land
Article
Assessment of Land Desertification in the Brazilian East Atlantic
Region Using the Medalus Model and Google
Earth Engine
Theilon Henrique de Jesus Macêdo 1 , Cristiano Tagliaferre 2 , Bismarc Lopes da Silva 3, *, Alessandro de Paula 2 ,
Odair Lacerda Lemos 2 , Felizardo Adenilson Rocha 4 , Rosilene Gomes de Souza Pinheiro 3
and Ana Carolina Santos Lima 3
Citation: Macêdo, T.H.d.J.; Tagliaferre,
C.; Lopes da Silva, B.; de Paula, A.;
Lacerda Lemos, O.; Adenilson Rocha,
F.; Gomes de Souza Pinheiro, R.;
Santos Lima, A.C. Assessment of Land
Desertification in the Brazilian East
Atlantic Region Using the Medalus
Model and Google Earth Engine. Land
2024, 13, 31. https://doi.org/
10.3390/land13010031
Academic Editor: Adrianos Retalis
Received: 3 September 2023
Revised: 2 October 2023
Accepted: 5 October 2023
Published: 26 December 2023
Copyright: © 2023 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
Land 2024, 13, 31. https://doi.org/10.3390/land13010031
1 Forest Sciences Graduate Program, State University of Southwestern Bahia, Vitória da Conquista 45029-000,
Bahia, Brazil
2 Department of Agricultural Engineering and Soils (DEAS), State University of Southwestern Bahia,
Vitória da Conquista 45029-000, Bahia, Brazil; apaula@uesb.edu.br (A.d.P.);
odairlacerda@hotmail.com (O.L.L.)
3 Graduate Program in Agronomy, State University of Southwestern Bahia, Vitória da Conquista 45029-000,
Bahia, Brazil
4 Federal Institute of Bahia, Vitória da Conquista 45029-000, Bahia, Brazil; felizardoar@hotmail.com
* Correspondence: bismarctra@gmail.com; Tel.: +55-779-9917-0190
Abstract: Many factors drive land desertification, especially in arid and semi-arid regions. However,
the sheer number of these driving factors of desertification makes analyses computer-intensive.
Cloud computing offers a solution to address this problem, especially in developing countries. The
objective of this work was to assess the sensitivity of the East Atlantic Basin, Brazil, to desertification
using the Mediterranean Desertification and Land Use (MEDALUS) model and Google Earth Engine
(GEE). The model is composed of four environmental Quality Indices (QIs) associated with soil
(SQI), vegetation (VQI), climate (CQI), and management (MQI), each encompassing factors that
influence the desertification process. Digital databases corresponding to these factors were pre-
processed and uploaded to GEE for analysis. We report Environmentally Sensitive Areas (ESAs) and
Environmentally Critical Factors (ECF) maps of the East Atlantic Basin, which show that most of the
basin is in either a critical (49.4%) or fragile (35.7%) state of sensitivity. In contrast, only a smaller
portion of the area is unaffected (5%) or potentially affected (10.1%). The analysis also revealed an
inverse correlation between desertification sensitivity and the presence of vigorous vegetation. A
joint evaluation of ESAs and ECF shed light on the importance of each factor in the sensitivity to
desertification. The East Atlantic Basin shows a high degree of sensitivity to desertification, thereby
demanding more attention and the establishment of measures to mitigate the negative impacts of the
desertification process.
Keywords: environmental sensitivity; cloud computing; GIS; land degradation; Brazil
1. Introduction
Among the several desertification definitions coined since the 1970s, a common factor
is the adverse environmental process that results in desert-like conditions [1]. Desertifi-
cation is a global land degradation phenomenon, affecting 6% (2.7 million km2 ) of the
44.5 million km2 of drylands [2]. The Aridity Index (AI), determined by the ratio of precip-
itation to evapotranspiration, is commonly used to classify a region as a dryland when the
AI is less than 0.65 [3]. In Brazil, drylands predominantly exist in the northeastern region
of the country and have been closely associated with extensive droughts and land degrada-
tion [2,4]. Despite some controversies arising from disagreements between scientific and
policy discourses [1,5], land desertification process remains a crucial topic, particularly in
increasing impacts from agricultural production, deforestation, and climate change [6–8].
https://www.mdpi.com/journal/land
Land 2024, 13, 31 2 of 16

In addition to the climate, several soil and socioeconomic factors play important roles
in driving environmental degradation, especially in areas located in arid and semi-arid
regions [9]. A multi-disciplinary understanding of desertification drivers is essential for
implementing monitoring systems to assess land degradation and for developing early
warning systems that allow timely mitigation of negative environmental impacts [10,11].
Remote sensing techniques have enabled large-scale, cost-effective monitoring in
environmentally sensitive areas [12–14]. Among desertification monitoring methods, a
relatively common approach involves the use of multi-disciplinary techniques to define
Environmentally Sensitive Areas (ESAs) [15]. Overall, this approach integrates qualitative
and quantitative factors—encompassing soil and climatic characteristics, as well as human
influence on a region—to assess an area’s susceptibility to desertification [16]. A well-
studied multi-disciplinary approach is the Mediterranean Desertification and Land Use
(MEDALUS) model, which has been conceived from the parameterization of environmental
factors driving the desertification process and is recommended due to its simplicity [17].
The model integrates climatic, soil, land management, and vegetation factors [18].
The MEDALUS model, initially proposed for Mediterranean Europe [19], has been
used in other regions of the world to assess land desertification due to the universal
characteristics of the model‘s parameters [20]. The application of this model provided
a better understanding of degradation processes in the Peruvian Amazon [16], revealed
regions sensitive to land degradation, and pinpointed management quality as the main
driver in Ethiopia [21]. It assessed the rapid increase in desertification sensitivity in
Mongolia between 2003 and 2008 [22], was adapted to map desertification severity in
Egypt [23], detected desertification risk trends in Greece over 45 years using enhanced
ESAs MEDALUS methodology [24], and underwent a significant update to provide a global
assessment and mapping of land degradation and desertification threats, further proving
its versatility for both regional and global scales [25].
In Brazil, the application of the MEDALUS model has shown good performance in
mapping areas susceptible to desertification in the Northeast region [26]. This region
is mainly characterized by a semi-arid climate and vast extensions of degraded land,
resulting from decades of inadequate land use, such as slash-and-burn, overgrazing, and
overexploitation of natural resources [27]. However, limited computer power has been
a major limitation to studies analyzing the many variables involved in desertification
and larger regions; therefore, studies have generally focused on small areas to reduce
computational requirements [28]. The high computer power necessary to analyze large
amounts of data can be addressed by cloud computing platforms, such as Google Earth
Engine, which allow convenient and effective data processing, analysis, and storage [29–32].
In this study, we used the MEDALUS model to assess the susceptibility to desertifi-
cation of the East Atlantic Basin, Brazil, using Google Earth Engine (GEE) in 2019. This
model has been widely used in the Mediterranean regions due to its flexibility and rapid
implementation strategy. Furthermore, it allows for an in-depth understanding of the
parameters causing the desertification threat at a certain location by delivering accurate
results and being easily implemented in arid areas [25]. The objectives were to (i) assess
the sensitivity of the East Atlantic Basin to desertification and (ii) evaluate the influence of
each component of the MEDALUS model on its overall performance.

