Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Dynamical spatial modeling to simulate the forest scenario in Brazilian dry forest landscapes

Dynamical spatial modeling to simulate the forest scenario in Brazilian dry forest landscapes GEOLOGY, ECOLOGY, AND LANDSCAPES 2019, VOL. 3, NO. 1, 46–52 INWASCON https://doi.org/10.1080/24749508.2018.1481658 Dynamical spatial modeling to simulate the forest scenario in Brazilian dry forest landscapes José Jorge Monteiro Junior, Emanuel Araújo Silva, Ana Luiza De Amorim Reis and João Pedro Mesquita Souza Santos Forestry Science Department, Federal Rural University of Pernambuco, Recife, Brazil ABSTRACT ARTICLE HISTORY Received 2 April 2018 Caatinga is a biome located in Brazilian northeast region well known by the dry forest Accepted 22 May 2018 vegetation. The analyzed area called charcoaling zone has a great suppression of native vegetation to supply the charcoal market. Through the images from the Operational Linescan KEYWORDS System sensor of LANDSAT 8, this paper demonstrates, in a spatial-temporal way, the Geoinformation/GI; land degradation in landscapes of the Caatinga biome and the evaluation of the forest scenario. distribution; legislation; Thus, it has been possible to evaluate the changes experienced by the vegetation (2013– remote sensing 2016) and to identify the land use patterns and coverage and to quantify them according to the obtained satellite images. Agriculture and the exposed soil represent about 70% of the area, considering the growing of the anthropic area and consequently the emergence of several exposed soil areas. The simulation of the future scenarios was used by modeling with the application DINAMICA EGO, generating projections of the region of the coal for the year of 2019. In addition, the assessment of the forest effacement demonstrates decreased devel- opment of the forest and increased levels of exposed soil. 1. Introduction In this context, this study establishes and identi- fies a possible relationship between the use of land The Caatinga biome (dry forest) is a predominant coverand theincreaseordecreaseinthe areas vegetation in the northeast region, covering 54.53% established as agriculture, bare soil, water, and forest of the 1,548,672 km of the total area, occupying 83% cover. The field surveys and remote-sensing techni- of the state of Pernambuco (IBGE, 2008). que data are used in the landscape modeling, and The semi-arid zone is an area where anthropogenic such information is relevant to the diffusion of for- action occurs with great intensity on forest resources, est management in Pernambuco. Contributing to indicating a need for strategic planning to contain the the sustainability of the micro-region called the devastation of its vegetation. The strategic planning for “Zona de Carvoejamento” (charcoaling zone) the application includes the application of sustainable bound by Sá (2003), the field surveys and remote- forest management planning techniques, remote sen- sensing technique data constitute important tools sing, and geoprocessing (CPRM, 2005). for the formulation of more effective public and These techniques increase the understanding of environmental policies in this region (Silva, 2015). the behavior of the Caatinga biome (dry forest) and This study aims to create a model to simulate its aspects related to geographic, agricultural, geopo- future forest scenarios of the Caatinga biome (dry litical, and environmental parameters (Silva, 2015). forests) in relation to changes in time and influence Thus, monitoring and planning for the sustainable of economic, social, and environmental variables. use of natural resources are necessary in all areas of Also, this study further projects and evaluates the societies associated with their management through changes suffered over the years, through the demon- agricultural, forestry, and urban development. In this stration of the dynamic mode representing changes in context, it is necessary to know the importance of land areas of forest cover and identifies which factors that cover and land use, seeking to identify, in landscapes, could influence such changes. subsidies to understand the physical, economic, and social aspects in the global-scale levels (Pereira, 2008). Wood and coal are the main products from the 2. Methodology Caatinga, but obtaining these energy sources is far from First, prior to forming the geographic database for sustainable. Deforestation originates around 80% of these the study, the search periods were defined according forest products in the northeast (Gariglio, 2010). to their availability and image quality. For the CONTACT José Jorge Monteiro Junior josejorgemonteirojunior@gmail.com Forestry Science Department, Federal Rural University of Pernambuco, Recife, Brazil © 2018 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. GEOLOGY, ECOLOGY, AND LANDSCAPES 47 execution of this work, four scenes from different Table 1. List of scenes covering the study region and their respective dates. dates of the LANDSAT 8 satellite OLS (Operational Scene Orbit Point Year 2013 Year 2016 Linescan System) sensor were used. The four scenes 1 216 65 30/10 23/11 were acquired in the format “.geotiff” free of charges 2 216 66 30/10 23/11 from the United States Geological Survey (www.earth 3 215 65 10/12 02/12 4 215 66 10/12 02/12 explorer.usgs.gov). The scenes covered the