Accuracy assessment of MODIS fire products in African

Accuracy Assessment Of Modis Fire Products In African-Free PDF

  • Date:30 Jul 2020
  • Views:2
  • Downloads:0
  • Pages:47
  • Size:2.03 MB

Share Pdf : Accuracy Assessment Of Modis Fire Products In African

Download and Preview : Accuracy Assessment Of Modis Fire Products In African


Report CopyRight/DMCA Form For : Accuracy Assessment Of Modis Fire Products In African


Transcription:

Acknowledgements, My greatest challenge was moving from a social science back ground into a purely scientific. line of thinking yet I consider it a privilege to have ventured into this pathway Despite all the. hurdles I had to jump all the way I learnt eventually that where there is a will there surely is. a way I therefore start by expressing my sincere gratitude to my family for all the support. they gave during the course of my Master of Science study. Let me begin by acknowledging the contributions of my thesis supervisor Doctor. Mhosisi Masocha who helped me realise my potential and develop a scientific independent. smart and hardworking mind To Masocha I am not able to thank you enough for helping me. realise the scope of my study and perfect its rationale I thank you most for the times when you. pushed me to work smart and discover things on my own it would be very ungrateful of me. not to admit the positive difference it made in me I am grateful I had you for my mentor You. taught me a lot from the basic GI science to scientific writing appreciation and yes I will. always remember to limit the content in my slides when I teach somewhere else or make. presentations out there in the world, To Henry Ndaimani your encouragement motivation and assistance during the course. of my thesis I appreciate with overwhelming gratitude. To my classmates I thank you all for all the contributions you made when things where. a bit hazy on my end The spirit we had as a team may it persist even out there in the world. Finally chief of all I thank God for providing for me continually life strength courage and. dedication throughout the course of my study My faith in you will abide forever Amen. Fire poses a continuous threat to forest ecosystems and can dramatically reduce valuable timber. species in forest woodland areas The occurrence of fire in Baikiaea plurijuga woodlands. warrants the need of an active and timely fire detection system Rudimentary methods of. detecting fires are still in use in most of Zimbabwe s forest reserves yet remote sensing has. played a pivotal role around the globe in detecting and monitoring both active incidences and. post fire burnt areas Several satellite systems have been validated in different biomes of the. world for both MODIS MOD14A1 and MCD14ML However there still remains a gap of. knowledge in the accurate detection of fires in Baikiaea plurijuga woodlands In this study we. evaluate the accuracy of MODIS fire products using confusion matrices kappa statistic true. skill statistic TSS remote sensing and Geographical Information techniques where employed. to assess the accuracy of MODIS MOD14A1 burnt area product and MCD14ML active fire. product in fire detection This is the first time results of accuracy assessment of fire products. are reported in Baikiaea plurijuga woodlands In both study sites course resolution 1 km. MODIS MOD14A1 burnt area fire product has a continually poor index of agreement with. ground data kappa 0 and TSS value is 0 However for MODIS MCD14ML we found high. kappa and true skill statistic values showing a high However we recorded high kappa and. TSS values for the MODIS MCDML active fire product These results are consistent with the. premise that increase in spatial resolution reduces the sensors ability to detect fires in African. Savanna woodlands,ii P a g e,Table of contents,Acknowledgements i. Abstract ii,Table of contents iii,List of Figures iv. List of Appendices vi,List of Tables vi,List of Abbreviations and Acronyms vii.
Chapter 1 Introduction 1,1 1 Problem statement 3,1 2 Objectives 3. 1 3 Hypothesis 3,1 4 Justification of the study 4,Chapter 2 Materials and methods 5. 2 1 Gwayi State Forest 5,2 1 2 Matusadona National Park 7. 2 2 Data Collection 8,2 3 Data Analysis 11,2 3 1 Pre processing and Image analysis 11. 2 3 2 Post Classification 12,2 4 Statistics employed 12.
Chapter 3 Results 15, 3 1 Validation of MODIS MOD14A1 in Gwayi State Forest 15. 3 2 Validation of MODIS MCD14ML 15, 3 3 Validation of MODIS MOD14A1 in Matusadona National Park 15. 3 4 Validation of MODIS MCD14ML 16,Chapter 4 Discussion 29. Chapter 5 Conclusion 31,Bibliography 33,Appendices 38. iii P a g e,List of Figures, Figure 1 Map showing the spatial distribution of Baikiaea plurijuga woodlands in Zimbabwe.
and the location of Gwayi State forest 6, Figure 2 Map showing fire points collected from the Forest Commission database for the years. 2000 to 2009 10,Figure 3 Landsat 5 false colour composite 10. Figure 4 Landsat 5 false colour composite 11, Figure 5 Map showing agreement between ground fire points and the classified MOD14A1 fire. product for the year 2000 17, Figure 6 Map showing agreement between ground fire points and the classified MOD14A1 fire. product for the year 2005 18, Figure 7 Map showing agreement between ground fire points and the classified MOD14A1 fire.