2. Materials and Methods

2.1. Study Site and Database
The East Atlantic Basin (Figure 1) is located within the states of Bahia, Espírito Santo,
Minas Gerais, and Sergipe, with a total area of approximately 386,511.95 km2 . The East
Atlantic Basin has four distinct climate regimes (arid, semi-arid, humid, and sub-humid), a
strong incidence of solar radiation, altitudes ranging from sea level to as high as 1100 m [33],
and average annual precipitation rates ranging from 600 and 2200 mm [34]. Due to its
diversity in climate, relief, and vegetation, the basin is an interesting choice for land
desertification assessment and for testing the GEE processing capacity. This basin spans
 the spatial domains of two distinct biomes: Caatinga, characterized by its arid condition
and unique vegetation, and Atlantic Forest, known for its wetter climate and lush forests
Identifying and measuring the impacts resulting from the occurrence o
desertification in regional ecosystems is an effective method of evaluating the effects o
Land 2024, 13, 31 3 of 16
sustainable soil management in rural communities [35]. In this scenario, the selected stud
area, a basin situated in a semi-arid region, offers a unique landscape characterized by
gradient of rainfall. Evaluating desertification against this backdrop provides insights int
the spatial domains of two distinct biomes: Caatinga, characterized by its arid conditions
the unique
and impacts of climatic
vegetation, andshifts on such
Atlantic Forest,ecosystems.
known for its wetter climate and lush forests.

Figure1.1.Location
Figure Location map
map of the
of the EastEast Atlantic
Atlantic Basin.Basin.

Identifying and measuring the impacts resulting from the occurrence of desertification
Digital geospatial databases were used to map the areas sensitive to desertificatio
in regional ecosystems is an effective method of evaluating the effects of sustainable soil
in the East Atlantic
management in rural Basin. Some of
communities these
[35]. data
In this underwent
scenario, the removal
the selected of topological
study area, a basin error
and the addition of population data. These adjustments were made using
situated in a semi-arid region, offers a unique landscape characterized by a gradient the open-sourc
of
software
rainfall. QGIS, version
Evaluating 3.14. Subsequently,
desertification these files
against this backdrop were insights
provides uploadedintotothe
theimpacts
GEE platform
of climatic shifts on such ecosystems.
Digital geospatial databases were used to map the areas sensitive to desertification in
the East Atlantic Basin. Some of these data underwent the removal of topological errors
and the addition of population data. These adjustments were made using the open-source
software QGIS, version 3.14. Subsequently, these files were uploaded to the GEE platform.
The research data for this study primarily consist of geospatial and environmental
datasets that enable the comprehensive application of the MEDALUS model in the Brazilian
East Atlantic Region. The datasets include variables related to soil quality, vegetation
condition, climatic factors, and land management practices, among others. These variables
are meticulously processed and analyzed using advanced geoprocessing techniques on
Land 2024, 13, 31 4 of 16

the GEE platform. Given the limited research previously conducted in this region, our
research data serves as one of the pioneering datasets that will hopefully stimulate further
scientific investigations. The data aims to offer a multifaceted view of land desertification
factors, thereby providing a robust foundation for the Quality Indices (QIs) employed in
our model.
Before geospatial classification, data preprocessing was conducted on the acquired
datasets. Climate determinants were derived from averaged precipitation and potential
evapotranspiration data spanning a decade, leading to the computation of the aridity
index. Terrain features, including slope and orientation, were deduced from the SRTM
digital elevation model. Using land use and land cover data, three risk indices were
produced, while the NDVI from Sentinel-2 imagery informed vegetation health assessments.
To maintain uniformity across the files, all datasets were projected to the EPSG: 31984
coordinate system. Such preprocessing is indispensable in spatial analyses, ensuring data
accuracy and consistency. It rectifies raw data discrepancies, harmonizes diverse datasets,
and contextualizes data, bolstering the reliability and depth of ensuing analyses.

2.2. The MEDALUS Model

The Environmentally Sensitive Area (ESA) index was based on the modified frame-
work of the MEDALUS model [19], which uses four quality indices (QI): climate quality
index (CQI), soil quality index (SQI), vegetation quality index (VQI), and management qual-
ity index (MQI). The framework has variations from that of the original method, such as
the addition of information on Soil Carbon Content and Normalized Difference Vegetation
Index (NDVI) (Table 1).

Table 1. Variables with respective weights used to compose the Quality Indices.

Drought Erosion Land Use

Description Fire Risk Drainage Quality Weight
Resistance Protection Intensity
Forest Training 1.7 1.1 1.0 1.0 Excessive 2.0
Savanna Formation 1.8 1.1 1.2 1.0 Slightly Excessive 1.7
Mangrove 1.2 1.0 1.1 1.0 Imperfect 1.4
Planted Forest 1.8 1.2 1.2 1.4 Moderately good 1.2
Flooded Field and Swampy Area 1.2 1.0 1.2 1.0 Good 1.0
Field Training 1.3 1.4 1.3 1.1
Other Non-Forest Formations 1.5 1.5 1.4 1.1
Pasture 1.5 1.6 1.4 1.7 NDVI Weight
Cane 1.6 1.5 1.5 1.5 <0.25 2.0
Mosaic of Agriculture and Pasture 1.5 1.6 1.6 1.5 0.25–0.32 1.8
Beach and Dune 1.0 2.0 2.0 2.0 0.32–0.40 1.5
Urban Infrastructure 1.0 2.0 2.0 2.0 0.40–0.50 1.3
Other Non-Vegetated Areas 1.0 2.0 2.0 2.0 >0.50 1.0
Not observed 1.0 1.0 1.0 1.0
Rocky Formation 1.0 2.0 1.0 1.0
Mining 1.0 2.0 2.0 2.0 Demography (inhab/ha) Weight
Aquaculture 1.0 1.0 1.8 2.0 >400 2.0
Apicum 1.0 1.5 1.7 1.3 400–200 1.8
River, Lake, and Ocean 1.0 1.0 2.0 1.0 200–100 1.6
Perennial Crop 1.5 1.4 1.3 1.5 100–50 1.4
Soybean 1.4 1.7 1.6 1.7 50–25 1.2
Other Temporary Crops 1.3 1.5 1.7 1.6 <25 1.0
Land 2024, 13, 31 5 of 16

Table 1. Cont.

Soil type Weight Precipitation Weight

Rocks, Luvisols, Planosols 2.0 <280 2.0 Carbon Content Weight
Argisols, Gleissolos, Neosols 1.7 280–650 1.5 <0.2 2.0
Cambisols, Chernosols, Spodosols 1.3 >650 1.0 0.2–0.6 1.5
Oxisols, Vertisols 1.0 0.6–1.2 1.3
Aridity
Weight 1.2–2 1.2
Index
Albedo Weight <0.5 2.0 >2 1.0
>0.25 2.0 0.5–0.65 1.5
0.25–0.2 1.5 >0.65 1.0
<0.2 1.0 Soil Texture Weight
Inclination
Weight s 2.0
(%)
Orientation Weight >35 2.0 C(h); SiC; C(l); yes 1.6
NW-NE 2.0 18–6 1.5 SiCL; SiL; SC 1.2
SW-SE 1.0 <6 1.0 CL; L; SCL; SL; LS 1.0

Driving factors of the desertification process were classified according to scores ranging
from 1 (least sensitive to desertification) to 2 (most sensitive to desertification) for the
determination of quality indices (QIs); values between 1 and 2 indicate intermediate
susceptibility. Thus, obtaining the ESAs took place in two stages: in the first stage, the QIs
were obtained using the geometric mean of the variables that compose the indices using
Equations (1) and (2) [19].

