texto en 
Inglés (pdf)
Artículo en XML
Referencias del artículo
Traducción automática
Enviar artículo por email
Citado por SciELO 
Similares en
SciELO 
Permalink
Introduction
Globally, the ongoing expansion of agriculture and livestock production, along with the rise in African oil palm (Elaeis guineensis Jacq.) plantations, as well as the occurrence of pests, wildfires, and illegal logging (De León et al., 2018; Gómez et al., 2023), are among the leading causes of land-use change. These activities have resulted in the loss of over 420 million hectares of forest cover, with an annual deforestation rate of 10 million hectares (Food and Agriculture Organization of the United Nations [FAO], 2024). Land-use change is driven by a combination of demographic, economic, technological, political, institutional, and cultural factors (Brovelli et al., 2020; Galicia-Sarmiento et al., 2007). In Mexico, land-use changes have been promoted by the implementation of social programs such as the Mexican Food System (SAM) in 1980, the National Food Program (PNA) in 1983, the Rain-Tequio-Food program in 1987, the Direct Support to the Countryside Program (PROCAMPO) in 1993, and the Special Program for Food Security (PESA) in 1994 (Guerrero-Arenas et al., 2010; Zarazúa-Escobar et al., 2011).
The loss of forest cover has serious consequences, including biodiversity loss, habitat degradation, increased local temperatures, and soil erosion. It also disrupts precipitation patterns and reduces the capacity for water infiltration, storage, and carbon sequestration- factors that intensify the impacts of climate change (Casiano-Domínguez et al., 2018; Rodríguez-Larramendi et al., 2016; Sahagún-Sánchez & Reyes-Hernández, 2018). The establishment of management programs that ensure biodiversity conservation in any geographic region is made possible through multitemporal analyses. These analyses facilitate monitoring the dynamics of vegetation and land-use change over time and also allow for the identification of areas under the greatest anthropogenic pressure, enabling the implementation of measures aimed at containing the degradation of natural resources (Sandoval-García et al., 2021).
The objective of this study was to determine forest cover loss through a multitemporal analysis using high-resolution imagery available for the period 1995-2022 in the Istmo-Costa, Maya, and Lacandon Jungle regions of the state of Chiapas. These three regions are important due to their high biodiversity, particularly the Lacandon Jungle, which is characterized by its species diversity, water recharge capacity, and carbon sequestration potential (Montoya et al., 2006). The hypothesis proposed is that agricultural expansion contributes to the loss of forest ecosystems and ecosystem services.
Materials and Methods
Study Area
The research was conducted in the Istmo-Costa region (Arriaga, Mapastepec, Pijijiapan, and Tonalá), the Maya region (Benemérito de las Américas, Marqués de Comillas, Palenque, Catazajá, and La Libertad), and the Lacandon Jungle region (Altamirano and Ocosingo) in the state of Chiapas, located in the southwestern part of Mexico (Figure 1).
Figure 1 Location of the study regions in the state of Chiapas, Mexico. Source: Compiled by the authors using QGIS 3.36.0 (QGIS Development Team, 2024).
Image collection
Orthophotos were downloaded from the “Espacios y Datos de México” platform (Instituto Nacional de Estadística y Geografía [INEGI, 2022a] ), and high-resolution satellite images from Airbus Defence and Space, GeoEye-1, and Birdseye were obtained using SASPlanet software (SASPlanet, 2022); both tools are open source, and the image downloads are free of charge. Four orthomosaics were generated, composed of 220 orthophotos with a spatial resolution of 1.5 m per pixel (1995), 1 930 Airbus Defence and Space images at 0.57 m per pixel (2008), 4 400 GeoEye-1 images at 0.28 m per pixel (2014), and 4 400 Birdseye images at 0.28 m per pixel (2022).
Image preprocessing
Prior to processing and subsequent analysis, each satellite image was prepared through geometric correction by selecting identifiable control points. These points were systematically distributed in areas with high levels of confusion caused by reflectance, exposure, noise, and cloud cover. Land cover changes were detected by resampling the images using the r.resamp.filter tool in the open-source software Quantum GIS QGIS 3.36.0 “Maidenhead” (QGIS Development Team, 2024).
