadalcazar. The rest of the zone remains stable in the values of NDVI (green color; Figure 7a).
Figure 7 Maps of total balance for a) NDVI; b) NDWI; c) ST; d) IO
The overall balance of NDWI allows discovering different levels of decrease in this parameter across the area (Figure 7b). In some areas, although the NDWI decrease is high, the vegetation is retained which indicates the presence of vegetation with low levels of humidity. This characteristic is common in the vegetation of the desert (Huber-Sannwald et al., 2006). The municipalities most affected by decreasing moisture are Santo Domingo, Villa de Ramos, Charcas, Venado, Villa de Guadalupe and Villa Hidalgo (blue color; Figure 7b).
The map of the total balance of ST shows that the area most affected by high increases of temperature is the western portion of the SLPP, where it is registered rises above 3°C (red color; Figure 7c). The municipalities most affected by high increases in temperature are Santo Domingo, Villa de Ramos, Salinas, Catorce, Vanegas, Matehuala, and Cedral. In contrast, the eastern and southern parts have high decreases in ST (light blue; Figure 7c).
The total balance of IO shows a status of stability for the most of the SLPP (Green color; Figure 7d). This outcome corroborates the non-removal of the layer of desert varnish for the most of the region and, only in some parts this layer decreases slightly. However, there are some municipalities affected by slight declines in IO, such as Santo Domingo, Real de Catorce, Cedral, Matehuala, Villa de Guadalupe, Guadalcázar and Villa de Arista (blue color; Figure 7 d). This effect is logical because these municipalities have an important farming and mining activity, which eases the removal of desert varnish.

3.2 Potential areas to develop desertification in semi-arid environments

Just parts of the region show adverse changes in some of the images of the final balance. However, with the purpose of establishing a more objective view of the phenomena of desertification in the region, a map of probabilities of desertification was produced based on the four images of the final balance. The resulting map shows small possibilities to develop processes of desertification in most of the field (blue color in Figure 8). Only some parts are more susceptible to problems of desertification (red color in Figure 8) because they demonstrate substantial decreases of NDVI, NDWI (decrease > 0.3), IO (decrease > 0.5) and significant increases for ST (increase < -3°C; Figure 7). In the field verification, the zones with very high desertification probabilities present values greater than 20% of the surface occupied by desiccation cracks. Vegetation is practically nonexistent; the ground temperature is above 28°C and the soil is dry (Figure 9e). The area covered by this level is 0.2%, and the most affected municipality is Santo Domingo (Figure 8 and Table 4).
Figure 8 Map showing the areas with the highest probability of developing processes of desertification in the SLPP.
Figure 9 Landscapes related to different stages of desertification probability: a) Very low (point SCh-3), b) Low (point Sch-12), c) Moderate (point Sch-9), d) High (point Sch-19) and e) Very high (point Sch-16). The pictures were acquired during the field verification.
Table 4 Surface affected by the level of probability of developing desertification
Probability of desertification Surface (km2) Surface (%)
Very high 1.20 0.02
High 2154.39 8.78
Moderate 4568.39 18.63
Low 8637.56 35.23
Very low 9154.67 37.34
Total 24516.21 100.00
Following, is the area with a high potential for desertification which occupies an area of 8.78% (Table 4). This stage shows the slight presence of vegetation. The desiccation cracks occupy 10%-20% of the soil surface, the increase of ST varies between -0.1°C and -3°C. NDVI and NDWI have decreases ranging between 0.01 and 0.3 and IO declines between 0.01 and 0.5 (Figure 7). On the ground, a limited coverage of desert varnish is observed with a ground temperatures of 25°C (Figure 9d). The municipalities most affected are Santo Domingo, Villa de Ramos, Real de Catorce, Salinas de Hidalgo, and Charcas (Figure 8).
In contrast, areas with very small possibility of desertification show a substantial coverage of green vegetation, ground temperatures less than 20°C, high levels of moisture and the desiccation cracks are non-existent (Figure 9a). This level of probability occupies the largest area of the region (37.34%; Table 4). In this case, the final balance of NDVI and NDWI show the highest increases (<-0.3), IO also shows significant increases (<-0.5) and ST presents the largest decreases (> 3°C; Figure 7). The moderate level of desertification occupies 18.63% of the total area (Table 4 and Figure 8). In field verification, this level is characterized by patches of green and woody vegetation. The range of ground temperature is between 22.5°C and 25°C. The coverage of desert varnish is variable, and few desiccation cracks are observed (Figure 9c). In this level, the final balance shows very slight variations in NDVI, NDWI, IO and ST (between -0.01 to 0.01; Figure 7).
The level of low probability is also widespread in the study area (35.23%; Table 4 and Figure 8). The zones that correspond to this level have ample coverage of woody vegetation, the desert varnish is well developed, the ground temperatures range between 20°C and 22.5°C, and there are no desiccation cracks on the ground (Figure 9b). NDVI, NDWI, and IO show increase in the final balance (-0.01 to -0.3 for NDVI and NDWI; -0.01 to -0.5 for IO). ST presents decreases, which varies between 0.01°C to 3°C (Figure 7).
The field validation confirms an accuracy of 85% between field observations and the estimated data (Table 5). Therefore, these outcomes suggest that if each factor is analyzed on an individual basis, it could be concluded that almost the whole region is susceptible to a high level of desertification. However, the combination of all factors allows establishing a more balanced scene of the real status of the study zone concerning the hazard of generating desertification.
Table 5 Confusion matrix showing the agreement between field observations (columns) and the estimated data (rows)
Very high High Moderate Low Very low Total User accuracy Error of commission
Very high 1 0 0 0 0 1 1.0 0.0
High 0 1 0 0 0 1 1.0 0.0
Moderate 0 0 3 0 0 3 1.0 0.0
Low 0 0 0 5 0 5 1.0 0.0
Very Low 0 0 1 2 7 10 0.7 0.3
Total 1 1 4 7 7 20
Producer accuracy 1.0 1.0 0.8 0.7 1
Error of omission 0.0 0.0 0.2 0.3 0.0