QIx = (variable_1ij × variable_2ij × variable_3ij . . . × variable_nij )1/n (1)

where: n = number of variables; x = quality indices (CQI, SQI, VQI, and MQI); iej = row
and column of a pixel of the used rasters, respectively.
In the second stage, the previously obtained QIs were used to calculate the ESA:

ESAij = (CQIij × SQIij × VQIij × MQIij )1/4 (2)

where: i and j = is the row and column, respectively, of a pixel of the rasters.
Thus, ESAs obtained from the listed QIs were classified into eight desertification
sensitivity classes [24]: unaffected (N), potentially affected (P), fragile (F1, F2, F3), and
critical (C1, C2 and C3) (Table 2).

Table 2. Types of Desertification Sensitivity Indices.

Susceptibility Weights Subclasses

Not affected <1.170 No
Potentially affected 1.170–1.225 P
1.225–1.275 F1
Fragile 1.275–1.325 F2
1.325–1.375 F3
1.375–1.425 C1
Critical 1.425–1.530 C2
>1.530 C3

2.3. Quality Scores

The respective QIs were obtained using digital databases in raster format. While some
of these databases were already available in the GEE catalog, those not available on the
 Land 2023, 12, x FOR PEER REVIEW 6 of 17

Land 2024, 13, 31 6 of 16

The respective QIs were obtained using digital databases in raster format. While
some of these databases were already available in the GEE catalog, those not available on
platform werewere
the platform acquired and reclassified
acquired according
and reclassified to the
according to weights (Table
the weights 1) established
(Table in
1) established
the
in methodology. Subsequently,
the methodology. they were
Subsequently, they uploaded to the GEE
were uploaded platform
to the for the creation
GEE platform for the
ofcreation
the ESAs
of map (Figure
the ESAs map2).(Figure 2).

Figure 2. Flowchart showing the components associated with climate (CQI), soil (SQI), vegetation
(VQI) and management (MQI) quality indexes to calculate the Environmental Sensitive Area (ESA).
Figure 2. Flowchart showing the components associated with climate (CQI), soil (SQI), vegetation
R—rainfall; Ar—aridity; As—basin orientation; St—soil type; Sl—basin slope; Tx—soil texture;
(VQI) and management (MQI) quality indexes to calculate the Environmental Sensitive Area (ESA).
Dd—drainage classes; OC—organic carbon content; Al—albedo; FR—fire risk; DR—drought re-
R—rainfall; Ar—aridity; As—basin orientation; St—soil type; Sl—basin slope; Tx—soil texture;
sistance; EP—erosion
Dd—drainage protection;
classes; NDVI—normalized
OC—organic difference
carbon content; vegetation
Al—albedo; FR—fireindex;
risk;LU—land use;
DR—drought
PD—population density.
resistance; EP—erosion protection; NDVI—normalized difference vegetation index; LU—land use;
PD—population density.
CQI was determined using the variables rainfall (R), aridity (Ar), and basin orientation
(As), based on thedetermined
CQI was GEE catalogusing
databases. The mean rainfall
the variables annual total
(R), precipitation
aridity (Ar),information
and basin
(mm) was derived from the TerraClimate global dataset with a spatial
orientation (As), based on the GEE catalog databases. The mean annual resolution of approx-
total precipitation
imately 4 km [36].
information (mm)Aridity was assessed
was derived using
from the the UNESCO
TerraClimate aridity
global index with
dataset [37], which is
a spatial
the ratio between precipitation (mm) and potential evapotranspiration (mm). The
resolution of approximately 4 km [36]. Aridity was assessed using the UNESCO aridity data on
potential evapotranspiration for the period 2010 to 2019, essential for obtaining Ar, was
index [37], which is the ratio between precipitation (mm) and potential evapotranspiration
sourced from the Terra Net Evapotranspiration digital database. This database utilizes
(mm). The data on potential evapotranspiration for the period 2010 to 2019, essential for
MODIS satellite images, which have a spatial resolution of 500 m and a temporal resolution
obtaining Ar, was sourced from the Terra Net Evapotranspiration digital database. This
of 8 days, provided by the Land Processes Distributed Active Archive Center (LPDAAC).
database utilizes MODIS satellite images, which have a spatial resolution of 500 m and a
The orientation of the basin was determined using SRTM images from the hole-filled SRTM
for the globe, version 4, with a spatial resolution of 90 m.
Land 2024, 13, 31 7 of 16

SQI was derived from several variables, including soil type (St), basin slope (Sl), soil
texture (Tx), drainage classes (Dd), organic carbon content (OC), and albedo (Al). The
basin slope was determined using the previously cited SRTM data. Details on soil texture,
drainage classes and organic carbon content were sourced from the Harmonized World Soil
Database, which has a spatial resolution of 1 km [38]. After reclassification based on the
predetermined weights, this information was uploaded to the GEE platform. The surface
albedo data was acquired from an annual mosaic created for the year 2019, composed with
pre-processed images from the Sentinel-2 (S2). These images have a spatial resolution of
10 m and a temporal resolution of 10 days. The albedo was calculated from the S2 images
using coefficients defined in an approach proposed by [39].
VQI is derived from several sub-indices fire risk (FR), drought resistance capacity (DR),
erosion protection capacity (EP) and the NDVI. The FR, DR and EP variables were sourced
by reclassifying (Table 1) an annual land use mosaic made available by the Mapbiomas
project, which boasts a spatial resolution of 30 m [27], from the year 2019. This image,
available on GEE, was subsequently cropped and reclassified according to the parameters
established for each variable that composes the VQI. The NDVI for the study area was
calculated using the mosaic of S2 images using the following equation:

NDVI = (IR − RED)/(IR − RED) (3)

where IR is the infrared surface reflectance on band eight; RED is the red surface reflectance
on band four of the S2 satellites.
MQI was composed of the variables intensity of land use (LU) and population density
(PD) of the municipalities within the study basin. The land use intensity data were sourced
from the land cover classification mosaic, with scores designated to each class as outlined
in Table 1. The PD variable was derived from the ratio of a municipality’s population to its
area in km2 . Given that the most recent population census in Brazil took place in 2010, it
was necessary to use the data for the projected 2020 population.
The indicators employed in this study were selected based on two primary criteria:
(i) the significance and relevance of each indicator to the study area; for example, due to the
region’s semi-arid climate, indicators directly associated with the area’s rainfall dynamics
were given priority; (ii) emphasis was placed on indicators for which consistent and reliable
data were readily available. This not only ensured the robustness of our findings but
was also imperative due to the limited research available on the basin dynamics of this
region. Given the paucity of studies concerning the climate characteristics of the basin in
this region, it was essential to adjust the parameter values sourced from other research:
R [19,25], Ar [26], As [18], St [26,40], Sl [19], Tx [19], Dd [25], OC [35], Al [36], FR [19,29],
DR [19,25], EP [19,25], NDVI [38], LU [19,25], and PD [25].
The maps based on the Environmentally Critical Factors (ECF) index were created to
complement the ESA map for the East Atlantic Basin. The composition of the ECF index
was performed with the following equation [25]:
15
∑n=1 f(variablenij


ECFij = (4)

1, if x > 1.425
f (x) = (5)
0, if x ≤ 1.425
Pixels with values below the critical threshold (C2) of 1.425 were excluded from the
processing. Thus, the composition of the ECF index was made from counting the number
of variables that exceeded the critical value of 1.425 for each pixel.