Digital classification
Images from each period were clipped and subjected to the unsupervised K-Means classification algorithm, which groups pixel values into four classes (primary vegetation, secondary vegetation, agriculture, and grassland) using a multivariate clustering approach. Subsequently, supervised classification was applied by converting raster files to vector format.
Information related to land use, vegetation cover, and vegetation types was generated and compared with the classification developed by INEGI (2022b). This comparison resulted in the identification of the following land uses and vegetation covers: cloud forest, oak-pine forest, fir forest, pine forest, pine-oak forest, mangrove, palm grove, grassland, savanna, tropical rainforest, low deciduous forest, low spiny deciduous forest, low spiny semi-evergreen forest, low deciduous forest, medium semi-deciduous forest, medium semi-evergreen forest, cattail marsh, secondary vegetation, agriculture, roads, and human settlements. These were distributed across the Istmo-Costa, Maya, and Lacandon Jungle regions in the state of Chiapas.
Data validation
To assess the agreement and accuracy of the high-resolution satellite image classification results, the r.kappa module in GRASS 7.6.0 (QGIS Development Team, 2024) was used. An error matrix was generated, and the Kappa index (K) was calculated using the following equation (Quezada et al., 2022):
where,
Po = Observed agreement
Pe = Expected agreement by chance
1 - Pe = Maximum possible agreement not attributable to chance.
Multitemporal analysis
To calculate changes in vegetation cover and land use, a cross-tabulation was performed across four time periods: 1995-2008, 2008-2014, 2014-2022, and 1995-2022. Percent change, net change, rate of change, and relative change were calculated for each type of vegetation cover over time.
Determination of vegetation cover losses and gains
To determine the rate of change (δ n , %) the equation developed by FAO (1996) and adapted by Palacio-Prieto et al. (2004) was used:
where,
s 1= Surface area (ha) at Time 1
s 2 = Surface area (ha) at Time 2
n = Difference in years between the two dates.
Annual deforestation rate
Changes in land cover were identified by comparing pairs of vegetation cover and land use maps, which allowed for the creation of new maps illustrating the transitions that occurred during the period 1995-2022. Using the data obtained from image processing, the annual deforestation rate (r) was calculated by comparing land cover at the same location across two periods, applying the equation proposed by Puyravaud (2003):
where,
A 1 = Vegetation cover (ha) or land use at the initial time
A 2 = Vegetation cover (ha) or land use at the final time
t 1 = Initial time period
t 2 = Final time period
A positive value of r indicates an increase in vegetation cover, while a negative value indicates a loss of cover.
Results
Four orthomosaics were generated (Figure 2), from which 21 land use and vegetation classes were obtained through supervised classification (Figure 3). However, the present multitemporal analysis considered 16 vegetation types for assessing land use change dynamics, focusing on agriculture and secondary vegetation, since elements such as water bodies, roads, and human settlements showed no statistically significant changes.
Figure 2 Orthomosaics of the study regions in the state of Chiapas corresponding to the years 1995 (A), 2008 (B), 2014 (C), and 2022 (D). Source: Compiled by the authors using QGIS 3.36.0 (QGIS Development Team, 2024).
Figure 3 Land use and vegetation in the years A) 1995, B) 2008, C) 2014, D) 2022 in three regions of the state of Chiapas. Source: Compiled by the authors using QGIS 3.36.0 (QGIS Development Team, 2024).
The Kappa index was 0.82, which, according to Landis and Koch (1977), represents almost perfect agreement, ensuring a reliable interpretation of the land use and vegetation type classification.
Among the main ecosystems distributed across the three regions, the tropical rainforest stands out with an area of 284 303.25 ha, representing 62.42 % of the Lacandon Jungle region, 87 096.38 ha (45.08 %) in the Maya region, and 54 089.08 ha (22.92 %) in the Istmo-Costa region. Grasslands cover 90,041.36 ha (46.61%) in the Maya region, 80,597.27 ha (34.15 %) in the Istmo-Costa region, and 56 376.04 ha (12.38 %) in the Lacandon Jungle. Cloud forests are present in the Lacandon Jungle and Istmo-Costa regions, with areas of 51 525.79 ha (11.31 %) and 22 446.72 ha (9.51 %), respectively (Table 1).