Global accuracy 85.0
Kappa coefficient 0.84

3.3 Discussion

The final result of this analysis can be explained according to climatic oscillations. For example, NDVI values from the year 2000 show a decrease related to precipitation values reported by the National Weather Service from Mexico (Figure 10a). This relationship be-tween rainfall and greening or decrease in vegetation has been discussed for San Luis Potosi and other parts of the world (Miranda-Aragon et al., 2012; Sun et al., 2013) and, it corresponds with the observed in this study.
Figure 10 Comparison of the behavior of a) NDVI, b) ST, c) IO, and d) NDWI with precipitation or temperature obtained from the National Weather Service. The data of temperature and precipitation are the average of 15 days before the dates analyzed in this work. The standards were obtained based on information from 50 climatological stations.
The temporal behavior of the average values of NDVI allows verifying a trend of relative stability characterized by a strong decrease of NDVI between 2000 and 2005 (> 0.1) followed by a slight recovery in the value of NDVI (<-0.01; Table 6a). This result is explained regarding the character of vegetation that predominates in the area, and the deforestation processes related to the development of agriculture and grassland areas in Mexico (Díaz-Padilla et al., 2011; Miranda-Aragón et al., 2013), and in other continents (Helldén and Trttrup, 2008).
Table 6 Average values obtained during each period analyzed for a) NDVI, b) ST, c) IO and d) NDWI. Subsection e) corresponds to the average value of the total balance for each factor
a) Years Minimum Maximum Average
1990-1995 -0.59 0.63 0.01
1995-2000 -0.61 0.58 0.02
2000-2005 -0.47 0.67 0.11
2005-2011 -0.62 0.52 -0.01
b) Years Minimum Maximum Average
1990-1995 -14.63 14.34 -1.93
1995-2000 -18.19 9.01 -0.85
2000-2005 -13.26 18.7 8.90
2005-2011 -28.01 -7.89 -5.86
c) Years Minimum Maximum Average
1990-1995 -3.37 1.50 -0.15
1995-2000 -1.63 2.75 -0.04
2000-2005 -0.65 0.70 0.08
2005-2011 -1.5 0.63 -0.1
d)Years Minimum Maximum Average
1990-1995 -1.03 1.94 0.51
1995-2000 -1.38 1.16 -0.06
2000-2005 -1.27 1.36 0.16
2005-2011 -1.28 1.08 -0.14
e) Variable Average
NDVI 0.03
ST -3.77
NDWI 0.26
IO -0.07
The average difference between the extreme values of ST is 5.86°C. Comparing this result with the environmental temperature (ET) registered by the National Meteorological Service (6.56°C), it can be seen that the behavior of both temperatures (ST vs. ET) is very similar (Figure 10b). Also, ST values obtained in this work from the thermal band of Landsat TM, are consistent with those reported by other studies carried out in arid and semi-arid zones (Sánchez et al., 2008; Tereshchenko et al., 2012). The results allow verifying that high ST variations are related to the El Niño and La Niña phenomena except the year 1990 which does not present an event of this kind (Null, 2013). Temperature difference analysis shows increases in almost all pairs of date, except for the pair 2000-2005 with a positive value of 8.9°C, i.e., there is a significant decrease in temperature between these two dates. The weather station also reported this reduction in temperature. With this result, it is noted the fluctuations in temperature and not a constant increase of this one (Table 6e).
The results of the IO dynamics show variations in the content of the iron oxides among dates analyzed. A comparison between the values of IO and data of rainfall, suggests increases in IO when the precipitation values decrease (Figure 10c). In contrast, the comparison between the values of IO and ST draws a direct correlation between the two parameters (Figure 10d). Figure 7d shows the areas with the highest increase in IO located mainly in the west of the SLPP. Some of these regions have been previously reported as areas with alteration of oxidation - silicification derived from hydrothermal processes (Moreira-Rivera et al., 1996).
On the other hand, INEGI (2012) reported for this zone, soils with high concentrations of organic matter and iron oxides (Phaeozem and Kastanozem). The variability of iron oxides can be related to soil degradation processes as other works have suggested (Haboudane et al., 2002). Also, the desert varnish is an indicator of recent disturbances in the desert regions. Therefore, the IO values decrease can be related to the removal of the desert varnish (Spatz et al., 1989; Ustin et al., 2009). Then, in the study area, the fluctuations in the values of IO (Table 6c) suggest the removal of desert varnish in particular periods of time due to the growth of crop fields, the opening of roads and other types of activities such as mining. This kind of remark has also been registered in other regions of the world (Haboudane et al., 2002; Ustin et al., 2009).
By comparing the behavior of the NDWI with precipitation data from the National Weather Service (Figure 10e), it is observed a similar behavior between both parameters. From the observations of Gao (1996) and based on the results obtained for the region, it can be proposed that the average values of NDWI indicate the presence of soil with sparse vegetation and little moisture retention. This result is consistent with reports for other semi-arid regions as the SLPP (Li et al., 2004; Wang et al., 2007; Xu et al., 2011). The analysis of pairs of dates (Table 6d) shows alternating periods of increases in NDWI (1995-2000 and 2005-2011), with cycles of decrements of this factor (1990-1995 and 2000-2005). These results demonstrate the fluctuations in moisture related to climatic oscillation phenomena (Figure 10e).
Table 6e shows the NDVI relatively stable from 1990 until 2011 (total balance = 0.03), i.e., this parameter has preserved very similar values at the time, and its changes depend on the oscillations of the precipitation and temperature. NDWI presents an overall decrease reported by its average value (0.26) which can be explained from the output of the total balance of ST (-3.77°C), suggesting a general increase in temperature, that's why the evaporation processes also increase. In the case of IO, the average value of the total balance (-0.07) shows some stability of this factor over time analyzed.
The results of the susceptibility map show very high probability of developing desertification in a small zone of the region, which occupies an area of 1.2 km2 (Table 4). In contrast, the lowest probability of desertification occupies the largest surface of the area (9154.67 km2) followed by the category of low probability of desertification with an extension of 8637.56 km2. These results reveal cyclical changes in vegetation and moisture in the region. Hence, most of the study area don't have a high probability of developing desertification processes. The result is consistent with the work of Miranda-Aragón and others (2013) which point out that the SLPP does not have severe decreases in vegetation and the principal reductions are related to the scrub. In addition, Roth et al. (2016) indicate a low degree of human disturbing for this region of Mexico. This fact is an essential element in the processes of desertification as pointed out by other studies (Tao, 2004; Huber-Sannwald et al., 2006; Santini et al., 2010;. Becerril-Piña et al., 2015). The result also confirms the approach of the re-greening of vegetation and the development of cyclical phenomena, rather than an advance of desertification, as it has been observed in other regions of the world such as the Sahel (Dardel et al., 2014; Ouedraogo et al., 2014). The result obtained in this work is important to promote better arguments for decision-making processes for an adequate management of semi-arid areas.