2.4. Statistical Analysis

Principal Component Analysis (PCA) was used on the matrix of variables to evaluate
the relationship between the component variables of the produced ESAs. The four QIs
and the ESA were used as supplementary parameters. The averages of the 20 variables
 2.4. Statistical Analysis
Principal Component Analysis (PCA) was used on the matrix of variables to evaluate
Land 2024, 13, 31 the relationship between the component variables of the produced ESAs. The four QIs 8 ofand
16
the ESA were used as supplementary parameters. The averages of the 20 variables (15
factors, Qis, and ESA) were obtained for each of the 541 municipalities located within the
limits
(15 of the
factors, Qis,basin’s area,were
and ESA) composing
obtainedoffor10,820 observations.
each of The PCA was
the 541 municipalities performed
located within
using the R software.
the limits of the basin’s area, composing of 10,820 observations. The PCA was performed
using the R software.
3. Results
3.3.1.
Results
Quality Scores
3.1. Quality Scores
The maps, generated using Equation (1) for the creation of QIs, are presented in
Figure 3.maps,
The Each generated usingpixel
raster contains Equation
values(1)ranging
for the from
creation
1 toof2.QIs, are presented
A higher value forina
Figure 3. Each raster contains pixel values ranging from 1 to 2. A higher value for a
particular index indicates a greater contribution to the basin’s sensitivity to desertification.
particular index indicates a greater contribution to the basin’s sensitivity to desertification.

Figure3.3.Spatial
Figure Spatial distribution
distribution of quality
of quality indices
indices in theinEast
the Atlantic
East Atlantic
Basin. Basin. CQI—climate
CQI—climate quality quality
index;
index; VQI—vegetation
VQI—vegetation qualitySQI—soil
quality index; index; SQI—soil qualityMQI—management
quality index; index; MQI—management quality index.
quality index.

Upon
Uponcloser
closerinspection
inspection of of
Figure 3, distinct
Figure spatial
3, distinct patterns
spatial emerge
patterns in thein
emerge QIs,
thehigh-
QIs,
lighting region-specific
highlighting triggers
region-specific for heightened
triggers sensitivity,
for heightened particularly
sensitivity, in the northern
particularly in the
regions.
northernFrom the From
regions. CQI map, the existing
the CQI map, theclimatic
existinggradient
climaticwithin thewithin
gradient basin becomes
the basin
apparent,
becomes revealing
apparent, potential
revealingmicro-climatic zones.
potential micro-climatic zones.
The
Thewestern
westernandandnorthern
northernparts
partsofofthe
thebasin
basinare
arecharacterized
characterizedby byaamore
morearidaridclimate.
climate.
However,
However, this area displays a notable presence of arboreal vegetation and lessintensive
this area displays a notable presence of arboreal vegetation and less intensive
land
landuse,
use,as
assuggested
suggestedby byVQI
VQIandandMQI.
MQI.Despite
Despitethetheconsiderable
considerablediversity
diversityininsoil
soilquality
quality
across the study area, the northwest side appears to be of lesser quality, as
across the study area, the northwest side appears to be of lesser quality, as indicated indicated by the
by
SQI. SomeSome
the SQI. regions also show
regions high high
also show vegetation quality
vegetation even even
quality with with
a lowaMQI.
low MQI.
The
Thespatial distribution
spatial distributionof the
of quality indicesindices
the quality (Figure(Figure
3) depicts3)a depicts
gradientaofgradient
sensitivityof
from
sensitivity from the southern to the northern parts of the basin. This patterninherent
the southern to the northern parts of the basin. This pattern hints at specific hints at
environmental or geomorphic characteristics influencing the desertification process.

3.2. Environmentally Sensitive Area (ESA)

Using the previously generated QIs and applying Equation (2), we produced an ESAs
map (Figure 4). The ESAs represent the geometric mean of the four QIs derived in the
initial phase of the study.
 3.2. Environmentally Sensitive Area (ESA)

Using the previously generated QIs and applying Equation (2), we produced an ESAs
Land 2024, 13, 31 map (Figure 4). The ESAs represent the geometric mean of the four QIs derived in the
9 of 16
initial phase of the study.

Figure
Figure 4.
4. Desertification
Desertification sensitivity
sensitivity map
map of
of the
the East
East Atlantic
Atlantic Basin
Basin in
in 2019.
2019.

In analyzing
analyzing the
the ESA
ESA map,
map, areas in the north and central-west
central-west regions
regions of of the
the study
study
area exhibit greater vulnerability to desertification. Conversely,
Conversely, areas located closer to the
coastal
coastal zone
zoneandandtotothe
thesouthwest
southwest of of
thethe
basin were
basin lessless
were susceptible.
susceptible.OnlyOnly
5.0% of theof
5.0% basin
the
area
basin(Table 3) can3)
area (Table becan
classified in the in
be classified “Not
the affected” category
“Not affected” (N), with
category (N),emphasis
with emphasis on areas
on
areas located
located in theinregion
the region further
further to theto the coast
coast andand southwest
southwest of of
thethebasin.
basin.ItItisisalso
also worth
mentioning when
mentioning when examining
examining the the ESA
ESA map
map that
that it
it is evident
evident that the central regions are
characterized as transition zones between low and high sensitivity areas.
The areas included in the “Potentially affected” (P) class comprise about 10.1% of the
3. Desertification
Tablearea
total susceptibility
of the basin and classes
are located, and their
as well as therespective
N class, areas
in thewithin
coastaltheandEast Atlantic
southern
Basin, Brazil.
regions. The three levels corresponding to the “Fragile” (F) class have a relatively uniform
spatial distribution throughout the basin, occupying around 49.3% of2 the entire area. In
Susceptibility Weights Subclasses Area (km ) Area (%)
general, subclasses F1 and F2 share similar spatial distribution to classes N and P. In
Not subclass
contrast, affected F3 is predominant
<1.170 No and central
in the northern 19,177.49
portion of the study 5.0% area,
potentially affected 1.170–1.225 P 38,883.42 10.1%
which aligns more closely with the “Critical” class (C). Class C corresponds to a portion
1.225–1.275 F1 56,033.22 14.5%
of 35.7%Fragile
of the basin and,1.275–1.325
as previously mentioned, F2 the areas included
68,675.25in this classification
17.8%
(C1, C2, C3) are located to the north and centerF3of the area under
1.325–1.375 study. The distribution
65,941.47 17.1%
of different desertification susceptibility classes,
1.375–1.425 C1 as shown54,968.20 in Table 3, highlights14.2% a
Critical 1.425–1.530 C2 70,176.42 18.2%
>1.530 C3 12,656.84 3.3%