entire study area from 2013 to 2016 (Table 1). Previously georeferenced metadata of the images was imported into the image manipulation software. The steps to land cover and land use modeling start The mosaic of the images was completed and later after the analysis of the evolution caused by the land cut within the limits of the study area defined by Sá use and land cover and their transitions. The projec- (2003)(Figure 1). Linear contrast was used to tion into the future scenarios (2019) used maps of land improve the image quality. From then on, the acqui- use and the cover model (2013 and 2016), which are sition of land cover and land use samples was carried continuous variables, in the DINAMICA EGO appli- out which defined the classes of forest, exposed soil, cation. The final results also visually compared with agriculture, and water. Also, static maps were created, the land use and coverage map for 2016 (Silva, 2015). and a wide range of variables was utilized that The maps used to demonstrate static variables were included cattle, sloping roads, logging, hydrography, determined by Evidence Weights (W+), the transition hypsometry (Shuttle Radar Topography Mission probabilities of land cover and land use classes. As you (National Aeronautics and Space Administration process the data, other files will be generated, and (NASA), 2012)), Human Development Index, urban these will be utilized in the following steps cores, Gross Domestic Project, and population. (Tramontina, 2016). The methodology was the same as used by Silva The first step is the calculation of transition (2015) and Tramontina (2016), being considered and matrices. According to Silva (2015), the “categorical adopted as the basic methodologies in this research. maps” set functors to calculate the transition matrix Figure 1. Studied area (Carvoejamento area/coaling zone) defined by Sá (2003). 48 J. J. MONTEIRO JUNIOR ET AL. for the initial and final maps of the analyzed periods. Thus, the “Uncertainty of Joint Information” of A Making a link with these data sets in the matrix and B, U (A, B), can be used as a measure of associa- “functor,” with the time differences between the initial tion, being defined by Equation (2): map and the final map. The maps were from 2013 to HAðÞþ HBðÞ HAðÞ ; B 2016, and the time window was 3 years. The results of UAðÞ ; B ¼ 2 (2) the matrix calculation were interconnected with two HAðÞþ HBðÞ single-step and multiple-step output “functors.” The first matrix involves the transitions that occur from According to Almeida (2003), the U index varies year to year, and the second matrix involves the tran- between 0 and 1, and when the two maps are com- sitions that occur throughout the analysis period. The pletely independent, H (A, B) = H (A) + H (B) and U results are saved in a Comma-Separated Values (A, B) is 0 (zero), and when the two maps are com- (CSV) file. pletely dependent, H (A) = H (B) = H (A, B) = 1 and The second step is the calculation of intervals for U (A, B) is 1 (one). Values that are less than 0.5, for categorization of continuous variables. In this step, both U and V, have a lower association and values “the calculation of evidence weights,” it was necessary above 0.5 show a high correlation. to group the static variables into a single file called Ferrari (2008) emphasizes that the above formula “Cube.” This is used to facilitate the insertion into the is indispensable, given that the Evidence Weights program of the maps that used to calculate the con- method is based on Bayes’ conditional probability tinuous variables. theorem. According to this theorem, the selection of Thethird step is thecalculation of Weightsof variables for modeling analysis should consider the Evidence coefficients. After calculations that define the evaluation of the independence between pairs of ranges of distances, the coefficients of the evidence explanatory variables selected to explain the same weights were used to select the variables that influence type of transition of land cover and land use (Silva, the dynamics of transitions from land cover and land use, 2015). creating local probabilities of differences. It requires: We used the initial land use and land cover (use) Input parameters (the initial and final maps and the map, the cube of static variables, the Weights of cube with static maps; the sliced file “Skeleton,” created Evidence file, and the “Determine Weights of in the previous step); the functor “Determine Weights of Evidence Correlation” functor. These, along with the Evidence Coefficients”; and as an output parameter sav- calculation of the distance map, the class numbers, ing the file of Weights of Evidence coefficients. and a functor, were generated into a table in the CSV The fourth step is the calculation of correlations format for further analysis. between variables. At this stage of modeling, the The fifth step is running the simulation model. spatially independent variables are observed that After the correlation step of the variables, we used include the spatial association between two variables, the functors with the use of the initial map, the cube eliminating from the model to those who were of variables, the transition matrix (from one year to strongly correlated with each other. The variables another), and the file of Weights of Evidence. A should consider the evaluation of the independence container called “Repeat” was added, which has the between these to and explain the same transition of function of executing the operations during the time land use and land cover. intervals, for example, between 2016 and 2019, the The parameters used to obtain the “correlation parameter used for 3 years. maps” are the Cramer (V) and Joint Information In this container, were inserted functors to carry Uncertainty (U) indexes that helped to decide which out the model process as follows: variables should be kept in the model. The Cramer “Mux Categorical Map” which is used to re-inject Index (V) is defined by Zuquette (2017) and Bonham the maps produced from one interaction to another Carter (1994) by Equation (1): and allows the feedback of the maps; “Calc Distance” which is used to calculate the distance rffiffiffiffiffiffiffiffiffiffiffi maps; V ¼ (1) “Modulate Change Matrix” it when the transitions T::M have been defined in percent change rates; “Expander” is used in the process for expansion or where contraction of spots that were already of a certain T = marginal totals of the cross-tabulation matrix class; between two maps A and B; “Patcher” has the function of generating or form- χ = chi-square statistics; and ing new spots through the sowing mechanism, that is, M = minimum of (n-1, m-1), where n is equal to looking for cells around the chosen location for a the number of rows and m is the number of columns joint transition; of the crosstab matrix between maps A and B. GEOLOGY, ECOLOGY, AND LANDSCAPES 49 “Calc W. OF. E. Probability Map” was entered to production, and a balance is needed between the calculate a transition probability map for each transi- demand for agricultural production and environmen- tion specified by adding the Evidence Weights; tal preservation. It can be observed in Figure 3. Finally, the exit parameters were inserted, one with In Figure 3, it is also noted that there is a greater the generation of the landscape of the annual maps predominance of the agricultural and livestock class and the other with the generation of probability maps and exposed soil where the largest cattle-producing of changes in the annual landscape. In this way, the municipalities in the region are found. This region is simulated maps were obtained for the final years and characterized by areas of higher altitudes, which may the same as the real maps (Figure 2) hinder the introduction of new agricultural areas. After the validation of the model, the scenario simu- When considering future simulation analysis and lation procedure was carried out for 2019, and the maps scenarios according to Benedetti (2010), the simula- created were quantified in terms of land use and land tion of maps through DINAMICA EGO is valid when cover. The calculation of the transitions matrices for the done under a calibrated model. In fact, this model years 2016–2019 was also carried out in order to adequately represents the transition processes that observe the trends of the changes from year to year have elapsed in the time interval considered, accord- and of the total changes in the period of 3 years. ing to the result obtained in the validation. Upon obtaining the simulation performed in the application (Figure 4), the visual comparison is used 3. Results and discussion between the classified map of 2016 and the simulated map of that year to observe the model’s condition. The land cover and land use mapping is based on the The simulated map bears similarity to the real map classification of images of LANDSAT 8. It was possi- optically, but the future studies may corroborate the ble to quantify the coverage and use of the quality of the simulated maps precision, in relation to Carvojamento area (coal area) defined by Sá (2003), the real maps obtained through a set of procedures in a satisfactory way. These indicate the decrease of and methods of digital image processing. the vegetation to the detriment of the increase in the Observing the simulation data for the year 2019 agricultural and cattle raising, which is predominant (Table 2)(Figures 5 and 6), the forest areas will be on in the central region to the west. These practices may the decline, given the increase in the exposed soil vari- be more common due to incentives for families and able areas. This is due to the intense use and extraction subsidies for agricultural growth from the federal and of wood in areas that do not have an adequate super- state governments. vision or monitoring or a sustainable forest manage- The progress of the thematic classes analyzed dur- ment plan. A management plan in the local forests ing the two years, specifically in the agriculture and would allow for well thought out and conscious con- exposed soil, were the highest values in hectares and sumption of the forest areas. If the logging continues, consequently in percentage; that is, they were the statistically, in 2025, the forest class will correspond to predominant classes in the region of study (Silva, 10.36% of the area, which would be equivalent to the 2015). This relationship is biased due to increased Figure 2. Running the simulation mode in DINAMICA EGO. 50 J. J. MONTEIRO JUNIOR ET AL. Figure 3. Land use and land cover maps of the coaling zone defined by Sá (2003), 2013 and 2016. Figure 4. Comparison of scenarios (in tones “blue” admits class “water”; “red” admits class “exposed soil”; “yellow” admits the class “agriculture”; “green” admits the class “forest”). “I” is the classified map of land use and land cover and “II” is the model (predicted) from the real map in 2016. Table 2. Quantification (point pixel) of land use and land respectively, to create models that proved to be effi- cover in the coaling zone and their respective prediction cient tools in monitoring the transition processes of percentages for 2019. the forest cover while incorporating important vari- 2019 Percentage Classes of land use and land cover (point pixel) over total area (%) ables in the zone of coaling. Water 9.86541 0.57 The decrease in the forest class was determined to be Bare soil 689.83619 39.57 Farming 678.52168 38.92 detrimental, and the increase of the exposed soil class Forest 364.95620 20.94 showed an alarming result that prompted the following proposals of intervention: zoning; creation of studies for the implementation of a management plan (for the Catimbau National Park. This park is the only pre- coaling zone); implementation of ecological corridors; served park in the area which is considered a national and also a massive plan of environmental education, conservation unit and has management plan. enabling a gradual cultural change that would reduce the illegal logging in the study area. Simulation of predictions for future scenarios will 4. Conclusion aid public policies to focus on the preservation of the In this study, it is possible to find the geoprocessing forest cover in the region. The simulation provides capacity for forest-related studies on its development determining factors that allow the expansion or and use in the Charcoaling Zone, using Landsat 8 retraction of these areas in the studied area. It is Sensor OLS, with images of 2013 and 2017, evident, therefore, that it is necessary to monitor the GEOLOGY, ECOLOGY, AND LANDSCAPES 51 Comparison of land use and land cover in 2013, 2016 and 2013(hectare) 2016 (hectare) 2019 (hectare) Water 17216.46 12886.29 9865.41 Bare Soil 361850.22 546161.22 689836.19 Farming 729356.85 726882.93 678521.68 Forest 634755.96 457249.05 364956.20 Figure 5. Graph of comparison of land use and land cover in 2013, 2016, and 2019. The agricultural class remains at the same level as observed in Figure 5. Figure 6. Simulated model for the year 2019. point-pixel exorbitance per map 52 J. J. MONTEIRO JUNIOR ET AL. area for the maintenance of the natural ecosystem CPRM - Serviço Geológico do Brasil. (2005). Projeto cada- stro de fontes de abastecimento por água subterrânea. and, thus, guiding other work both in the area of Diagnóstico do município de Ibimirim, estado de forest protection and its dynamism as well as in the Pernambuco. Retrieved from http://www.cprm.gov.br area of geospatial monitoring. Ferrari, R. (2008). Modelagem de Uso e Cobertura da terra da Quarta Colônia, RS. 2008. 127 f. (Masters in Geomatics): Universidade Federal de Santa Maria, RS. Gariglio, M. A. (2010). Uso sustentável e conservação dos Disclosure statement recursos florestais da caatinga (pp. 368). Brasília: Serviço No potential conflict of interest was reported by the Florestal Brasileiro. authors. INSTITUTO BRASILEIRO DE GEOGRAFIA E ESTATÍSTICA (IBGE). (2008). Biomas [Online]. Retrived from http://www.ead.codai.ufrpe.br/index.php/ apca/article/viewFile/178/161 Funding National Aeronautics and Space Administration (NASA). (2012). Global Modeling and Assimilation Office Annual This work was supported by the Conselho Nacional de Report & Research Highlights 2011-2012 [Online]. Desenvolvimento Científico e Tecnológico - CNPq [R$ Retived from https://gmao.gsfc.nasa.gov/research/ 29.910,20]. reports/GMAO_2011-2012Highlights.pdf Pereira, L. V. (2008). Análise espacial da evolução da cobertura e uso da terra no distrito de Santa Flora. Santa Maria, RS. References 100 f. Dissertação (Mestrado em Geomática): Universidade Federal de Santa Maria. Almeida, C. M. (2003). Modelagem da dinâmica espacial Sá,I.B.(2003). Fatores abióticos: Áreas e ações prioritárias para como uma ferramenta auxiliar ao planejamento: conservação da caatinga [Online]. Retrived from: http:// Simulação de mudanças de uso da terra em áreas urba- www.mma.gov.br/estruturas/chm/_arquivos/parte1caa.pdf nas para as cidades de Bauru e Piracicaba (SP). 351 f. Silva, E. A. (2015). Modelagem dinâmica para simulação no Tesis (Doctoring in Remote sensing) - INPE: São José processo de arenização e cobertura florestal na Campanha dos Campos, SP. Ocidental-RS. 165 f. Tesis (Doctoring in Forestry): Benedetti, A. C. P. (2010). Modelagem dinâmica para Universidade Federal de Santa Maria, RS. 2010. simulação de mudanças na cobertura florestal das Serras Tramontina, J. (2016). Modelagem dinâmica para estudo de do Sudeste e Campanha Meridional do Rio Grande do uso e cobertura do terra em área de influência da usina Sul. 166 f. Tesis (Doctoring in Remote sensing): hidrelétrica dona francisca. 2015. 149 f. (Masters): Forest Universidade Federal de Santa Maria, RS. Engineering, Universidade Federal de Santa Maria, Santa Bonham Carter, G. F. (1994). Geographic information sys- Maria. tems for geoscientists: Modelling with GIS (pp. 305). Ontario: Pergamon. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Geology Ecology and Landscapes Taylor & Francis

Dynamical spatial modeling to simulate the forest scenario in Brazilian dry forest landscapes

Download PDF
7 pages

Loading next page...