4 Conclusions

This work focused on the analysis of Landsat TM images with the purpose of performing the temporary monitoring of the evolution of four factors (NDVI, ST, IO, and NDWI) to detect possible zones of developing desertification processes in the SLPP on the southern edge of the Chihuahuan Desert.
A clear trend for any of the factors analyzed are visible; rather they have a fluctuating behavior. Therefore, a new equation of total balance is proposed to hold the overall evolution of each factor.
The map of probability obtained in this study helped to corroborate that the most of the SLPP has a very low probability of developing desertification processes, and this category covers an area of 9154.67 km2 (37.34%). The next largest category corresponds to the low risk of desertification, which occupies an area of 8637.56 km2 (35.23%).
The results of this study confirm the approach of vegetation re-greening and the dynamics controlled by cyclical changes of climate typical of semi-desert zones. For that reason, only 8.8% of the study area requires more attention for its preservation and the rest of the region doesn’t show significant advances of desertification. The municipalities with a greater probability of developing desertification processes are Santo Domingo, Villa de Ramos, Real de Catorce, Salinas de Hidalgo, and Charcas.
The methodology proposed in this paper proved to be useful to establish more concise settings to generate an adequate management plans for semi-desert zones.

The authors have declared that no competing interests exist.

References

Publishing order | Descend order by publishing year | Descend order by cited within
[1]
Becerril-Piña R, Mastachi-Loza C A, González-Sosa E et al., 2015. Assessing desertification risk in the semi-arid highlands of central Mexico.Journal of Arid Environments, 120: 4-13. doi: 10.1016/j.jaridenv.2015.04.006.This study presents the integration of a set of indices representing four desertification factors: vegetation, soil, climate and anthropic disruptors. The Desertification Trend Risk Index (DTRI) adds the mean slope of each factor in a time series, thereby classifying the risk of desertification according to four categories: low, medium, high and extreme. This index is based on Landsat images and population data from 1993 to 2011, as well as climatological data from 1980 to 2010. An analysis using a combination matrix is also presented to identify the driving forces in each category. In addition, a change vector analysis (CVA) is performed to evaluate changes in land use in the semi-arid highlands of central Mexico. The results show that the semi-arid areas of the municipalities of Quer taro, Santa Catarina, Corregidora, Apaseo el Alto and Celaya represent extreme alert regions. In general, anthropic activities such as changes in land use and deforestation are the primary driving forces in the desertification process in the region. Based on the results of this study, the use of the DTRI is recommended as a low-cost and easily applied tool to assess and monitor desertification.

DOI

[2]
Chander G, Markham B L, Helder D L, 2009. Summary of current radiometric calibration coefficients for Landsat MSS, TM, ETM+, and EO-1 ALI sensors.Remote Sensing of Environment, 113(5): 893-903. doi: 10.1016/j.rse.2009.01.007.This paper provides a summary of the current equations and rescaling factors for converting calibrated Digital Numbers (DNs) to absolute units of at-sensor spectral radiance, Top-Of-Atmosphere (TOA) reflectance, and at-sensor brightness temperature. It tabulates the necessary constants for the Multispectral Scanner (MSS), Thematic Mapper (TM), Enhanced Thematic Mapper Plus (ETM+), and Advanced Land Imager (ALI) sensors. These conversions provide a basis for standardized comparison of data in a single scene or between images acquired on different dates or by different sensors. This paper forms a needed guide for Landsat data users who now have access to the entire Landsat archive at no cost.

DOI

[3]
Collado D A, Chuvieco E, Camarasa A, 2002. Satellite remote sensing analysis to monitor desertification processes in the crop-rageland boundary of Argentina.Journal of Arid Environments, 52(1): 121-133. doi: 10.1006/jare.2001.0980.Remote sensing digital image analysis is applied to monitor desertification processes in central San Luis Province (Argentina), where signs of severe landscape degradation have been observed in the last decades. Two Landsat images, acquired in 1982 and 1992 were used to evaluate the potential of using remote sensing analysis in desertification monitoring. After geometric and radiometric correction of both images, multitemporal comparison techniques were utilized to emphasize areas of greater degradation. Spectral unmixing of the vegetation, water and sand components facilitated the analysis of areas of heterogeneous cover from satellite images. Simple differences between unmixed images of sand or water revealed dune movement, re-vegetation trends and variations in water bodies, as a result of changing rainfall and land use patterns.

DOI

[4]
Dardel C, Kergoat L, Hiernaux P et al., 2014. Re-greening Sahel: 30 years of remote sensing data and field observations (Mali, Niger).Remote Sensing of Environment, 140: 350-364. doi: 10.1016/j.rse.2013.09.011.Desertification of the Sahel region has been debated for decades, while the concept of a “re-greening” Sahel appeared with satellite remote sensing data that allowed vegetation monitoring across wide regions and over increasingly long series of years (nowadays 3002years with the GIMMS-3g dataset). However, the scarcity of long-term field observations of vegetation in the Sahel prevents ground validation and deeper analysis of such trends. After assessing the consistency of the new GIMMS-3g NDVI product by comparison to three other AVHRR-NDVI datasets and MODIS NDVI, regional GIMMS-3g NDVI trends over 1981–2011 are analyzed. Trends are found positive and statistically significant almost everywhere in Sahel over the 1981–2011 period. Long-term field observations of the aboveground herbaceous layer mass have been collected within the Gourma region in Mali (1984–2011) and within the Fakara region in western Niger (1994–2011). These observations sample ecosystem and soil diversity, thus enabling estimation of averaged values representative of the Gourma and Fakara. NDVI measurements are found in good agreement with field observations, both over the Gourma and Fakara regions where re-greening and negative trends are observed respectively. A linear regression analysis performed between spatially averaged seasonal NDVI and a weighted average of field measurements explains 59% of the variability for the Gourma region over 1984–2011, and 38% for the Fakara region over 1994–2011. In the Gourma, which is a pastoral region, the re-greening trend is mainly observed over sandy soils, and attests for the ecosystem's resilience to the 1980s' drought, able to react to the more favorable rainfall of the 1990s and 2000s. However, contrasted changes in the landscape's functioning have occurred locally. An increase in erosion and run-off processes in association with decreasing or stable vegetation cover was observed over shallow soils, which occupy 30% of the area. In the agro-pastoral Fakara, the decreasing trends observed both from satellite NDVI and field assessments of herbaceous mass are hardly explained by rainfall. These results give confidence in the dominant positive trends in Sahelian greenness, but indicate that degradation trends can also be observed, both in situ and from satellite time series.

DOI

[5]
Díaz-Padilla G, Sánchez-Cohen I, Guajardo-Panes R A et al., 2011. Mapping of the aridity index and its populations distribution in Mexico.Revista Chapingo. Serie Ciencias Forestales y del Ambiente, 17(SPE): 267-275. doi: 10.5154/r.rchscfa.2010.09.069.

