Gis based mcda forest fire susceptibility and risk zone mapping in cho moi, bac kan, vietnam

THAI NGUYEN UNIVERSITY UNIVERSITY OF AGRICULTURE AND FORESTRY RONNIECA MAE ESPINOSA GERMAN GIS-BASED MCDA AND FREQUENCY RATIO: FOREST FIRE SUSCEPTIBILITY AND RISK ZONE MAPPING IN CHỢ MỚI BẮC KẠN, VIETNAM _ BACHELOR THESIS Study Mode : Full-time Major : Environmental Science and Management Faculty : Advance Education Program Office Batch : K49-AEP (2018-2021) Thai Nguyen, 05/11/2021 DOCUMENTATION PAGE WITH ABSTRACT Thai Nguyen University of Agriculture and Forestry Degree Program Student name Student ID Bachelor of Environmental Science and Management Ronnieca Mae Espinosa German DTN1754290024 Thesis Title GIS-Based MCDA: Forest Fire Susceptibility and Risk Zone Mapping in Cho Moi, Bac Kan, Vietnam Dr Nguyen Van Hieu, Mr Mui MinhTung Supervisor (s) Supervisor’s Signature Abstract: The improvement of forest management in anthropogenic landscapes is significant and a must, specifically for forest fire mitigation measures Forest fire mapping identifies risk levels for each zone while presenting ideal alternatives for the projected occurrence This study collected meteorological, geographical, and socioeconomic data sets to assess the possibility of fire occurrences in an area The researcher used nine influencing factors: temperature, normalized difference vegetation index (NDVI), land use and land cover (LULC), slope, aspect, elevation, and distance to settlements, population density, and distance to roads The map was generated through GIS-based multi-criteria decision analysis (MCDA) to create multiple thematic models of the study area The weights of the influencing factors were based on an expert’s knowledge, and they were computed through the analytic hierarchy process (AHP) and applied to the Cho Moi district in Bac Kan, Vietnam Collected forest fire news articles from Cho Moi district are listed and overlaid with the forest fire risk map to interconnect the zones and assess the reliability of the produced map Frequency ratio modeling was also applied for another validation and to perform FR models and maps The AHP technique results show that an area of 79.79 square kilometers (13%) received a very low risk of fire occurrence and the area that was projected to receive the highest risk of forest fire covered almost 60.76 square kilometers (10.18%) The FR technique results show the best factor map to be ranked and suitable to use for future similar studies On the final forest fire risk map using FR method, forest fire incidents fall into high-risk zones and very high-risk zones The methodology applied in this study can help decisionmakers, concerned authorities of the study area, and forest managers optimize forest management and fire mitigation measures The whole process is applicable to other regions (with the same culture and values regarding anthropogenic needs) forest fire risk map, GIS-based, analytic hierarchy process (AHP), forest Keywords: management, multi-criteria decision analysis (MCDA), frequency ratio vale Number of pages: 100 pages Date of Submission: November 5, 2021 i ACKNOWLEDGEMENT This whole thesis was made with whole strength provided by our Lord for that I am grateful I am expressing the highest praise to Him for I am able to continue writing this paper despite my own life challenges Completing the whole study starting from the collection of data up until the writing part clearly needs a lot of motivation and hard works I am clearly obtaining those things from my parents and brothers in the Philippines Priceless gratitude for you all I am also giving my gratitude to our great supervisors Dr Nguyen Van Hieu and Mr Mui Minh Tung for the non-stop support of needed data and valuable pieces of training for our study My supervisor and Mr Tung that also served as my supervisor always lend their time and technological devices every time that I process my data Thank you so much! I also want to give thanks to all of my