The areas included in the “Potentially affected” (P) class comprise about 10.1% of the
total area of the basin and are located, as well as the N class, in the coastal and southern
regions. The three levels corresponding to the “Fragile” (F) class have a relatively uniform
spatial distribution throughout the basin, occupying around 49.3% of the entire area. In
general, subclasses F1 and F2 share similar spatial distribution to classes N and P. In
contrast, subclass F3 is predominant in the northern and central portion of the study area,
which aligns more closely with the “Critical” class (C). Class C corresponds to a portion of
 Susceptibility Weights Subclasses Area (km2) Area (%)
Not affected <1.170 No 19,177.49 5.0%
potentially affected 1.170–1.225 P 38,883.42 10.1%
1.225–1.275 F1 56,033.22 14.5%
Land 2024, 13, 31 Fragile 1.275–1.325 F2 68,675.25 17.8%
10 of 16
1.325–1.375 F3 65,941.47 17.1%
1.375–1.425 C1 54,968.20 14.2%
35.7% ofCritical
the basin and, as previously
1.425–1.530 mentioned,C2 the areas included in
70,176.42 this classification
18.2%
(C1, C2, C3) are located to the north and center of the area under study. The distribution of
>1.530 C3 12,656.84 3.3%
different desertification susceptibility classes, as shown in Table 3, highlights a predominant
presence of the ‘Fragile’ and ‘Critical’ classes, covering nearly 85% of the basin, indicating a
In the north of the basin, when analyzing areas characterized by high sensitivity to
significant portion of the basin requires management intervention.
desertification, the presence of low-quality soils and high rates of aridity are predominant
In the north of the basin, when analyzing areas characterized by high sensitivity to
factors. A joint analysis of the ECF index (Figure 5) and ESA enhances our understanding
desertification, the presence of low-quality soils and high rates of aridity are predominant
of each model component’s contribution to the desertification process, pinpointing areas
factors. A joint analysis of the ECF index (Figure 5) and ESA enhances our understanding
where
of each multiple environmental
model component’s stresses to
contribution converge. This analysis
the desertification highlights
process, that areas
pinpointing even
certain areas with identical ESAs can exhibit variables that respond differently
where multiple environmental stresses converge. This analysis highlights that even certain to the
threat of desertification [25]. When analyzing the ECF maps, it becomes
areas with identical ESAs can exhibit variables that respond differently to the threat ofevident that
variables linked
desertification to the
[25]. Whenregion’s climate
analyzing and soil
the ECF maps,areitdominant determinants
becomes evident of sensitivity
that variables linked
to desertification in the East Atlantic Basin, underscoring the emergence of both
to the region’s climate and soil are dominant determinants of sensitivity to desertification soil and
climate
in factors
the East as pivotal
Atlantic Basin,ECFs. Moreover,the
underscoring theemergence
analysis reveals
of both potential
soil andregional
climateclusters
factors
as pivotal ECFs. Moreover, the analysis reveals potential regional clusters desertification
of heightened sensitivity, suggesting specific zones with advanced of heightened
vulnerability.
sensitivity, suggesting specific zones with advanced desertification vulnerability.

Figure 5.
Figure 5. The
Thenumber
numberofofenvironmentally
environmentally critical
critical factors
factors (ECF)
(ECF) in the
in the EastEast Atlantic
Atlantic Basin
Basin in 2019—
in 2019—ESA
ESAclimate
(A), (A), climate (B), vegetation
(B), vegetation (C),(D),
(C), soil soiland
(D),management
and management(E). (E).

The results from the PCA (Table 4 and Figure 6) provided a more detailed evaluation
of the contributions from each edaphoclimatic and anthropic variable utilized in creating
the ESAs maps for the East Atlantic Basin. This analysis paves the way for refining the
model in future research.
The PCA results identify three principal components that collectively account for
61.31% of the total variance. This considerable percentage suggests that these components
capture much of the data’s variability, providing key insights into the desertification
dynamics of the region. Principal component 1 (PC1), which explains 29.65% of the
variance, is associated with variables from VQI, including Fire Risk, Resistance to Drought,
and Resistance to Erosion, as well as MQI’s Land Use Intensity. PC1 was also associated
with albedo, which is not usually a primary factor in such analyses and showed a significant
correlation with ESAs in our study, indicating its under-explored potential in desertification
assessments, although it did not show a high correlation. Principal component 2 (PC2),
which accounts for 18.92% of the variance, correlated prominently with climatic variables
Land 2024, 13, 31 11 of 16

from the CQI, namely Rainfall and Aridity, and had an association with the NDVI. Lastly,
PC3, responsible for 12.75%, was directly tied to the SQI’s component variables: Soil
Texture, Drainage Classes, and Organic Carbon Content.

Table 4. Principal Components Analysis (PCA) of the variables composing the Environmentally
Sensitive Areas (ESA) index for the East Atlantic Basin.

Variables PC1 PC2 PC3

Precipitation 0.12 0.86 * 0.04
Aridity −0.15 0.89 * 0.00
Guidance −0.37 −0.17 0.14
Fire risk −0.82 * 0.34 −0.16
Drought resistance 0.91 * −0.17 0.14
NDVI 0.33 0.70 0.04
Erosion resistance 0.90 * 0.00 0.18
Soil texture −0.19 0.06 0.84 *
Drainage class −0.13 0.12 0.76 *
Albedo 0.61 * 0.35 −0.17
Soil type 0.40 0.01 −0.02
Slope −0.52 −0.51 −0.04
OC content −0.15 −0.11 0.69 *
LU intensity 0.89 * −0.13 0.11
Population density 0.41 −0.46 −0.08
CQI −0.10 0.92 * 0.03
Land 2023, 12, x FOR PEER REVIEW VQI 0.75 * 0.46 0.11 11 of 17
SQI −0.06 0.02 0.88 *
MQI 0.76 * −0.39 0.00
ESA 0.51 0.55 0.43
TheEigenvalues
results from the PCA (Table4.45 4 and Figure 6) provided
2.84 a more detailed1.91
evaluation
of the contributions
Variance (%) from each edaphoclimatic
29.65 and anthropic
18.92variable utilized12.75
in creating
Accumulated
the ESAs maps variance
for the(%)
East Atlantic29.65
Basin. This analysis48.56
paves the way for 61.31
refining the
*model
Indicatesinfactor loadings
future > 0.6.
research.

Figure6.
Figure 6. Principal
Principal Component
Component Analysis
Analysis (PCA)
(PCA) for
for the
the component
component variables
variables of
of the
the Environmentally
Environmentally
Sensitive Area (ESA) of the East Atlantic Basin.
Sensitive Area (ESA) of the East Atlantic Basin.

TableBy4. analyzing
Principal Components
the resultsAnalysis
in Figure(PCA)
6, weof the
canvariables
have a composing the Environmentally
better understanding of the
Sensitive Areas (ESA) index for the East Atlantic Basin.
relationship between the selected variables, QIs, and ESAs, showing that NDVI, albedo,
and VQI are positively correlated with ESAs.
Variables PC1 PC2 PC3
Precipitation 0.12 0.86 * 0.04
Aridity −0.15 0.89 * 0.00
Guidance −0.37 −0.17 0.14
Fire risk −0.82 * 0.34 −0.16
Drought resistance 0.91 * −0.17 0.14
Land 2024, 13, 31 12 of 16

4. Discussion
Analyzing satellite images has often been hindered by limited access to high comput-
ing power. The challenge is particularly pronounced in land desertification assessments
given the large amount of data involved in the model creation and assessment, such as the
MEDALUS. However, our results demonstrate the feasibility of assessing desertification
sensitivity in extensive areas, such as the East Atlantic Basin.