[6]
De Waroux Y L P, Lambin E F, 2012. Monitoring degradation in arid and semi-arid forests and woodlands: The case of the argan woodlands (Morocco).Applied Geography, 32(2): 777-786. doi: 10.1016/j.apgeog.2011.08.005.Arid and semi-arid forests and woodlands (hereafter called dryland forests ), in spite of their ecological and social importance, have received little attention in land change studies. Growing evidence shows that these forests have been receding at very high rates in many places, suggesting a need for a better understanding of the processes and causes of dryland forest degradation. Changes in the extent of dryland forests are debated in part because estimates of forest and woodland areas in drylands are uncertain. Causal explanations of the degradation tend to draw on the literature on desertification and tropical deforestation, and to emphasize either local or remote, and either social or biophysical drivers. This study contributes to a better understanding of dryland forest degradation as a basis for conservation policies. Firstly, we argue that monitoring arid and semi-arid forests and woodlands using area estimates may lead to an underestimation of the severity of change because tree density change often exceeds area change. Secondly, we argue that the analysis of degradation processes in these multifunctional landscapes should integrate both local and remote, and both social and biophysical factors. We use a case study of degradation in the argania woodlands in semi-arid to arid Southwest Morocco to test these two claims. We used gridded tree counts on aerial photographs and satellite images to estimate forest change between 1970 and 2007, and we tested several possible causes of change on the basis of original socio-economic field surveys and climatic and topographic data. We found that forest density declined by 44.5% during this period, a figure that is significantly underestimated if forest area change is used as a measure of degradation. Increasing aridity and, to a lesser extent, fuelwood extraction were related to forest decline. No effect of grazing by local livestock was found.

DOI

[7]
Dorn R I, 1986. Rock varnish as an indicator of aeolian environmental change. In: Nickling W G (ed.). Aeolian Geomorphology. Boston: Allen & Unwin, p. 291-307.

[8]
Drury S, 1987. Image Interpretation in Geology. London: Allen and Unwin, p. 243.

[9]
Ezcurra E, 2006. Global Deserts Outlook. United Nations Environment Programme (UNEP): Nairobi, 148 p.

[10]
Galindo A S, García M E, 1986. Usos del mezquite (Prosopis L.) en el Altiplano Potosino.Agrociencia, 63: 7-16.

[11]
Gamo M, Shinoda M, Maeda T, 2013. Classification of arid lands, including soil degradation and irrigated areas, based on vegetation and aridity indices.International Journal of Remote Sensing, 34(19): 6701-6722. doi: 10.1080/01431161.2013.805281.It is preferable to prepare internally consistent maps of arid regions on a global scale in order to understand the present conditions of arid regions, especially deserts and soil degradation areas. We attempted to delimit arid regions at a global scale by combining climate data, i.e. aridity index (AI), and vegetation data, i.e. vegetation index. The annual AI was estimated by the ratio of mean annual precipitation to mean annual potential evapotranspiration, using the Thornthwaite method. The long-term mean of yearly maximum normalized difference vegetation index (NDVI) was used as an indicator of the vegetation condition. Arid regions of the world were classified into four categories, namely A, severe deserts, where both aridity and vegetation indices are very small; G, semi-arid regions, where the vegetation index is proportionally related to the AI; I, irrigated areas and oases, where the vegetation is relatively abundant despite severe dryness; and S, soil degradation areas, where the vegetation is poor despite relatively humid conditions. The Sahel from Niger to Chad, the Sahel in Darfur, and the Ordos Plateau in China are within Category S. The standard deviation of NDVIis very small/large in severe deserts/semi-arid areas, respectively. Thus, the Sahara desert was clearly distinguished from the Sahel; the latter belongs to Category G and drought occurs frequently here. In Category S zones, the standard deviation of NDVIis relatively small compared with that within the Category G zone because the return rainfall does not seem to promptly restore productivity. Category S was divided into three subdivisions according to the degree of degradation, expressed by the ratio of the AI to vegetation index. Category G was also divided into four classes, according to degree of vegetation (or aridity). The distribution of Category S is comparable to the soil degradation areas mapped by Global Assessment of Human-Induced Soil Degradation (GLASOD) data. True deserts, where the standard deviation of NDVIis very small, were selected from the evere desert group. Desert areas were classified as true deserts, severe deserts, grassland deserts (Category G), and soil degradation deserts (Category S).

DOI

[12]
Gao B C, 1996. NDWI: A normalized difference water index for remote sensing of vegetation liquid water from space.Remote Sensing of Environment, 58(3): 257-266. doi: 10.1016/S0034-4257(96)00067-3.The normalized difference vegetation index (NDVI) has been widely used for remote sensing of vegetation for many years. This index uses radiances or reflectances from a red channel around 0.66 μm and a near-IR channel around 0.86 μm. The red channel is located in the strong chlorophyll absorption region, while the near-IR channel is located in the high reflectance plateau of vegetation canopies. The two channels sense very different depths through vegetation canopies. In this article, another index, namely, the normalized difference water index (NDWI), is proposed for remote sensing of vegetation liquid water from space. NDWI is defined as (09(0.86 μm) 61 09(1.24 μm))(09(0.86 μm) + 09(1.24 μm)), where 09 represents the radiance in reflectance units. Both the 0.86-μm and the 1.24-μm channels are located in the high reflectance plateau of vegetation canopies. They sense similar depths through vegetation canopies. Absorption by vegetation liquid water near 0.86 μm is negligible. Weak liquid absorption at 1.24 μm is present. Canopy scattering enhances the water absorption. As a result, NDWI is sensitive to changes in liquid water content of vegetation canopies. Atmospheric aerosol scattering effects in the 0.86–1.24 μm region are weak. NDWI is less sensitive to atmospheric effects than NDVI. NDWI does not remove completely the background soil reflectance effects, similar to NDVI. Because the information about vegetation canopies contained in the 1.24-μm channel is very different from that contained in the red channel near 0.66 μm, NDWI should be considered as an independent vegetation index. It is complementary to, not a substitute for NDVI. Laboratory-measured reflectance spectra of stacked green leaves, and spectral imaging data acquired with Airborne Visible Infrared Imaging Spectrometer (AVIRIS) over Jasper Ridge in California and the High Plains in northern Colorado, are used to demonstrate the usefulness of NDWI. Comparisons between NDWI and NDVI images are also given.

DOI

[13]
Granados-Sánchez D, Hernández-García M Á, Vázquez-Alarcón A et al., 2013. The processes of desertification and arid regions.Revista Chapingo. Serie Ciencias Forestales y del Ambiente, 19(1): 45-66. doi: 10.5154/r.rchscfa.2011.10.077.