Table 1 Distribution of vegetation types in the three study regions of the state of Chiapas
| Vegetation | Istmo-Costa | Maya | Lacandon Jungle | |||
|---|---|---|---|---|---|---|
| ha | % | ha | % | ha | % | |
| Oak-pine forest | 1 033.92 | 0.44 | 0.00 | 0.00 | 712.89 | 0.16 |
| Fir forest | 622.63 | 0.26 | 0.00 | 0.00 | 0.00 | 0.00 |
| Pine forest | 1 133.68 | 0.48 | 2 554.42 | 1.32 | 16 244.65 | 3.57 |
| Pine-oak forest | 772.25 | 0.33 | 0.00 | 0.00 | 41 781.63 | 9.17 |
| Cloud forest | 22 446.72 | 9.51 | 0.00 | 0.00 | 51 525.79 | 11.31 |
| Mangrove | 24 826.24 | 10.52 | 0.00 | 0.00 | 0.00 | 0.00 |
| Palm grove | 870.64 | 0.37 | 0.00 | 0.00 | 0.00 | 0.00 |
| Grassland | 80 597.27 | 34.15 | 90 041.36 | 46.61 | 56 376.043 | 12.38 |
| Savanna | 2 682.89 | 1.14 | 1 129.62 | 0.58 | 22.44 | 0.00 |
| Tropical rainforest | 54 089.08 | 22.92 | 87 096.38 | 45.08 | 284 303.25 | 62.42 |
| Low evergreen rainforests | 0.00 | 0.00 | 0.00 | 0.00 | 216.11 | 0.05 |
| Low deciduous forest | 3 660.98 | 1.55 | 0.00 | 0.00 | 0.00 | 0.00 |
| Low spiny deciduous forest | 852.79 | 0.36 | 4 437.05 | 2.30 | 0.00 | 0.00 |
| Medium semi-deciduous forest | 8 533.83 | 3.62 | 0.00 | 0.00 | 0.00 | 0.00 |
| Medium semi-evergreen forest | 33 694.73 | 14.28 | 2 245.26 | 1.16 | 3 566.37 | 0.78 |
| Cattail marsh | 220.34 | 0.09 | 5 688.40 | 2.94 | 690.33 | 0.15 |
| Total | 236 037.98 | 100.00 | 193 192.48 | 100.00 | 455 439.50 | 100.00 |
During the period 1995-2022, the three regions collectively experienced land cover changes across 374 050.93 ha, with signs of recovery observed only in oak-pine forest (−265.74 ha) and pine forest (−718.96 ha). Forest cover changes were calculated in relation to the expansion of secondary vegetation and agriculture; therefore, a negative value indicates ecosystem recovery. In this context, the greatest losses were recorded in tropical rainforest, with 136 482.78 ha (69.45 %) in the Lacandon Jungle region, 37 680.81 ha (41.86 %) in the Maya region, and 6 100.70 ha (6.97 %) in the Istmo-Costa region. This was followed by grassland, with 49 695.09 ha (56.57 %) lost in Istmo-Costa, 46 665.07 ha (51.54 %) in the Maya region, and 25 625.66 ha (13.04 %) in the Lacandon Jungle. Losses were also significant in cloud forest, with 24 977.24 ha (12.71 %) in the Lacandon Jungle and 4 638.19 ha (5.30 %) in Istmo-Costa (Table 2).
Table 2 Changes in cover by vegetation type in the three study regions of the state of Chiapas during the period 1995-2022.