4.1. Quality Indices (QIs)

The CQI exhibits a relatively well-defined spatial distribution. Specifically, regions to
the east of the basin, characterized by having lower altitudes and proximity to the coastal
region, are recognized as areas of better climate quality. This characteristic is often attributed
to the substantial rainfall in the region, which can exceed 2000 mm in total per year [34].
However, regions close to Chapada Diamantina, in the state of Bahia, are an exception to
this spatial distribution. These areas, especially those near prominent conservation units
like the Chapada Diamantina National Park and the Marimbus/Iraquara Environmental
Protection Area, exhibit high CQIs. The presence of fauna and flora conservation areas
works as a mechanism for maintaining and conserving environmental resources, with a
particular emphasis on water resources.
From a climatic perspective, the areas located to the west and north of the basin
stand out as the most sensitive. These drier regions have a semi-arid climate, with the
annual aridity index reaching a value of 0.2 and are distributed across northeastern Agreste
and Sertão [26]. These areas are notably associated with low annual average rainfall
(around 600 mm) and higher altitudes in comparison to the broader East Atlantic Basin,
consequently being exposed to a greater incidence of solar energy [29].
The vegetation cover within the study displays a diverse spatial distribution, predom-
inantly consisting of pastures, agricultural lands, and forests. Regarding the VQI, one can
notice the occurrence of patches or zones of high environmental sensitivity stretching from
the northernmost to the southernmost parts of the basin. These patches correspond to areas
with arid characteristics and minimal rainfall, such as northern Minas Gerais, western
Sergipe, south-central and northern Bahia. These regions also host significant agricultural
activities, notably horticulture and cattle raising, which have been expanding in recent
decades [26].
With a pattern similar to that observed for CQI, nearly the entire coastal region of
the basin, located within the Atlantic Forest biome, shows high vegetation quality, except
in densely urbanized and populated areas. Across the entire eastern side of the basin,
agricultural and forestry activities are predominant. Forestry, in particular, serves as a
primary economic driver in this region, largely due to eucalyptus plantations and the
pulp and paper industry. Therefore, the presence of vigorous vegetation cover, whether
cultivated or native, tends to protect the soil against erosion and offer greater resistance to
water scarcity when compared to areas with less healthy vegetation cover [18,41].
Based on the characteristics used to construct the SQI, most of the basin’s soils range
from medium to good quality. The soils generally have good drainage and texture classes.
Additionally, a significant portion of the area has low albedo values (<0.20), indicating a
minimal bare surface presence. The albedo is indicative of the presence of the forest cover
removal process in semi-arid regions; high reflectance values might hint at the presence
of environmental degradation and increased sensitivity to desertification, especially in
regions with drier climate regimes [42]. The areas in the basin with the lowest SQI are
located in regions that have sandy soils (S), which are, by nature, highly vulnerable to the
erosion process and have a low water retention capacity [20]. These lower-quality areas are
found in the northeast, northwest, and south-central portions of the East Atlantic Basin.
Compared to the other quality indices, SQI has low temporal variability, especially the
variables St, Sl, Tx, and Dd, which take a considerable period to change their characteristics.
Thus, when establishing a continuous desertification monitoring program using cloud
Land 2024, 13, 31 13 of 16

processing applications supported by edaphoclimatic variables, the soil parameters require

only punctual updates in their configurations.
The MQI of the East Atlantic Basin shows, in the western portion, the areas with
lower environmental susceptibilities. These areas, located further inland on the continent,
are characterized by having low population density rates (<100 inhabitants/km2 ). The
western areas of the basin also exhibit low intensity of land use, particularly concerning
grazing, suggesting a tendency to maintain natural vegetation cover, especially the savanna
forest formation [26]. Additionally, these regions have experienced minimal soil distur-
bance, although this trend has been shifting in recent years, primarily for the expansion of
pastures [43].
Municipalities located closer to the coastal zone of the basin have historically shown
higher population densities (>100 inhabitants/km2 ). The high rainfall that occurs in most of
these regions makes them suitable for the development of rural activities, as in the extreme
south of Bahia [26].

4.2. Environmentally Sensitive Areas (ESAs)

Overall, in the East Atlantic Basin, sensitivity to desertification is closely associated
with the distribution of rainfall and the presence of vigorous vegetation. A considerable
portion of the western face of the basin, near the protection areas in Chapada Diamantina,
shows reduced susceptibility to desertification. This indicates that such areas contribute to
the mitigation of impacts and desertification expansion on these ecosystems. In contrast,
areas to the north of the basin, which are characterized by high sensitivity to desertifica-
tion, have predominantly low-quality soils and high rates of aridity. Regions with high
potential evapotranspiration rates and low precipitation rates tend to be more susceptible
to degradation from desertification [44,45].
Both VQI and MQI jointly contributed to the definition of levels of sensitivity to deser-
tification. This relationship is likely due to the Land Use variable being derived from the
basin’s vegetation classification. Conversely, SQI and CQI shaped the basin’s environmental
sensitivity independently, without a synergistic effect with the other quality indices. These
findings diverge from other studies, in which the CQI and VQI had a correlated influence
on the desertification classifications while the MQI operated separately [40]. Nevertheless,
in both scenarios, the SQI had an autonomous contribution compared to the other indices.
Using the modified MEDALUS model framework, we effectively identified areas
within the East Atlantic Basin that are vulnerable to desertification. The model offers
valuable insights for shaping public policies and measures against the environmental
degradation impacts resulting from desertification. The creation of online and universally
accessible tools for monitoring and assessment of desertification is necessary to help mit-
igate the impacts of this phenomenon in regions with few economic resources. Hence,
facilitated access to geoprocessing technologies with high computational capacity for
public institutions may significantly help and improve decision-making strategies. For
methodological improvements, assessing and evaluating the contribution of additional
socio-ecological factors in the emergence of environmental degradation processes due to
desertification is crucial. Such enhancements can potentially be integrated into future itera-
tions of the MEDALUS model [25], contingent upon the development and dissemination of
up-to-date databases regarding these parameters.
Spectral indices, such as the NDVI, are effective tools for monitoring and understand-
ing environmental degradation due to desertification in arid regions. Conversely, it is
possible to observe the existence of a negative correlation between slope and sensitivity
to desertification. The slope plays a pivotal role in analyzing the existence of degradation
and desertification processes in an area, primarily because it is instrumental in the onset
of erosion processes [19]. Just as with NDVI, slope can be used to monitor areas under
environmental pressure and degradation processes [42]. The identification of the existence
of a correlation between ESA, slope, and NDVI variables is important given the potential of
using these factors as responses to desertification-induced environmental degradation [46].
Land 2024, 13, 31 14 of 16

When assessing susceptibility to desertification in Mongolia in different years, a similar

trend has been observed between ESA and NDVI over time. This suggests that areas with
low NDVI values may signify regions in critical situations of desertification [22]. Under
conditions of low information availability, knowing the most important variables can
facilitate the construction of environmental sensitivity models, treating these factors as
independent variables [47–49]. However, it is important to mention that the monitoring of
the desertification process using the NDVI does not allow obtaining information about the
causes and factors influencing this phenomenon, thereby constraining the planning and
management of specific regions.
Compared to traditional GIS software, using cloud GEE offers distinct advantages,
especially in terms of processing speed and capacity [28,30,31]. Further, GEE also stands
out because of the high-quality digital databases already available in its public catalog,
significantly streamlining the workflow. The development of an environmental monitoring
model related to desertification in a cloud processing system allows the creation and
analysis of historical series from smaller time intervals, given its fewer constraints on
processing and storage capacity.
A considerable portion of the East Atlantic Basin showed susceptibility to deserti-
fication, requiring greater attention from public entities and civil society to mitigate the
possible impacts of this phenomenon on the local communities. The regions further north
and west of the basin, spanning areas in the states of Bahia and Sergipe, are in a more
critical state. Thus, it is vital to prioritize the establishment of public policies interven-
tions in municipalities within these areas. Given the significance of vegetation-related
parameters for the ESA, implementing measures to promote, recover, and conserve forested
areas, combined with sustainable soil management practices, can help reduce the area’s
vulnerability to degradation.