[14]
Haboudane D, Bonn F, Royer A et al., 2002. Land degradation and erosion risk mapping by fusion of spectrally-based information and digital geomorphometric attributes.International Journal of Remote Sensing, 23(18): 3795-3820. doi: 10.1080/01431160110104638.The Guadalentin basin, located in the SE of Spain, has a semiarid climate and presents typical characteristics of Mediterranean landscapes vulnerable to land degradation processes and desertification risks. In such an environment, when the vegetation cover is low, the signal received by satellites is dominated by the spectral properties of soils. Changes in these properties can be interpreted in terms of varying soil surface conditions. These optical changes underline the major modifications affecting soil surface under land degradation processes. The present research uses remote sensing techniques to characterise land degradation based on two approaches: spectral mixture analysis and a set of indices describing the spectrum shape. It also presents an integrated approach for evaluating ecosystem vulnerability to land degradation, through the combined analysis of spectrally-derived land units and geomorphometric units. Specific objectives consist of evaluating the potential of extending the indices describing the spectrum shape to the short-wave infrared region, and of identifying landscape units according to their sensitivity to land degradation. Our results demonstrate that the spatial distribution of regional patterns of land degradation can be reliably mapped by using both indices describing the spectrum shape and spectral unmixing. The latter holds great potential for operational mapping of soil conditions and erosion features from optical images. Moreover, landscape-unit analysis shows that DEM (Digital Elevation Model) variables combined with spectral information are very useful for land degradation assessment. This approach allowed us to segment the landscape into different units according to their lithology and vegetation characteristics, as well as their susceptibility to water erosion.

DOI

[15]
Helldén U, 1992. Desertification: Time for an assessment. Ambio, 20: 372–383.

[16]
Helldén U, Tottrup C, 2008. Regional desertification: A global synthesis.Global and Planetary Change, 64(3/4): 169-176. doi: 10.1016/j.gloplacha.2008.10.006.The paper presents results on the use of NOAA AVHRR data for desertification monitoring on a regional–global level. It is based on processing of the GIMMS 802km global NDVI data set. Time series of annually integrated and standardized annual NDVI anomalies were generated and compared with a corresponding rainfall data set (1981–2003). The regions studied include the Mediterranean basin, the Sahel from the Atlantic to the Red Sea, major parts of the drylands of Southern Africa, China–Mongolia and the drylands of South America, i.e. important parts of the desertification prone drylands of the world. It is concluded that the suggested methodology is a robust and reliable way to assess and monitor vegetation trends and related desertification on a regional–global scale. A strong general relationship between NDVI and rainfall over time is demonstrated for considerable parts of the drylands. The results of performed trend analysis cannot be used to verify any systematic generic land degradation/desertification trend at the regional–global level. On the contrary, a “greening-up” seems to be evident over large regions.

DOI

[17]
Hernández H M, Gómez-Hinostrosa C, Bárcenas R T, 2001. Diversity, spatial arrangement, and endemism of Cactaceae in the Huizache area, a hot-spot in the Chihuahuan Desert.Biodiversity & Conservation, 10(7): 1097-1112. doi: 10.1023/A:1016606216041.We studied several aspects of the cactus diversity in the Huizache, an area located in the southern extreme of the Chihuahuan Desert Region, in San Luis Potosí, Mexico. Fieldwork was conducted in a square-shaped polygon (22°30′–23°00′ N, 100°00′–100°30′ W), where a total of 80 systematically-chosen localities were sampled for cactus species. Results showed that the Huizache is an important focal point for the conservation of the Cactaceae, due to the remarkably high diversity of members of this plant family in the area. With 75 species recorded, this area has the highest concentration of cactus species in the American Continent, in comparison with other comparably-sized regions. The outstanding diversity of cactus species in the Huizache area may be explained by the relatively favorable climate of this desert area, its environmental heterogeneity, and its location in a transitional position between three regions with distinct cactus floras: Chihuahuan Desert Region, Queretaroan-Hidalgoan Arid Zone, and Tula-Jaumave Valleys. Species richness is not distributed evenly in the area; the zones of highest species concentration are primarily located in the lowland valleys, in the northwestern portion of the grid square, where typical Chihuahuan Desert conditions prevail. Of the 75 cactus species found in the area, 63% are endemic to the Chihuahuan Desert. Among these, ten species are outstanding for being endemic or nearly endemic to the Huizache area. The Huizache grid square is part of a recently declared natural reserve, The Real de Guadalcázar Natural Reserve.

DOI

[18]
Huber-Sannwald E, Maestre F T, Herrick J E et al., 2006. Ecohydrological feedbacks and linkages associated with land degradation: A case study from Mexico.Hydrological Processes, 20(15): 3395-3411. doi: 10.1002/hyp.6337.Abstract Land degradation and desertification are major environmental problems of human societies in drylands of North and South America. Mexico is one of the most severely affected countries. An assessment of how both biophysical and socio-economic processes simultaneously affect, and are affected by, land degradation is recognized as one of the most important and challenging topics in the research on global change. Towards meeting this challenge, in June 2004 an interdisciplinary mix of scientists assembled in Mexico to participate in a workshop convened by the ARID net network. The focus of the workshop was to apply a new conceptual framework he Dahlem Desertification Paradigm (DDP) o La Amapola, a small rural community located in the Central Plateau of Mexico. The DDP aims to advance understanding of global desertification issues by focusing on the interrelationships within coupled human-environment systems that cause desertification. In this paper we summarize the conclusions of the La Amapola workshop. First, we present a brief review of some of the broader issues and concerns of global desertification, which led to the formation of ARID net and to the DDP. Second, we provide an overview of land degradation issues in La Amapola, highlighting examples of hydrological linkages between biophysical and socio-economic factors. Third, we summarize our findings in a conceptual model, which highlights linkages between biophysical and socio-economic factors in La Amapola, and the role of hydrology in desertification. Lastly, we discuss the results derived from the application of the major assertions of the DDP to La Amapola. The numerous feedbacks, linkages, and causal pathways between the biophysical and human dimensions suggest that hydrology is the fundamental component of the livelihoods of rural communities in this region of Mexico, and thus it is of central importance when evaluating desertification. Copyright 2006 John Wiley & Sons, Ltd.