| Vegetation | Istmo-Costa | Maya | Lacandon Jungle | |||
|---|---|---|---|---|---|---|
| ha | % | ha | % | ha | % | |
| Oak-pine forest | -265.74 | -0.30 | 0.00 | 0.00 | 151.02 | 0.08 |
| Fir forest | 137.75 | 0.16 | 0.00 | 0.00 | 0.00 | 0.00 |
| Pine forest | 584.72 | 0.67 | 905.27 | 1.01 | -718.96 | -0.37 |
| Pine-oak forest | 376.33 | 0.43 | 0.00 | 0.00 | 3 945.44 | 2.01 |
| Cloud forest | 4 638.19 | 5.30 | 0.00 | 0.00 | 24 977.24 | 12.71 |
| Mangrove | 2 439.55 | 2.79 | 0.00 | 0.00 | 0.00 | 0.00 |
| Palm grove | 710.15 | 0.81 | 0.00 | 0.00 | 0.00 | 0.00 |
| Grassland | 49 695.09 | 56.79 | 46 665.07 | 51.84 | 25 625.66 | 13.04 |
| Savanna | 296.47 | 0.34 | 1 934.04 | 2.15 | 66.26 | 0.03 |
| Tropical rainforest | 6 100.70 | 6.97 | 37 680.81 | 41.86 | 136 482.78 | 69.45 |
| Low evergreen rainforests | 0.00 | 0.00 | 0.00 | 0.00 | 319.78 | 0.16 |
| Low deciduous forest | 1 785.62 | 2.04 | 0.00 | 0.00 | 0.00 | 0.00 |
| Low spiny deciduous forest | 370.95 | 0.42 | 161.22 | 0.18 | 0.00 | 0.00 |
| Medium semi-deciduous forest | 2 627.11 | 3.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| Medium semi-evergreen forest | 16 394.80 | 18.73 | 34.21 | 0.04 | 2 984.66 | 1.52 |
| Cattail marsh | 1 618.48 | 1.85 | 2 637.56 | 2.93 | 2 688.70 | 1.37 |
| Total | 87 510.16 | 100.00 | 90 018.18 | 100.00 | 196 522.59 | 100.00 |
Note: This table represents changes in forest cover in relation to the expansion of secondary vegetation and agriculture; therefore, a negative value indicates ecosystem recovery.
During the period 1995-2008, in the Istmo-Costa region, land use and vegetation dynamics showed that the low deciduous thorn forest experienced the highest loss rate of 14 % due to agriculture, while this rate decreased by 3.58 % in the oak-pine forest. Meanwhile, secondary vegetation generated a maximum change rate of 10.01 % in the medium semi-evergreen forest and a minimum of 0.30 % in the fir forest (Figure 4A)
Figure 4 Change rates of forest cover in the Istmo-Costa region during the periods A) 1995-2008, B) 2008-2014, C) 2014-2022, D) 1995-2022.
In the period 2008-2014, change rates were higher, with a maximum of 38.97% loss of medium semi-evergreen forest due to agriculture and a recovery of 19.05 % in the oak-pine forest. The low deciduous thorn forest showed a maximum change rate of 37.63 % caused by secondary vegetation, and the oak-pine forest recovered 3.94 % of primary vegetation (Figure 4B).
During the period 2014-2022, a recovery process occurred, primarily in the cloud forest, with a maximum rate of 16.72 % in areas previously used for agriculture-an activity that shifted toward grassland, which showed a maximum loss of 9.82 %. Meanwhile, secondary vegetation decreased by 20.62 % in the tropical rainforest, likely due to the establishment of African oil palm plantations; however, secondary vegetation increased by 12.04 % in the medium semi-evergreen forest (Figure 4C).
Between 1995 and 2022, the medium semi-evergreen forest had a maximum loss rate of 14.03 % due to agriculture and 14.51 % due to secondary vegetation. In contrast, the oak-pine forest recovered 9.58 % in areas previously occupied by agriculture, and the tropical rainforest gained 3.33 % of secondary vegetation areas (Figure 4D).
In the Lacandon Jungle region, during the period 1995-2008, the tropical rainforest experienced a maximum loss rate of 6.72 % and 9.85 % due to agriculture and secondary vegetation, respectively. On the other hand, there was a recovery of 2.10 % in pine forest in areas previously used for agriculture, and 0.32 % in areas of secondary vegetation (Figure 5A).
Figure 5 Forest cover change rates in the Lacandon Jungle region during the periods 1995-2008 (A), 2008-2014 (B), 2014-2022 (C), and 1995-2022 (D).
During the period 2008-2014, change rates were similar to the previous period, with 9.15 % and 9.45 % of medium semi-evergreen forest replaced by agriculture and secondary vegetation, respectively. The pine forest recovered 6.46 % and 0.86 % of areas formerly occupied by agriculture and secondary vegetation, respectively (Figure 5B).
In the period 2014-2022, there was a gradual decrease in change compared to the previous period, with a maximum rate of 4.0 % of agricultural expansion in the tropical rainforest and 4.28 % of secondary vegetation in the low evergreen rainforest. The cover types with the highest recovery were pine-oak forest, with 2.11 % in relation to agriculture, and pine forest with 0.54 % corresponding to secondary vegetation (Figure 5C).