5. Conclusions
In the present study, we have used the Google Earth Engine platform to carry out
the first land desertification assessment of the East Atlantic Basin, Brazil. The study
underscores the possibility of using an open-access cloud computing platform to perform
relevant analysis of an extensive area, yielding key insights. For instance, (i) approximately
36% of the total area of the East Atlantic Basin exhibits a critical state of sensitivity to
desertification, with areas located in the north and midwest portions exhibiting the highest
levels of sensitivity to desertification. Conversely, coastal regions had the lowest levels of
environmental vulnerability across the basin; (ii) the NDVI, slope, and VQI emerged as
the edaphoclimatic variables most closely correlated with the East Atlantic Basin’s ESA
Index; (iii) the analysis of the Environmentally Critical Factors Index, in parallel with the
Environmentally Sensitive Index, offers valuable insights for understanding the importance
of each variable in the MEDALUS model’s composition.
The study sets the stage for future research using cloud computing platforms to
best describe, monitor, and assess land degradation not only in Brazil but also in other
global regions where such studies have historically been hindered due to computational
limitations. Furthermore, we believe that incorporating ground-based measurements could
enhance the analysis, addressing what was a primary limitation of this study.

Author Contributions: T.H.d.J.M., software, writing—original draft; B.L.d.S., writing—review and

editing; C.T., supervision, writing—review; A.d.P., methodology; A.d.P., formal analysis; O.L.L., data
curation; F.A.R., formal analysis, validation; A.C.S.L., writing—review; R.G.d.S.P., visualization. All
authors have read and agreed to the published version of the manuscript.
Funding: This research received no external funding.
Data Availability Statement: Data will be available upon request.
Conflicts of Interest: The authors declare no conflict of interest.
Land 2024, 13, 31 15 of 16

References
1. Sterk, G.; Stoorvogel, J.J. Desertification-Scientific versus political realities. Land 2020, 9, 156. [CrossRef]
2. Burrell, A.L.; Evans, J.P.; De Kauwe, M.G. Anthropogenic climate change has driven over 5 million km2 of drylands towards
desertification. Nat. Commun. 2020, 11, 3853. [CrossRef] [PubMed]
3. Beceriil-Piña, R.; Mastachi-Loza, C.A.; González-Sosa, E.; Díaz-Delgado, C.; Bâ, K.M. Assessing desertification risk in the semi-arid
highlands of central Mexico. J. Semi-Arid Environ. 2015, 120, 4–13. [CrossRef]
4. Souza, D.C.; Oyama, M.D. Climatic consequences of gradual desertification in the semi-arid area of Northeast Brazil. Theor. Appl.
Climatol. 2011, 103, 345e357. [CrossRef]
5. Dronin, N. Reasons to rename the UNCCD: Review of transformation of the political concept through the influence of science.
Environ. Dev. Sustain. 2023, 25, 2058–2078. [CrossRef]
6. Feng, S.; Zhao, W.; Zhan, T.; Yan, Y.; Pereira, P. Land degradation neutrality: A review of progress and perspectives. Ecol. Indic.
2022, 144, 109530. [CrossRef]
7. D’Odorico, P.; Bhattachan, A.; Davis, K.F.; Ravi, S.; Runyan, C.W. Global desertification: Drivers and feedback. Adv. Water Resour.
2013, 51, 326–344. [CrossRef]
8. Wang, Y.; Yan, X. Climate change induced by Southern Hemisphere desertification. Phys. Chem. Earth 2017, 102, 40–47. [CrossRef]
9. Liu, Z.; Si, J.; Deng, Y.; Jia, B.; Li, X.; He, X.; Zhou, D.; Wang, C.; Zhu, X.; Qin, J.; et al. Assessment of land desertification and its
drivers in semi-arid alpine mountains: A case study of Qilian Mountains Region, northwest China. Remote Sens. 2023, 15, 3836.
[CrossRef]
10. Wijitkosum, S. Reducing Vulnerability to Desertification by Using the Spatial Measures in a Degraded Area in Thailand. Land
2020, 9, 49. [CrossRef]
11. Ren, Y.; Liu, X.; Zhang, B.; Chen, X. Sensitivity assessment of land desertification in China based on multi-source remote sensing.
Remote Sens. 2023, 15, 2674. [CrossRef]
12. Bakr, N.; Weindorf, D.C.; Bahnassy, M.H.; El-Badawi, M.M. Multi-temporal assessment of land sensitivity to desertification in a
fragile agro-ecosystem: Environmental indicators. Ecol. Indic. 2012, 15, 271–280. [CrossRef]
13. Djeddaoui, F.; Chadli, M.; Gloaguen, R. Desertification susceptibility mapping using logistic regression analysis in the Djelfa area,
Algeria. Remote Sens. 2017, 9, 1031. [CrossRef]
14. Wang, W.; Jiang, Y.; Wang, G.; Guo, F.; Li, Z.; Liu, B. Multi-Scale LBP texture feature learning network for remote sensing
interpretation of land desertification. Remote Sens. 2022, 14, 348. [CrossRef]
15. Xu, D.; You, X.; Xia, C. Assessing the spatial-temporal pattern and evolution of areas sensitive to land desertification in North
China. Ecol. Indic. 2019, 97, 150–158. [CrossRef]
16. Meza Mori, G.; Torres Guzmán, C.; Oliva-Cruz, M.; Salas López, R.; Marlo, G.; Barboza, E. Spatial Analysis of Environmentally
Sensitive Areas to Soil Degradation Using MEDALUS Model and GIS in Amazonas (Peru): An Alternative for Ecological
Restoration. Sustainability 2022, 14, 14866. [CrossRef]
17. Abuzaid, A.S.; Abdelatif, A.D. Assessment of desertification using modified MEDALUS model in north Nile Delta, Egypt.
Geoderma 2022, 405, 115400. [CrossRef]
18. Lahlaoi, H.; Rhinane, H.; Hilali, A.; Lahssini, S.; Moukrim, S. Desertification assessment using MEDALUS model in watershed
Oued El Maleh, Morocco. Geosciences 2017, 7, 50. [CrossRef]
19. Kosmas, C.; Ferrara, A.; Briassouli, H.; Imeson, A. Methodology for Mapping Environmentally Sensitive Areas (ESAs) to Desertification;
European Commission: Luxembourg, 1999.
20. Wijitkosum, S. The impact of land use and spatial changes on desertification risk in degraded areas in Thailand. Sustain. Environ.
Res. 2016, 26, 84–92. [CrossRef]
21. Song, C.; Kim, W.; Kim, J.; Gebru, B.M.; Adane, G.B.; Choi, Y.E.; Lee, W.-K. Spatial assessment of land degradation using
MEDALUS focusing on potential afforestation and reforestation areas in Ethiopia. Land Degrad. Dev 2022, 33, 79–93. [CrossRef]
22. Lee, E.; Piao, D.; Song, C.; Kim, J.; Lim, C.H.; Kim, E. Assessing environmentally sensitive land to desertification using MEDALUS
method in Mongolia. For. Sci. Technol. 2019, 15, 210–220. [CrossRef]
23. Fadl, M.E.; Abuzaid, A.S.; AbdelRahman, M.A.E.; Biswas, A. Evaluation of Desertification Severity in El-Farafra Oasis, Western
Desert of Egypt: Application of Modified MEDALUS Approach Using Wind Erosion Index and Factor Analysis. Land 2022, 11, 54.
[CrossRef]
24. Karamesouti, M.; Panagos, P.; Kosmas, C. Model-based spatio-temporal analysis of land desertification risk in Greece. Catena
2018, 167, 266–275. [CrossRef]
25. Ferrara, A.; Kosmas, C.; Salvati, L.; Padula, A.; Mancino, G.; Nolè, A. Updating the MEDALUS-ESA Framework for Worldwide
Land Degradation and Desertification Assessment. Land Degrad. Dev. 2020, 31, 1593–1607. [CrossRef]
26. Vieira, R.M.S.P.; Tomasella, J.; Alvalá, R.C.S.; Sestini, M.F.; Affonso, A.G.; Rodriguez, D.A.; Barbosa, A.A.; Cunha, A.P.M.A.;
Valles, G.F.; Crepani, E.; et al. Identifying areas susceptible to desertification in the Brazilian northeast. Solid Earth 2015, 6, 347–360.
[CrossRef]
27. Tomasella, J.; Vieira, R.M.S.P.; Barbosa, A.A.; Rodriguez, D.A.; Santana, M.O.; Sestini, M.S. Desertification trends in the Northeast
of Brazil over the period 2000–2016. Int. J. Appl. Earth Obs. Geoinf. 2018, 73, 197–206. [CrossRef]
Land 2024, 13, 31 16 of 16