DOI

[19]
Huete A R, Didan K, Miura T et al., 2002. Overview of the radiometric and biophysicial performance of the MODIS Vegetation Indices.Remote Sensing of Environment, 83: 195-213. doi: 10.1016/S0034-4257(02) 00096-2.We evaluated the initial 12 months of vegetation index product availability from the Moderate Resolution Imaging Spectroradiometer (MODIS) on board the Earth Observing System-Terra platform. Two MODIS vegetation indices (VI), the normalized difference vegetation index (NDVI) and enhanced vegetation index (EVI), are produced at 1-km and 500-m resolutions and 16-day compositing periods. This paper presents an initial analysis of the MODIS NDVI and EVI performance from both radiometric and biophysical perspectives. We utilize a combination of site-intensive and regionally extensive approaches to demonstrate the performance and validity of the two indices. Our results showed a good correspondence between airborne-measured, top-of-canopy reflectances and VI values with those from the MODIS sensor at four intensively measured test sites representing semi-arid grass/shrub, savanna, and tropical forest biomes. Simultaneously derived field biophysical measures also demonstrated the scientific utility of the MODIS VI. Multitemporal profiles of the MODIS VIs over numerous biome types in North and South America well represented their seasonal phenologies. Comparisons of the MODIS-NDVI with the NOAA-14, 1-km AVHRR-NDVI temporal profiles showed that the MODIS-based index performed with higher fidelity. The dynamic range of the MODIS VIs are presented and their sensitivities in discriminating vegetation differences are evaluated in sparse and dense vegetation areas. We found the NDVI to asymptotically saturate in high biomass regions such as in the Amazon while the EVI remained sensitive to canopy variations.

DOI

[20]
INEGI, 2012. Anuario de estadísticas por entidad federativa. Aguascalientes, México. Available at: Accessed 15 August 2013.

[21]
Li H, Wang X F, Gao Y Q, 2004. Analysis and assessment of land desertification in Xinjiang based on RS and GIS.Journal of Geographical Sciences, 14(2): 159-166. doi: 10.1007/BF02837531.Landdesertificationreferstolanddegradationinarid,semi-aridanddrysub-humidareasresultingfromvariousfactors,includingclimaticvariations,humanactivities,etc.(CCICCD,1996).XinjiangislocatedatthenorthwestborderareaofChina.Withthelengthof2000kmfromeasttowestan

DOI

[22]
Miranda-Aragón L, Treviño-Garza E J, Jiménez-Pérez J et al., 2012. NDVI-rainfall relationship using hyper-temporal satellite data in a portion of North Central Mexico (2000-2010).African Journal of Agricultural Research, 7(6): 1023-1033. doi: 10.5897/AJAR11.1674.Nowadays, availability of hyper-temporal satellite data offers options for studying natural environments that were unattainable by traditional measurement approaches. This work focuses on monitoring vegetation using the Moderate Resolution Imaging Spectroradiometer (MODIS) sensor and rainfall data in the state of San Luis Potosi, Mexico. A hyper-temporal dataset comprising 489 MODIS MCD43A4 imagery (500 m spatial resolution) taken over the last 11 years (2000 to 2010) was used. Ten temporal yearly composites were constructed by fusing normalized difference vegetation indices (NDVI) data along each year using the lsquo;maximum composite value approachrsquo;. The time-series dataset was also fused using the principal component analysis (PCA) technique; the second principal component captured the main inter-annual anomalies from maximum yearly NDVI values. The analysis of this time-series indicates two periods: a) 2000 to 2005 characterized by a drought during 2000, 2001 and 2005, and b) 2006 to 2010 marked by a trend indicating wetter conditions with maxima in 2008 and 2010. Finally, an upward trend in NDVI time-series was observed during the period studied and successfully validated with ancillary historical data of crop production area. Key words: Moderate resolution imaging spectroradiometer, MCD43A4, time-series, principal component analysis, maximum value composite.

DOI

[23]
Miranda-Aragón L, Treviño-Garza E J, Jiménez-Pérez J et al., 2013. Tasa de deforestación en San Luis Potosí, México (1993-2007).Revista Chapingo. Serie Ciencias Forestales y del Ambiente, 19(2): 201-215. doi: 10.5154/r.rchscfa.2011.06.044.

[24]
Moreira-Rivera F, Martínez-Rodríguez L, Palacios-García R et al., 1996. Carta Geológico-Minera Matehuala (F14-1). 1a edición. Servicio Geológico Mexicano, Pachuca, Hidalgo, México.

[25]
Null J, 2013. El Niño and La Niña years and intensities. Available at Accessed 20 September 2013.

[26]
Ouedraogo I, Runge J, Eisenberg J et al., 2014. The re-greening of the Sahel: Natural cyclicity or human-induced change?Land, 3(3): 1075-1090. doi: 10.3390/land3031075.The Sahel has been the focus of scientific interest in environmental-human dynamics and interactions. The objective of the present study is to contribute to the recent debate on the re-greening of Sahel. The paper examines the dynamics of barren land in the02Sahel of Burkina Faso through analysis of remotely-sensed and rainfall data from 1975–2011. Discussions with farmers and land management staff have helped to understand the anthropogenic efforts toward soil restoration to enable the subsistence farming agriculture. Results showed that area of barren land has been fluctuating during the02study period with approximately 10-year cyclicity. Similarly, rainfall, both at national and local levels has followed the same trends. The trends of the area of barren land and rainfall variability suggest that when rainfall increases, the area of barren land decreases and barren land increases when rainfall decreases. This implies that rainfall is one of the02main factors driving the change in area of barren land. In addition, humans have contributed positively and negatively to the change by restoring barren lands for agriculture using locally known techniques and by accelerating land degradation through intensive and inappropriate land use practices.

DOI

[27]
Oyama M D, Nobre C A, 2004. Climatic consequences of a large-scale desertification in Northeast Brazil: A GCM simulation study.Journal of Climate, 17(16): 3203-3213. doi: <ext-link ext-link-type="doi" xlink:href="1009-637X-27-4-420/1520-0442(2004)017<3203: CCOALD>2.0.CO">10.1175/1520-0442(2004)017<3203: CCOALD>2.0.CO;2.The climatic impacts of a large-scale desertification in northeast Brazil (NEB) are assessed by using the Center for Weather Forecasting and Climate Studies Center for Ocean Land Atmosphere Studies (CPTEC COLA) AGCM. Two numerical runs are performed. In the control run, NEB is covered by its natural vegetation (most of NEB is covered by a xeromorphic vegetation known as caatinga); in the desertification run, NEB vegetation is changed to desert (bare soil). Each run consists of five 1-yr numerical integrations. The results for NEB wet season (March May) are analyzed. Desertification results in hydrological cycle weakening: precipitation, evapotranspiration, moisture convergence, and runoff decrease. Surface net radiation decreases and this reduction is almost evenly divided between sensible and latent heat flux. Atmospheric diabatic heating decreases and subsidence anomalies confined at lower atmospheric levels are found. The climatic impacts result from the cooperative action of feedback processes related to albedo increase, plant transpiration suppression, and roughness length decrease. On a larger scale, desertification leads to precipitation increase in the oceanic belt close to the northernmost part of NEB (NNEB). In the NEB NNEB dipole, the anomalies of vertical motion and atmospheric circulation are confined to lower atmospheric levels, that is, 850 700 hPa. At these levels, circulation anomalies resemble the linear baroclinic response of a shallow atmospheric layer (850 700 hPa) to a tropical heat sink placed over NEB at the middle-layer level. Therefore, NEB climate does show sensitivity to a vegetation change to desert. The present work shows the possibility of significant and pronounced climate impacts, on both regional and large scales, if the environmental degradation in NEB continues unchecked.<HR ALIGN="center" WIDTH="30%">

DOI

[28]
Pando-Moreno M, Gutiérrez G, Maldonado H et al., 2002. Evaluación de los procesos de desertificación en una cuenca hidrológica del NE de México. Ciencia UANL, 5(4): 519-524.