During the period 1995-2022, the highest loss rate was recorded in the tropical rainforest, with 5.42 % due to agriculture and 7.67 % due to secondary vegetation. In contrast, the pine forest recovered 2.75 % of areas previously occupied by agriculture and 0.51 % of areas of secondary vegetation (Figure 5D).
In the Maya region, during the period 1995-2008, agriculture and secondary vegetation caused maximum loss rates of 12.84 % and 7.36 %, respectively, in the tropical rainforest, and there was no recorded recovery in any vegetation type. However, the lowest change values were recorded in the pine forest due to agriculture (0.60 %) and in the savanna due to secondary vegetation (0.62 %).
During the period 2008-2014, change rates were higher than in the previous period, with a maximum of 27.21 % loss due to agriculture and 13.22 % due to secondary vegetation in the tropical rainforest, along with a 0.37 % recovery of pine forest resulting from a reduction in secondary vegetation.
From 2014 to 2022, a notable decline was recorded in the tropical rainforest, with a maximum change rate of 14.03 % due to agriculture and 8.31 % due to secondary vegetation. Meanwhile, the highest recovery was 1.56 % in the savanna due to agricultural retreat and 10.43 % in the tular due to secondary vegetation.
During the period 1995-2022, the highest loss rate due to agriculture was 16.33 %, and 8.93 % due to secondary vegetation in the tropical rainforest, while the pine forest recovered 1.27 % of areas previously occupied by secondary vegetation.
Among the three regions, deforestation rates for the 1995-2022 period indicate that the ecosystem most affected by the loss of its original cover was the cattail marsh, which lost 92.11 % of its surface area. During this period, cattail marsh declined from 2 792.31 ha to 220.34 ha, with an annual deforestation rate of 9.41 % in the Istmo-Costa region; in the Lacandon Jungle region, the loss was 87.62 % with an annual rate of 7.74%; while in the Maya region, the affected area was 36.99 %, with an annual rate of 3.68 % (Figure 6).
Figure 6 Annual deforestation rate by vegetation type in the three study regions of the state of Chiapas during the period 1995-2022.
A similar situation occurred in the savanna, which lost 91.40 % of its area in the Lacandon Jungle region, decreasing from 261.19 ha to 22.44 ha during the 1995-2022 period, with an annual deforestation rate of 9.09 %. During the same period, the savanna lost 79.96 % in the Maya region, with an annual rate of 5.95 %, and 64.09 % in the Istmo-Costa region, with an annual rate of 3.79 %.
Discussion
The results show that over a 27-year period (1995-2022), the three study regions lost a total of 374 050.93 ha of forest cover, representing 42.28 % of the study area and 5.12 % of the total surface area of the state of Chiapas. The tropical rainforest was among the most affected ecosystems, with a loss of 180 264.29 ha. Only oak-pine and pine forests showed signs of recovery. This trend is consistent with findings by Sandoval and Cantú (2021), who reported recovery of oak-pine and pine forests in the Copalita River sub-basin in Oaxaca.
In the Istmo-Costa region, the tropical rainforest was the most affected vegetation type, likely due to the introduction of crops such as African oil palm. This region includes the municipalities with the largest planted areas in the state: Acapetahua, with 11 292.84 ha, and Mapastepec, with 8 898.24 ha (Federación Mexicana de Palma de Aceite [FEMEXPALMA], 2021; Sistema de Información Agrícola y Pecuaria [SIAP], 2021). As in other countries, the cultivation of African oil palm has increased due to its rapid growth and high profitability, which has led to high levels of deforestation and environmental degradation (Borras et al., 2013; OECD/FAO, 2017). These ecosystem impacts are expected to intensify, as Chiapas is the leading African oil palm-producing state in Mexico, and the planted area continues to expand annually. These land cover changes contribute to the increase in greenhouse gas emissions, resulting in a hotter environment and a warmer, drier atmosphere (Alvarado et al., 2021).
In the Lacandon Jungle region, tropical forest vegetation (low, medium, and high forest) and cloud forest experienced the greatest loss of cover between 1995 and 2014. These data are associated with the establishment of African oil palm plantations between 2001 and 2009, with an annual growth rate of 11.11 % in grown area (SIAP, 2010). Between 2019 and 2020, the crop expanded by approximately 5 000 ha, representing a 4.5 % increase.