 
/lp/taylor-francis/dynamical-spatial-modeling-to-simulate-the-forest-scenario-in-Q5jBgd9VtA

References (11)

Publisher
Taylor & Francis
Copyright
© 2018 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group.
ISSN
2474-9508
DOI
10.1080/24749508.2018.1481658
Publisher site
See Article on Publisher Site

Abstract

GEOLOGY, ECOLOGY, AND LANDSCAPES 2019, VOL. 3, NO. 1, 46–52 INWASCON https://doi.org/10.1080/24749508.2018.1481658 Dynamical spatial modeling to simulate the forest scenario in Brazilian dry forest landscapes José Jorge Monteiro Junior, Emanuel Araújo Silva, Ana Luiza De Amorim Reis and João Pedro Mesquita Souza Santos Forestry Science Department, Federal Rural University of Pernambuco, Recife, Brazil ABSTRACT ARTICLE HISTORY Received 2 April 2018 Caatinga is a biome located in Brazilian northeast region well known by the dry forest Accepted 22 May 2018 vegetation. The analyzed area called charcoaling zone has a great suppression of native vegetation to supply the charcoal market. Through the images from the Operational Linescan KEYWORDS System sensor of LANDSAT 8, this paper demonstrates, in a spatial-temporal way, the Geoinformation/GI; land degradation in landscapes of the Caatinga biome and the evaluation of the forest scenario. distribution; legislation; Thus, it has been possible to evaluate the changes experienced by the vegetation (2013– remote sensing 2016) and to identify the land use patterns and coverage and to quantify them according to the obtained satellite images. Agriculture and the exposed soil represent about 70% of the area, considering the growing of the anthropic area and consequently the emergence of several exposed soil areas. The simulation of the future scenarios was used by modeling with the application DINAMICA EGO, generating projections of the region of the coal for the year of 2019. In addition, the assessment of the forest effacement demonstrates decreased devel- opment of the forest and increased levels of exposed soil. 1. Introduction In this context, this study establishes and identi- fies a possible relationship between the use of land The Caatinga biome (dry forest) is a predominant coverand theincreaseordecreaseinthe areas vegetation in the northeast region, covering 54.53% established as agriculture, bare soil, water, and forest of the 1,548,672 km of the total area, occupying 83% cover. The field surveys and remote-sensing techni- of the state of Pernambuco (IBGE, 2008). que data are used in the landscape modeling, and The semi-arid zone is an area where anthropogenic such information is relevant to the diffusion of for- action occurs with great intensity on forest resources, est management in Pernambuco. Contributing to indicating a need for strategic planning to contain the the sustainability of the micro-region called the devastation of its vegetation. The strategic planning for “Zona de Carvoejamento” (charcoaling zone) the application includes the application of sustainable bound by Sá (2003), the field surveys and remote- forest management planning techniques, remote sen- sensing technique data constitute important tools sing, and geoprocessing (CPRM, 2005). for the formulation of more effective public and These techniques increase the understanding of environmental policies in this region (Silva, 2015). the behavior of the Caatinga biome (dry forest) and This study aims to create a model to simulate its aspects related to geographic, agricultural, geopo- future forest scenarios of the Caatinga biome (dry litical, and environmental parameters (Silva, 2015). forests) in relation to changes in time and influence Thus, monitoring and planning for the sustainable of economic, social, and environmental variables. use of natural resources are necessary in all areas of Also, this study further projects and evaluates the societies associated with their management through changes suffered over the years, through the demon- agricultural, forestry, and urban development. In this stration of the dynamic mode representing changes in context, it is necessary to know the importance of land areas of forest cover and identifies which factors that cover and land use, seeking to identify, in landscapes, could influence such changes. subsidies to understand the physical, economic, and social aspects in the global-scale levels (Pereira, 2008). Wood and coal are the main products from the 2. Methodology Caatinga, but obtaining these energy sources is far from First, prior to forming the geographic database for sustainable. Deforestation originates around 80% of these the study, the search periods were defined according forest products in the northeast (Gariglio, 2010). to their availability and image quality. For the CONTACT José Jorge Monteiro Junior josejorgemonteirojunior@gmail.com Forestry Science Department, Federal Rural University of Pernambuco, Recife, Brazil © 2018 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. GEOLOGY, ECOLOGY, AND LANDSCAPES 47 execution of this work, four scenes from different Table 1. List of scenes covering the study region and their respective dates. dates of the LANDSAT 8 satellite OLS (Operational Scene Orbit Point Year 2013 Year 2016 Linescan System) sensor were used. The four scenes 1 216 65 30/10 23/11 were acquired in the format “.geotiff” free of charges 2 216 66 30/10 23/11 from the United States Geological Survey (www.earth 3 215 65 10/12 02/12 