[29]
Reynolds J F, Smith D M S, Lambin E F et al., 2007. Global desertification: Building a science for dryland development.Science, 316(5826): 847-851. doi: 10.1126/science.1131634.In this millennium, global drylands face a myriad of problems that present tough research, management, and policy challenges. Recent advances in dryland development, however, together with the integrative approaches of global change and sustainability science, suggest that concerns about land degradation, poverty, safeguarding biodiversity, and protecting the culture of 2.5 billion people can be confronted with renewed optimism. We review recent lessons about the functioning of dryland ecosystems and the livelihood systems of their human residents and introduce a new synthetic framework, the Drylands Development Paradigm (DDP). The DDP, supported by a growing and well-documented set of tools for policy and management action, helps navigate the inherent complexity of desertification and dryland development, identifying and synthesizing those factors important to research, management, and policy communities.

DOI PMID

[30]
Roth D, Moreno-Sanchez R, Torres-Rojo J M et al., 2016. Estimation of human induced disturbance of the environment associated with 2002, 2008 and 2013 land use/cover patterns in Mexico.Applied Geography, 66: 22-34. doi: 10.1016/j.apgeog.2015.11.009.Land use/cover types have an associated level of human induced disturbance of the environment. Previous studies in Mexico document the extent of land use/cover types and their changes over time. This information is important but not sufficient to support land use planning and environmental conservation decisions. This article explores the informational value of generating a numerical quantification of the level of human induced disturbance of the environment associated with land use/cover classes and their patterns at the national and state levels in Mexico. Estimates for land use/cover patterns existing in 2002, 2008 and 2013 are created. A disturbance scale created in a previous study was adapted and calibrated for 14 broad land use/cover types through domain expert opinions. The results indicate that the disturbance of the environment is increasing at the national level, but at a significantly slower rate between 2002 and 2008 (1.9%) and 2008鈥2013 (0.8%). Six states report a decrease in total environmental disturbance. A land use/cover transition analysis conducted for these states indicates a large proportion of transitions from anthropogenic land use/covers to natural ones.

DOI

[31]
Sánchez J M, Kustas W P, Caselles V et al., 2008. Modelling surface energy fluxes over maize using a two-source patch model and radiometric soil and canopy temperature observations.Remote Sensing of Environment, 112(3): 1130-1143. doi: 10.1016/j.rse.2007.07.018.Models estimating surface energy fluxes over partial canopy cover with thermal remote sensing must account for significant differences between the radiometric temperatures and turbulent exchange rates associated with the soil and canopy components of the thermal pixel scene. Recent progress in separating soil and canopy temperatures from dual angle composite radiometric temperature measurements has encouraged the development of two-source (soil and canopy) approaches to estimating surface energy fluxes given observations of component soil and canopy temperatures. A Simplified Two-Source Energy Balance (STSEB) model has been developed using a “patch” treatment of the surface flux sources, which does not allow interaction between the soil and vegetation canopy components. A simple algorithm to predict the net radiation partitioning between the soil and vegetation is introduced as part of the STSEB patch modelling scheme. The feasibility of the STSEB approach under a full range in fractional vegetation cover conditions is explored using data collected over a maize (corn) crop in Beltsville Maryland, USA during the 2004 summer growing season. Measurements of soil and canopy component temperatures as well as the effective composite temperature were collected over the course of the growing season from crop emergence to cob development. Comparison with tower flux measurements yielded root-mean-square-difference values between 15 and 50W m 61022 for the retrieval of the net radiation, soil, sensible and latent heat fluxes. A detailed sensitivity analysis of the STSEB approach to typical uncertainties in the required inputs was also conducted indicating greatest model sensitivity to soil and canopy temperature uncertainties with relative errors reaching 650230% in latent heat flux estimates. With algorithms proposed to infer component temperatures from bi-angular satellite observations, the STSEB model has the capability of being applied operationally.

DOI

[32]
Santini M, Caccamo G, Laurenti A et al., 2010. A multi-component GIS framework for desertification risk assessment by an integrated index.Applied Geography, 30(3): 394-415. doi: 10.1016/j.apgeog.2009.11.003.The implementation, performance of the methodology and benefits provided by the modelling approach to land management authorities for monitoring processes of land degradation are fully described in the paper.

DOI

[33]
Spatz D M, Taranik J V, Hsu L C, 1989. Differetiating volcanic rock assemblages using Landsat Thematic Mapper data-Influence of petrochemistry and desert varnish.Advances in Space Research, 9(1): 93-98. doi: 10.1016/0273-1177(89)90472-9.Composition, distribution, spectral properties, and Landsat thematic mapper (TM) imagery influences of desert varnish were investigated at three study sites in southern Nevada, U.S.A. Thematic mapper signatures of diverse volcanic rock assemblages were found to depend largely on primary petrochemical characteristics. Results indicate that desert varnish exerts minimal influence on TM imagery at longer wavelengths (bands 4–7) but absorbs higher frequency radiation (TM bands 1–3), leading to relatively high TM band 52 values and dark contrast on 52 images over units with a high rock-varnish albedo difference (RVAD). Highly evolved volcanic deposits exhibit steep positive spectral slopes in the TM band 5–7 region, resulting in low 57 values and dark tonal contrast on 57 images.