The most affected vegetation type in the Maya region was the tropical rainforest. In this region, African oil palm is one of the most important crops for local producers, with a planted area of 17 263.68 ha (FEMEXPALMA, 2021; SIAP, 2021).
Mazariegos-Sánchez et al. (2014) note that oil palm production is a significant source of income for residents of Acapetahua municipality, located in the Istmo-Costa region of Chiapas. The crop has expanded by 70 % in the states of Chiapas, Campeche, Tabasco, and Veracruz, often replacing other production systems such as traditional agriculture and livestock. This shift has led to the loss of natural vegetation in some evaluated municipalities, with a measurable impact on biodiversity (Hernández-Rojas et al., 2018). In the Coastal region, natural vegetation has been increasingly lost due to the expansion of agro-industrial crops and human settlements. These land use changes have affected the hydrological regime and altered natural resources (Escobar & Castillo, 2021).
The results of the present study are consistent with the findings of Villatoro-Arreola et al. (2022) conducted in the coastal plain of the southern Pacific region of the state of Chiapas, where changes in agricultural and livestock activities were observed. These authors attribute the change primarily to the establishment of African oil palm plantations, which has led to alterations and fragmentation of the lagoon system landscape. Camacho et al. (2017) state that land cover changes are due to the expansion of areas designated for agricultural production; however, although this activity is being implemented in greenhouses, its impact on vegetation persists.
Ávila et al. (2014) mention that African oil palm plantations are established in areas of high biological diversity and in regions inhabited by indigenous communities, which affects their food systems. This crop largely depends on government subsidies to remain profitable; however, the economic benefits are concentrated among business groups rather than landowners. In addition, these plantations cause significant environmental impacts.
Morales et al. (2019) state that deforestation in the state of Chiapas, along with the incompatibility between the natural composition of these landscapes-such as steep slopes and shallow soils-and agricultural and livestock activities, has created the most critical environmental scenario in the region. According to these authors, it is estimated that, in the short term, 55 % of the state's territory could face high levels of degradation, leading to biodiversity loss, a decline in ecosystem services, and issues related to geological and geomorphological phenomena.
The rate of change in the three regions of Chiapas is a result of the prevailing agricultural and livestock activities, which have increased drastically in recent years and have caused biodiversity loss in forest ecosystems (Paterno et al., 2024). The expansion of crops has impacted vegetation as well as the in-situ fauna. Chiapas ranks as the second state with the highest terrestrial mammal diversity, with a total of 210 species, nine of which are endemic to the state (Lorenzo et al., 2017).
The conversion of tree vegetation to African oil palm plantations results in large proportions of species being unable to adapt, leading to significant losses in fauna (Aratrakorn et al., 2006). On the other hand, extensive livestock farming in Chiapas has affected tree cover and is one of the main causes of deforestation. This activity has expanded into buffer zones of protected natural areas, where environmentally unfriendly practices are carried out (Vargas, 2018). For this reason, the implementation of climate-smart practices is necessary to conserve forested areas and prevent large-scale ecosystem loss.
The effects of climate change-such as rising temperatures, drought, heavy rainfall, and low temperatures-can reduce land productivity by up to 50 % in some regions. Based on this, farmers engaged in monoculture practices are at greater risk compared to those with diversified crops on their plots, making producers in the state of Chiapas highly vulnerable to this situation. Monoculture African oil palm plantations have emerged as a development strategy in the state of Chiapas (Fletes & Bonanno, 2015); however, despite plans and intentions to improve the population’s economy, create alternative energy sources, and protect the environment, the intended objectives have not been achieved, as evidenced by various studies on the subject (Alvarado et al., 2021; Ávila et al., 2014; Escobar & Castillo, 2021).
Conclusions
The assessment of land use change in the Istmo-Costa, Maya, and Lacandon Jungle regions of the state of Chiapas shows that there has been a significant shift in forest cover, mainly due to the establishment of agricultural areas. The Lacandon Jungle and Istmo-Costa regions reported the greatest vegetation changes, particularly in savanna and cattail marsh ecosystems. These two vegetation types harbor high biodiversity and play a key role in water regulation. The loss of cover affects ecosystem services and puts species at risk. The present study serves as a useful tool for decision-making in the planning and management of these regions, and for mitigating the effects caused by land use change and the loss of natural cover.