4 215 66 10/12 02/12 explorer.usgs.gov). The scenes covered the entire study area from 2013 to 2016 (Table 1). Previously georeferenced metadata of the images was imported into the image manipulation software. The steps to land cover and land use modeling start The mosaic of the images was completed and later after the analysis of the evolution caused by the land cut within the limits of the study area defined by Sá use and land cover and their transitions. The projec- (2003)(Figure 1). Linear contrast was used to tion into the future scenarios (2019) used maps of land improve the image quality. From then on, the acqui- use and the cover model (2013 and 2016), which are sition of land cover and land use samples was carried continuous variables, in the DINAMICA EGO appli- out which defined the classes of forest, exposed soil, cation. The final results also visually compared with agriculture, and water. Also, static maps were created, the land use and coverage map for 2016 (Silva, 2015). and a wide range of variables was utilized that The maps used to demonstrate static variables were included cattle, sloping roads, logging, hydrography, determined by Evidence Weights (W+), the transition hypsometry (Shuttle Radar Topography Mission probabilities of land cover and land use classes. As you (National Aeronautics and Space Administration process the data, other files will be generated, and (NASA), 2012)), Human Development Index, urban these will be utilized in the following steps cores, Gross Domestic Project, and population. (Tramontina, 2016). The methodology was the same as used by Silva The first step is the calculation of transition (2015) and Tramontina (2016), being considered and matrices. According to Silva (2015), the “categorical adopted as the basic methodologies in this research. maps” set functors to calculate the transition matrix Figure 1. Studied area (Carvoejamento area/coaling zone) defined by Sá (2003). 48 J. J. MONTEIRO JUNIOR ET AL. for the initial and final maps of the analyzed periods. Thus, the “Uncertainty of Joint Information” of A Making a link with these data sets in the matrix and B, U (A, B), can be used as a measure of associa- “functor,” with the time differences between the initial tion, being defined by Equation (2): map and the final map. The maps were from 2013 to HAðÞþ HBðÞ HAðÞ ; B 2016, and the time window was 3 years. The results of UAðÞ ; B ¼ 2 (2) the matrix calculation were interconnected with two HAðÞþ HBðÞ single-step and multiple-step output “functors.” The first matrix involves the transitions that occur from According to Almeida (2003), the U index varies year to year, and the second matrix involves the tran- between 0 and 1, and when the two maps are com- sitions that occur throughout the analysis period. The pletely independent, H (A, B) = H (A) + H (B) and U results are saved in a Comma-Separated Values (A, B) is 0 (zero), and when the two maps are com- (CSV) file. pletely dependent, H (A) = H (B) = H (A, B) = 1 and The second step is the calculation of intervals for U (A, B) is 1 (one). Values that are less than 0.5, for categorization of continuous variables. In this step, both U and V, have a lower association and values “the calculation of evidence weights,” it was necessary above 0.5 show a high correlation. to group the static variables into a single file called Ferrari (2008) emphasizes that the above formula “Cube.” This is used to facilitate the insertion into the is indispensable, given that the Evidence Weights program of the maps that used to calculate the con- method is based on Bayes’ conditional probability tinuous variables. theorem. According to this theorem, the selection of Thethird step is thecalculation of Weightsof variables for modeling analysis should consider the Evidence coefficients. After calculations that define the evaluation of the independence between pairs of ranges of distances, the coefficients of the evidence explanatory variables selected to explain the same weights were used to select the variables that influence type of transition of land cover and land use (Silva, the dynamics of transitions from land cover and land use, 2015). creating local probabilities of differences. It requires: We used the initial land use and land cover (use) Input parameters (the initial and final maps and the map, the cube of static variables, the Weights of cube with static maps; the sliced file “Skeleton,” created Evidence file, and the “Determine Weights of in the previous step); the functor “Determine Weights of Evidence Correlation” functor. These, along with the Evidence Coefficients”; and as an output parameter sav- calculation of the distance map, the class numbers, ing the file of Weights of Evidence coefficients. and a functor, were generated into a table in the CSV The fourth step is the calculation of correlations format for further analysis. between variables. At this stage of modeling, the The fifth step is running the simulation model. spatially independent variables are observed that After the correlation step of the variables, we used include the spatial association between two variables, the functors with the use of the initial map, the cube eliminating from the model to those who were of variables, the transition matrix (from one year to strongly correlated with each other. The variables another), and the file of Weights of Evidence. A should consider