friends, bunsos, besties, ates and kuyas (Hanna Ordanza, Aira Grace Cantor, Mary Lucky Hapon, Khaye Membrebe, Rania Aguinaldo, Princess Tamo, Kyla Gumate, Pauline Damaso, Elisha Ramos, Huynh Duc, Jemimah Valladores, Stefanie Joy Necesito, Pastor Ralph de Lara, Ella Elejorde, Angelica Adriano, Apple Villamor, Divine Vibares, Jan Carlo Mercado, Isaiah de Tomas, John Henry Papa, Raphael Recto, Francis Sta Ana, Gio Mercaida, Emerson Alvarado, James Eduard Dizon, Bernard Montano, Rizlyn Custodio, Michelle dela Rosa, Sherry Rafael, Andrei Cebanico, Ariel Magdurulang, Jacqueline Aquino, Myr Prades, Kaye Desiree Hernandez, Ka Andy, Ka Loren, Ka Mariell Mash, Ka Mac, Ka Joanna, Ate Wendy, Tito Abemelech Edquila, Tita Alma Mamangun, Mama Emma Espinosa, Linh Nguyen, Giang Doan, Thao Phuong Trinh, Maica Villanueva, Jhareca Patalen, Aiko Estrella, and Hanna Galindo) who inspired me a lot in studying, inspired ii me to become academically active, and made me learn to work things one-step at a time All planted seeds bounded with love of my friends and families bore the best fruits in my life According to Psalm 133:1, ―How good and pleasant it is when God’s people live together in unity!‖ I would also like to state that all of my Vietnamese friends also had a part and contributed to the knowledge that I have this time Dr Quang Binh Bui also contributed a big part in my GIS software skills and knowledge Great great thanks to him and to our course subject My whole thesis making is like an adventurous journey, working all day and night with Hanna Thank you! Thank you, (Mame) Michelle Espinosa German, (Dade Manny Pacute) Ronnie Cadorna German, Carl Angel German, and Cale German! You made me completely inspired to accomplish this thesis To God be the Glory! iii TABLE OF CONTENTS DOCUMENTATION PAGE WITH ABSTRACT i ACKNOWLEDGEMENT ii TABLE OF CONTENTS iv LIST OF FIGURES vi LIST OF TABLES viii LIST OF EQUATIONS x LIST OF ABBREVIATIONS xi DEFINITION OF TERMS xiii CHAPTER I INTRODUCTION 1.1 Research rationale 1.2 Research objectives 1.3 Research questions and hypothesis 1.4 Scope and Limitation .4 CHAPTER II LITERATURE REVIEW .6 2.1 Introduction 2.2 Forest fire and its influencing factors 2.2.1 Introduction 2.2.2 Natural Factors 2.2.3 Social Factors 2.3 Forest Fire, a natural disaster for forest in Vietnam 10 2.4 GIS-based MCDA Fire Risk Management 11 2.4.1 Introduction 11 iv 2.4.2 Analytic Hierarchy Process 12 2.4.3 Multi-criteria Decision Analysis (MCDA) 13 2.5 Frequency Ratio Modelling 14 2.6 Conclusion 14 CHAPTER III 16 METHODS 16 3.1 Study Area 16 3.2 Data Collection 19 3.3 Software .21 3.4 Analytical hierarchy process 21 3.5 GIS-MCDA model .22 3.6 Influencing factors .32 3.7 Criteria, weights and data processing 33 3.8 Frequency ratio model 48 CHAPTER IV 52 RESULT .52 4.1 Final Map .52 4.2 Validation .55 CHAPTER V DISCUSSION AND CONCLUSION .67 REFERENCES 72 APPENDICES 85 v LIST OF FIGURES Figure 1.Map of Cho Moi District, Bac Kan, Vietnam 17 Figure Structure of Saaty’s Analytical Hierarchy process (AHP) .22 Figure Flowchart of the methodology using GIS MCDA and AHP 23 Figure Rural commune and township map of Cho Moi district 34 Figure a Temperature map of Cho Moi district 35 Figure b Normalized difference vegetation index of Cho Moi district .37 Figure c Land use and land cover of Cho Moi district 38 Figure d Slope of Cho Moi district .39 Figure e Aspect of Cho Moi district 40 Figure f Elevation of Cho Moi district 41 Figure g Distance to settlement of Cho Moi district 43 Figure h Population density of Cho Moi district .45 Figure i Distance to road of Cho Moi district 46 Figure Flowchart of Frequency ratio modelling 48 Figure Tabulation of area in Arcmap environment 49 Figure Summing up all factor maps with an assigned FR values 50 Figure Forest fire risk map of Cho Moi district 52 Figure Very high and high forest fire risk zone map of Cho Moi district 54 Figure 10 Factor maps integrated with fire points buffered with km (a) temperature map, (b) normalized