DOI

[34]
Sun W, Shao Q, Liu J, 2013. Soil erosion and its response to the changes of precipitation and vegetation cover on the Loess Plateau.Journal of Geographical Sciences, 23(6): 1091-1106. doi: 10.1007/s11442-013-1065-z.Soil erosion is a major threat to our terrestrial ecosystems and an important global environmental problem. The Loess Plateau in China is one of the regions that suffered more severe soil erosion and undergoing climate warming and drying in the past decades. The vegetation restoration named Grain-to-Green Program has now been operating for more than10 years. It is necessary to assess the variation of soil erosion and the response of precipitation and vegetation restoration to soil erosion on the Loess Plateau. In the study,the Revised Universal Soil Loss Equation(RUSLE) was applied to evaluate annual soil loss caused by water erosion. The results showed as follows. The soil erosion on the Loess Plateau between2000 and 2010 averaged for 15.2 t hm 2a 1and was characterized as light for the value less than 25 t hm 2a 1. The severe soil erosion higher than 25 t hm 2a 1was mainly distributed in the gully and hilly regions in the central,southwestern,and some scattered areas of earth-rocky mountainous areas on the Loess Plateau. The soil erosion on the Loess Plateau showed a deceasing trend in recent decade and reduced more at rates more than 1 t hm 2a 1in the areas suffering severe soil loss. Benefited from the improved vegetation cover and ecological construction,the soil erosion on the Loess Plateau was significantly declined,especially in the east of Yulin,most parts of Yan'an prefectures in Shaanxi Province,and the west of Luliang and Linfen prefectures in Shanxi Province in the hilly and gully regions. The variation of vegetation cover responding to soil erosion in these areas showed the relatively higher contribution than the precipitation. However,most areas in Qingyang and Dingxi prefectures in Gansu Province and Guyuan in Ningxia Hui Autonomous Region were predominantly related to precipitation.

DOI

[35]
Tao W, 2004. Progress in sandy desertification research of China.Journal of Geographical Sciences, 14(4): 387-400. doi: 10.1007/BF02837482.

DOI

[36]
Tereshchenko I, Zolotokrylin A N, Titkova T B et al., 2012. Seasonal variation of surface temperature-modulating factors in the Sonoran Desert in northwestern Mexico.Journal of Applied Meteorology and Climatology, 51(8): 1519-1530. doi: 10.1175/JAMC-D-11-0160.1.ABSTRACT The authors explore a new approach to monitoring of desertification that is based on use of results on the relation between albedo and surface temperature for the Sonoran Desert in northwestern Mexico. The criteria of predominance of radiation by using the threshold value of Advanced Very High Resolution Radiometer (AVHRR) and Moderate Resolution Imaging Spectroradiometer (MODIS) normalized difference vegetation index (NDVI) were determined. The radiation mechanism for regulating the temperature of the surface and the definition of threshold values for AVHRR and MODIS NDVI have an objective justification for the energy budget, which is based on the dominance of radiation surface temperature regulation in relation to evapotranspiration. Changes in the extent of arid regions with AVHRR NDVI of <0.08 and MODIS NDVI of <0.10 can be considered to be a characteristic in the evolution of desertification in the Sonoran Desert region. This is true because, in a certain year, the time span of the period when radiation factor predominates is important for the desertification process.

DOI

[37]
Tucker C J, Dregne H E, Newcombe W W, 1991. Expansion and contraction of the Sahara desert from 1980 to 1990.Science, 253: 299-301. doi: 10.1126/science.253.5017.299.Data from polar-orbiting meteorological satellites have been used to determine the extent of the Sahara Desert and to document its interannual variation from 1980 to 1990. The Sahara Desert ranged from 8,633,000 square kilometers in 1980 to 9,982,000 square kilometers in 1984. The greatest annual north-south latitudinal movement of the southern Saharan boundary was 110 kilometers from 1984 to 1985 and resulted in a decrease in desert area of 724,000 square kilometers.

DOI PMID

[38]
USDA-NRCS, 1998. Estimating soil moisture by feel and appearance. Program Aid No.1619. USDA-NRCS, Washington, DC.ABSTRACT Prior to the collection of samples for estimating soil moisture, the producer must determine the soil type, texture and available water holding capacity of each layer sampled. Soil texture, which is the relative amounts of sand, silt, and clay contained in soil, plays an important role in determining the amount of water a soil will hold. The portion of water in the soil the can be readily used by plants is the available water capacity (AWC) of the soil. The AWC ranges shown in Table - 1 for various textural groups may be used as a guide in estimating soil moisture. Soil maps, soil texture, and AWC for each soil type can be found in a published Soil Survey Report available through the USDA, Natural Resources Conservation Service office in each county or state.

[39]
Ustin S L, Palacios-Orueta A, Whiting M L et al., 2009. Remote sensing based assessment of biophysical indicators for land degradation and desertification. In: Röder A, Hill J (eds.). Tecent Advances in Remote Sensing and Geoinformation Processing for Land Degradation Assessment. London: CRC Press, 15-44ABSTRACT Desert ecosystems spanning moisture conditions from dry grasslands to barren hyper-arid landscapes are the largest terrestrial biome with more than 40% of the terrestrial landmass. Remote sensing data provide an efficient cost-effective means to assess biophysical indicators of land degradation and desertification, providing that essential ecosystem properties can be monitored. We review the spectral characteristics of plants and soils that are detectable using optical sensors and methods to identify and quantify properties that have potential for monitoring arid ecosystem processes. Vegetation indexes have little sensitivity at low leaf area, particularly when the soil background is highly variable, as is characteristic of many arid systems. Additionally, accumulated dry plant material on the soil surface challenges measurement. Although the absorption characteristics of the major biochemical constituents of plants and soils are generally understood, the methods to retrieve this information from reflectance data and to understand the significance of how the structural organization alters the absorption features remains an area of active research. The overlapping absorption features of plants and soils preclude direct assessment of many biogeochemicals of interest. New biophysical methods that take the full spectral shape into account, including the effect of one compound on the spectral absorption of another, are needed to reduce uncertainty in their estimates. As a result, despite significant progress in developing fundamental understanding of ecosystem processes and optical properties, more research is needed before fully predictable quantitative methods are available.

[40]
Xu D, Li C, Zhuang D et al., 2011. Assessment of the relative role of climate change and human activities in desertification: A review.Journal of Geographical Sciences, 21(5): 926-936. doi: 10.1007/s11442-011-0890-1.Climate change and human activities are the two kinds of driving forces in desertification,and assessing their relative role in desertification is of great significance to deeply understanding the driving mechanisms and preventing desertification expansion.This paper has systematically reviewed the progress of the researches on assessing the relative role of climate change and human activities in desertification from qualitative,semi-quantitative and quantitative aspects respectively.The authors found that there were still some problems in the previous researches.For example,the subjectivity in assessment was obvious,the assessment cannot be easily repeated,and the assessment and its results were always based on administrative regions and less taken and expressed in a continuous space.According to the progress of previous researches and the works conducted by the authors recently,we put forward a quantitative approach by selecting NPP as a common indicator to measure the relative role of climate change and human activities in desertification and dividing the ecological process of "driving force effect-dynamic response of desertified land" into several scenarios.Meanwhile,validation and scale of assessment should be taken into account when quantitative assessment of the relative role of climate change and human activities in desertification are carried out.

DOI

Options
Outlines

/

〈 〉