the evaluation of the independence container called “Repeat” was added, which has the between these to and explain the same transition of function of executing the operations during the time land use and land cover. intervals, for example, between 2016 and 2019, the The parameters used to obtain the “correlation parameter used for 3 years. maps” are the Cramer (V) and Joint Information In this container, were inserted functors to carry Uncertainty (U) indexes that helped to decide which out the model process as follows: variables should be kept in the model. The Cramer “Mux Categorical Map” which is used to re-inject Index (V) is defined by Zuquette (2017) and Bonham the maps produced from one interaction to another Carter (1994) by Equation (1): and allows the feedback of the maps; “Calc Distance” which is used to calculate the distance rffiffiffiffiffiffiffiffiffiffiffi maps; V ¼ (1) “Modulate Change Matrix” it when the transitions T::M have been defined in percent change rates; “Expander” is used in the process for expansion or where contraction of spots that were already of a certain T = marginal totals of the cross-tabulation matrix class; between two maps A and B; “Patcher” has the function of generating or form- χ = chi-square statistics; and ing new spots through the sowing mechanism, that is, M = minimum of (n-1, m-1), where n is equal to looking for cells around the chosen location for a the number of rows and m is the number of columns joint transition; of the crosstab matrix between maps A and B. GEOLOGY, ECOLOGY, AND LANDSCAPES 49 “Calc W. OF. E. Probability Map” was entered to production, and a balance is needed between the calculate a transition probability map for each transi- demand for agricultural production and environmen- tion specified by adding the Evidence Weights; tal preservation. It can be observed in Figure 3. Finally, the exit parameters were inserted, one with In Figure 3, it is also noted that there is a greater the generation of the landscape of the annual maps predominance of the agricultural and livestock class and the other with the generation of probability maps and exposed soil where the largest cattle-producing of changes in the annual landscape. In this way, the municipalities in the region are found. This region is simulated maps were obtained for the final years and characterized by areas of higher altitudes, which may the same as the real maps (Figure 2) hinder the introduction of new agricultural areas. After the validation of the model, the scenario simu- When considering future simulation analysis and lation procedure was carried out for 2019, and the maps scenarios according to Benedetti (2010), the simula- created were quantified in terms of land use and land tion of maps through DINAMICA EGO is valid when cover. The calculation of the transitions matrices for the done under a calibrated model. In fact, this model years 2016–2019 was also carried out in order to adequately represents the transition processes that observe the trends of the changes from year to year have elapsed in the time interval considered, accord- and of the total changes in the period of 3 years. ing to the result obtained in the validation. Upon obtaining the simulation performed in the application (Figure 4), the visual comparison is used 3. Results and discussion between the classified map of 2016 and the simulated map of that year to observe the model’s condition. The land cover and land use mapping is based on the The simulated map bears similarity to the real map classification of images of LANDSAT 8. It was possi- optically, but the future studies may corroborate the ble to quantify the coverage and use of the quality of the simulated maps precision, in relation to Carvojamento area (coal area) defined by Sá (2003), the real maps obtained through a set of procedures in a satisfactory way. These indicate the decrease of and methods of digital image processing. the vegetation to the detriment of the increase in the Observing the simulation data for the year 2019 agricultural and cattle raising, which is predominant (Table 2)(Figures 5 and 6), the forest areas will be on in the central region to the west. These practices may the decline, given the increase in the exposed soil vari- be more common due to incentives for families and able areas. This is due to the intense use and extraction subsidies for agricultural growth from the federal and of wood in areas that do not have an adequate super- state governments. vision or monitoring or a sustainable forest manage- The progress of the thematic classes analyzed dur- ment plan. A management plan in the local forests ing the two years, specifically in the agriculture and would allow for well thought out and conscious con- exposed soil, were the highest values in hectares and sumption of the forest areas. If the logging continues, consequently in percentage; that is, they were the statistically, in 2025, the forest class will correspond to predominant classes in the region of study (Silva, 10.36% of the area, which would be equivalent to the 2015). This relationship is biased due to increased Figure 2. Running the simulation mode in DINAMICA EGO. 50 J. J. MONTEIRO JUNIOR ET AL. Figure 3. Land use and land cover maps of the coaling zone defined by Sá (2003), 2013 and 2016. Figure 4. Comparison of scenarios (in tones “blue” admits class “water”; “red” admits class “exposed soil”; “yellow” admits the class “agriculture”; “green” admits the class “forest”). “I” is the classified map of land use and land cover and “II” is the model (predicted) from the real map in 2016. Table 2. Quantification (point pixel) of land use and land respectively, to create models that proved to be effi- cover in the coaling zone and their respective prediction cient tools in monitoring the transition processes of percentages for 2019. the forest cover while incorporating important vari- 2019 Percentage Classes of land use and land cover (point pixel) over total area (%) ables in the zone of coaling. Water 9.86541 0.57 The decrease in the forest class was determined to be Bare soil 689.83619 39.57 Farming 678.52168 38.92 detrimental, and the increase of the exposed soil class Forest 364.95620 20.94 showed an alarming result that prompted the following proposals of intervention: zoning; creation of studies for the implementation of a management plan (for the Catimbau National Park. This park is the only pre- coaling zone); implementation of ecological corridors; served park in the area which is considered a national and also a massive plan of environmental education, conservation unit and has management plan. enabling a gradual cultural change that would reduce the illegal logging in the study area. Simulation of predictions for future scenarios will 4. Conclusion aid public policies to focus on the preservation of the In this study, it is possible to find the geoprocessing forest cover in the region. The simulation provides capacity for forest-related studies on its development determining factors that allow the expansion or and use in the Charcoaling Zone, using Landsat 8 retraction of these areas in the studied area. It is Sensor OLS, with images of 2013 and 2017, evident, therefore, that it is necessary to monitor the GEOLOGY, ECOLOGY, AND LANDSCAPES 51 Comparison of land use and land cover in 2013, 2016 and 2013(hectare) 2016 (hectare) 2019 (hectare) Water 17216.46 12886.29 9865.41 Bare Soil 361850.22 546161.22 689836.19 Farming 729356.85 726882.93 678521.68 Forest 634755.96 457249.05 364956.20 Figure 5. Graph of comparison of land use and land cover in 2013, 2016, and 2019. The agricultural class remains at the same level as observed in Figure 5. Figure 6. Simulated model for the year 2019. point-pixel exorbitance per map 52 J. J. MONTEIRO JUNIOR ET AL. area for the maintenance of the natural ecosystem CPRM - Serviço Geológico do Brasil. (2005). Projeto cada- stro de fontes de abastecimento por água subterrânea. and, thus, guiding other work both in the area of Diagnóstico do município de Ibimirim, estado de forest protection and its dynamism as well as in the Pernambuco. Retrieved from http://www.cprm.gov.br area of geospatial monitoring. Ferrari, R. (2008). Modelagem de Uso e Cobertura da terra da Quarta Colônia, RS. 2008. 127 f. (Masters in Geomatics): Universidade Federal de Santa Maria, RS. Gariglio, M. A. (2010). Uso sustentável e conservação dos Disclosure statement recursos florestais da caatinga (pp. 368). Brasília: Serviço No potential conflict of interest was reported by the Florestal Brasileiro. authors. INSTITUTO BRASILEIRO DE GEOGRAFIA E ESTATÍSTICA (IBGE). (2008). Biomas [Online]. Retrived from http://www.ead.codai.ufrpe.br/index.php/ apca/article/viewFile/178/161 Funding National Aeronautics and Space Administration (NASA). (2012). Global Modeling and Assimilation Office Annual This work was supported by the Conselho Nacional de Report & Research Highlights 2011-2012 [Online]. Desenvolvimento Científico e Tecnológico - CNPq [R$ Retived from https://gmao.gsfc.nasa.gov/research/ 29.910,20]. reports/GMAO_2011-2012Highlights.pdf Pereira, L. V. (2008). Análise espacial da evolução da cobertura e uso da terra no distrito de Santa Flora. Santa Maria, RS. References 100 f. Dissertação (Mestrado em Geomática): Universidade Federal de Santa Maria. Almeida, C. M. (2003). Modelagem da dinâmica espacial Sá,I.B.(2003). Fatores abióticos: Áreas e ações prioritárias para como uma ferramenta auxiliar ao planejamento: conservação da caatinga [Online]. Retrived from: http:// Simulação de mudanças de uso da terra em áreas urba- www.mma.gov.br/estruturas/chm/_arquivos/parte1caa.pdf nas para as cidades de Bauru e Piracicaba (SP). 351 f. Silva, E. A. (2015). Modelagem dinâmica para simulação no Tesis (Doctoring in Remote sensing) - INPE: São José processo de arenização e cobertura florestal na Campanha dos Campos, SP. Ocidental-RS. 165 f. Tesis (Doctoring in Forestry): Benedetti, A. C. P. (2010). Modelagem dinâmica para Universidade Federal de Santa Maria, RS. 2010. simulação de mudanças na cobertura florestal das Serras Tramontina, J. (2016). Modelagem dinâmica para estudo de do Sudeste e Campanha Meridional do Rio Grande do uso e cobertura do terra em área de influência da usina Sul. 166 f. Tesis (Doctoring in Remote sensing): hidrelétrica dona francisca. 2015. 149 f. (Masters): Forest Universidade Federal de Santa Maria, RS. Engineering, Universidade Federal de Santa Maria, Santa Bonham Carter, G. F. (1994). Geographic information sys- Maria. tems for geoscientists: Modelling with GIS (pp. 305). Ontario: Pergamon.

Journal

Geology Ecology and Landscapes – Taylor & Francis

Published: Jan 2, 2019

Keywords: Geoinformation/GI; land distribution; legislation; remote sensing

There are no references for this article.