difference vegetation index map, (c) aspect map, (d) land cover and land use map, (e) slope map, (f) elevation map, (g) distance to settlements map, (h) population density, and (i) distance to road 61 vi Figure 11 Area under the curve of FR model 64 Figure 12 Final forest fire risk map using frequency ratio method 65 Figure 13 Forest fire susceptibility level 66 Figure 14 Nong Ha commune surrounded with anthropogenic properties .93 Figure 15 Na May village, Nong Ha commune with its sloped vegetation area .93 Figure 16 Two sides of Hoa Muc commune and the river 94 Figure 17 Cho Moi district map capture through google pro 94 Figure 18 Sample forest fire news article in Hoa Muc commune 95 Figure 19 Forest fire risk zone map of Cho Moi district 100 Figure 20 Integrated forest fire incidents, afforestation, and forest fire risk map 101 vii LIST OF TABLES Table Historical monthly average weather conditions for Cho Moi, Bac Kan 18 Table Scale for a pair-wise comparison matrix 24 Table Pair-wise comparison matrix of this study 26 Table Normalized pair-wise comparison matrix of this study .29 Table 5.Computed criteria weights of each influencing factors 29 Table Consistency matrix of the study 30 Table 7.The weighted sum value, criteria weights and ratio 30 Table Random inconsistency values .31 Table Ratio maximum, consistency index and consistency ratio 32 Table 10 List of criteria, data type, format, and source 32 Table 11 List of thematic layers with its normalized weight and sub-criteria with its final weights 33 Table 12 List communes planted with new trees and the covered number of hectares 47 Table 12 a Temperature weight values and risk levels .35 Table 12 b Normalized difference vegetation index weight values and risk levels .36 Table 12 c Land use and land cover weight values and risk levels 38 Table 12 d Slope weight values and risk levels 39 Table 12 e Aspect weight values and risk levels 40 Table 12 f Elevation weight values and risk levels .41 Table 12 g Distance to settlements weight values and risk levels .42 Table 12 h Communes with its location and population density 43 viii Table 12 i Population density weight values and risk levels .44 Table 12 j Distance to road weight value and risk levels 45 Table 13 List of forest fire incidents of villages, cause of incident, and date and time 47 Table 14 Summary of Forest fire risk map using GIS based MCDA and AHP technique 52 Table 15 No-very high and high forest fire risk zone, high forest fire risk zone, and very high forest fire risk zone of Cho Moi district in square meters 53 Table 16 Highest to lowest very high and high forest fire risk zone area in square meters 53 Table 17 Frequency ratio value of forest fire incident area to class area 55 Table 18 Factor maps' predictive rate 62 Table 19 Relative frequency and prediction rate of factor maps 62 Table 20 Temperature pair-wise and normalized comparison matrix 85 Table 21 NDVI pair-wise and normalized comparison matrix .86 Table 22 LULC pair-wise and normalized comparison matrix .87 Table 23 Slope pair-wise and normalized comparison matrix 88 Table 24 Aspect pair-wise and normalized comparison matrix 90 Table 25 Elevation pair-wise and normalized comparison matrix 91 Table 26 Distance to settlements pair-wise and normalized comparison matrix 92 Table 27 Population density pair-wise and normalized comparison matrix 93 Table 28 Distance to roads pair-wise and normalized comparison matrix 94 Table 29.Thematic layers pair-wise and normalized comparison matrix 91 ix Table 26 Distance to settlements pair-wise and normalized comparison matrix PAIR-WISE COMPARISON 400 0.032877333 SUM 0.2 N0RMALIZED INDEX 400 0.087591 0.061224 0.04878 0.047619 0.066667 Normalized Weight Final Weight 0.416212445 0.020886306 0.261787988 0.013137003 0.161050407 0.008081806 0.098572773 0.004946563 0.062376387 0.003130162 Table 27 Population density pair-wise and normalized comparison matrix PAIR-WISE COMPARISON >100 >100 100-80 80-60 60-30