28. Chen, R.; Yang, H.; Yang, G.; Liu, Y.; Zhang, C.; Long, H.; Xu, H.; Meng, Y.; Feng, H. Land-Use Mapping with Multi-Temporal
Sentinel Images Based on Google Earth Engine in Southern Xinjiang Uygur Autonomous Region, China. Remote Sens. 2023,
15, 3958.
29. Souza, C.M., Jr.; Shimbo, J.Z.; Rosa, M.R.; Parente, L.L.; Alencar, A.A.; Rudorff, B.F.T.; Hasenack, H.; Matsumoto, M.; Ferreira, L.G.;
Souza-Filho, P.W.M.; et al. Reconstructing Three Decades of Land Use and Land Cover Changes in Brazilian Biomes with Landsat
Archive and Earth Engine. Remote Sens. 2020, 12, 2735. [CrossRef]
30. Kadri, N.; Jebari, S.; Augusseau, X.; Mahdhi, N.; Lestrelin, G.; Berndtsson, R. Analysis of Four Decades of Land Use and Land
Cover Change in Semiarid Tunisia Using Google Earth Engine. Remote Sens. 2023, 15, 3257. [CrossRef]
31. Velastegui-Montoya, A.; Montalván-Burbano, N.; Carrión-Mero, P.; Rivera-Torres, H.; Sadeck, L.; Adami, M. Google Earth Engine:
A Global Analysis and Future Trends. Remote Sens. 2023, 15, 3675.
32. Gorelick, N.; Hancher, M.; Dixon, M.; Ilyushchenko, S.; Thau, D.; Moore, R. Google Earth Engine: Planetary-scale geospatial
analysis for everyone. Remote Sens. Environ. 2017, 202, 18–27. [CrossRef]
33. Santos, L.S.; Mendes, L.A. Influence of the Equatorial Pacific Ocean Temperature Oscillation on the Precipitation Regime of the
East Atlantic Hydrographic Region. Braz. J. Phys. Geogr. 2020, 13, 1502–1518. [CrossRef]
34. Brasil Ministry of the Environment. East Atlantic Hydrographic Region Notebook; Ministry of Health: Brasília, Brazil, 2006.
35. Greenhalgh, S.; Samarasinghe, O.; Curran-Cournane, F.; Wright, W.; Brown, P. Using ecosystem services to underpin cost–benefit
analysis: Is it a way to protect finite soil resources? Ecosyst. Serv. 2017, 27, 1–14. [CrossRef]
36. Abatzoglou, J.T.; Dobrowski, S.Z.; Parks, S.A.; Hegewisch, K.C. TerraClimate, a high-resolution global dataset of monthly climate
and climatic water balance from 1958–2015. Sci. Data 2018, 5, 170191. [CrossRef] [PubMed]
37. United Nations Organization for Education, Science and Culture. Map of the World Distribution of Arid Regions: Explanatory Note;
UNESCO: Paris, France, 1979.
38. FAO/IIASA/ISRIC/ISSCAS/JRC. Harmonized World Soil Database (Version 1.2); FAO: Rome, Italy; IIASA: Laxenburg, Austria, 2012.
39. Bonafoni, S.; Sekertekin, A. Albedo Retrieval from Sentinel-2 by New Narrow-to-Broadband Conversion Coefficients. IEEE Geosci.
Remote Sens. Lett. 2020, 17, 618–1622. [CrossRef]
40. Silva, R.M.; Santos, C.A.G.; Maranhão, K.U.A.; Lima, R.F.P. Geospatial assessment of eco-environmental changes in desertification
area of the Brazilian semi-arid region. Earth Sci. Res. J. 2018, 22, 175–186. [CrossRef]
41. Lamqadem, A.A.; Pradhan, B.; Saber, H.; Rahimi, A. Desertification sensitivity analysis using Medalus model and GIS: A case
study of the oases of middle draa valley, Marocco. Sensors 2018, 18, 2230. [CrossRef]
42. Cunha, J.; Nóbrega, R.L.B.; Rufino, I.; Erasmi, S.; Galvão, C.; Valente, F. Surface albedo as a proxy for land-cover clearing in
seasonally dry forests: Evidence from the Brazilian Caatinga. Remote Sens. Environ. 2020, 238, 111250. [CrossRef]
43. Nabiollahi, K.; Golmohamadi, F.; Taghizadeh-Mehrjardi, R.; Kerry, R.; Davari, M. Assessing the effects of slope gradient and land
use change on soil quality degradation through digital mapping of soil quality indices and soil loss rate. Geoderma 2018, 318,
16–28. [CrossRef]
44. Vieira, R.M.S.P.; Cunha, A.P.M.A.; Alvalá, R.C.S.; Carvalho, V.C.; Ferraz Neto, S. Sestini, M.S. Land use and land cover map of a
semiarid region of Brazil for meteorological and climatic models. Braz. J. Meteorol. 2013, 28, 129–138. [CrossRef]
45. Mallick, J.; AlMesfer, M.K.; Singh, V.P.; Falqi, I.I.; Singh, C.K.; Alsubih, M.; Kahla, N.B. Evaluating the NDVI–Rainfall Relationship
in Bisha Watershed, Saudi Arabia Using Non-Stationary Modeling Technique. Atmosphere 2021, 12, 593. [CrossRef]
46. Fokeng, R.M.; Fogwe, Z.N. Landsat NDVI-based vegetation degradation dynamics and its response to rainfall variability and
anthropogenic stressors in Southern Bui Plateau, Cameron. Geosyst. Geoenviron. 2020, 1, 100075. [CrossRef]
47. Budak, M.; Günal, H.; Çelik, I.; Yildiz, H.; Acir, N.; Acar, M. Environmental sensitivity to desertification in northern Mesopotamia;
application of modified MEDALUS by using analytical hierarchy process. Arab. J. Geosci. 2018, 11, 481. [CrossRef]
48. Dutra, J.D.; Elmiro, M.A.T.; Garcia, R.A. Comparative analysis of methods applied in vegetation cover delimitation using Landsat
8 images. Soc. Nat. 2020, 32, 699–710. [CrossRef]
49. Liu, S.; Wang, T.; Kang, W.; David, M. Several challenges in monitoring and assessing desertification. Environ. Earth Sci. 2015, 73,
7561–7570. [CrossRef]

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.