product for the year 2006 19, Figure 8 Map showing agreement between ground fire points and the classified MOD14A1 fire. product for the year 2009 20, Figure 9 Map showing agreement between ground fire points and the classified MCD14ML. fire product for the year 2000 21, Figure 10 Map showing agreement between ground fire points and the classified MCD14ML. fire product for the year 2005 22, Figure 11 Map showing agreement between ground fire points and the classified MCD14ML. fire product for the year 2006 23, Figure 12 Map showing agreement between ground fire points and the classified MCD14ML.
fire product for the year 2009 24, Figure 13 Map showing agreement between ground fire data and the MOD14A1 fire product. for the year 2015 day 1 25,iv P a g e, Figure 14 Map showing agreement between ground fire data and the MOD14A1 fire product. for the year 2015 day 2 26, Figure 15 Map showing agreement between ground fire data and the MCD14ML fire product. for the year 2015 day 1 27, Figure 16 Map showing agreement between ground fire data and the MOD14A1 fire product. for the year 2015 28,List of Appendices, Appendices 1 Error matrix tables for MODIS MOD14A1 in Gwayi Forest 38.
Appendices 2 Error matrix tables for MODIS MCD14ML in Gwayi Forest 39. Appendices 3 Error matrix tables for MODIS MCD14ML in Matusadona National Park 39. Appendices 4 Error matrix tables for MODIS MOD14A1 in Matusadona National Park 39. List of Tables, Table 1 Shows the two satellites validated in this study and their characteristics 9. Table 2 Accuracy statistics for MODIS MOD14A1 and MODIS MCD14ML fire products in. Gwayi State forest in Zimbabwe 16, Table 3 Accuracy statistics for MODIS MOD14A1 and MODIS MCD14ML fire products in. Mutusadona National Park in Zimbabwe 17,vi P a g e. List of Abbreviations and Acronyms,MODIS Moderate Resolution Spectroradiometer. FIRMS Fire Information for Resource Management,TSS True Skill Statistic.
ASTER Advanced Spaceborne Thermal Emission and Reflection Radiometer. NOAA National Oceanic and Atmospheric Administration. vii P a g e,Chapter 1 Introduction, Fire has the capacity to break sustainable environments and can dramatically reduce the living. biomass in tropical forests even to the same degree as logging activities Weber and Flannigan. 1997 Fire can significantly affect biomass composition and content of tropical trees and the. composition of flora which are not adapted to this disturbance Martins et al 2012. Ecosystems subjected to frequent and severe fire disturbance can offset this balance resulting. in the ecosystem s loss of recovery ability and major ecological functions Held 2006. Empirical studies have demonstrated the higher mortality of trees with smaller diameter in. areas which experienced recent burning Stroppiana et al 2003 Whereas an increase in. mortality of larger trees 50 cm dbh has been reported 1 3 years after a fire Stroppiana et. al 2003 attributed to both natural and anthropogenic causes Africa is referred to as the Fire. Continent and biomass burning is recognised as an important and extensive function of. African grasslands and savannas However 168 million hectares burn annually and savanna. burning accounts for 50 of this total In African savannas fire is a key component playing a. pivotal ecological role in controlling vegetation patterns Houghton 2007. Satellite based remote sensing observations have been used to generate global scale burned. area BA products providing estimates of the extent and severity of damages Brink and Eva. 2009 Satellite remote sensing is an essential technology for gathering post fire related data in. an affordable and time conserving manner given the extremely broad spatial expanse and often. limited accessibility of the areas affected by forest fire Veraverbeke et al 2012 The. assessment of fire regimes across broad areas is done competently using data obtained from. remote sensors such as Moderate Resolution Spectroradiometer SPOT and National Oceanic. and Atmospheric Administration fire products which are specifically designed for this purpose. Korontzi et al 2006 Fire detection is performed using an algorithm that exploits the strong. emission of mid infrared radiation from fires Giglio et al 1999 Burnt area detection is. critical for forest reserves to estimate in automated manner the aggregate expanse of the burnt. surface in time and space and is highly demanded for statistical inventories Moreover it is a. step towards more detailed analysis such as burn frequency and severity Giglio et al 2009. Several studies Padilla et al 2014 have validated different fire products in various parts and. different biomes of the world Padilla et al 2015 for example compared the accuracies of. six remote sensing global burned area products using stratified random sampling and. estimation at a global scale in 2008 Padilla et al 2014 carried out a performance study of. the MODIS MCD45A1 in automated burned area detection in the Brazilian savanna. Schroeder et al 2008 validated GOES and MODIS active fire detection products using. ASTER and ETM data A spatio temporal analysis of fire relapse and magnitude for semi. arid savanna ecosystems in Southern Africa was also done by Pricope and Binford 2012. using moderate resolution satellite imagery In 2000 Silva et al 2003 used SPOT. VEGETATION satellite data to estimate during the dry seasons areas in Southern Africa Silva. et al 2003 carried out a study on burned area mapping in Greece using SPOT 4 HRVIR. images and Object Based image analysis and suggested that accurate information relating the. impact of fire environment is a key factor in quantifying the impact of fires on landscapes. selecting and prioritizing treatments applied on site planning and monitoring restoration and. recovery activities and providing baseline information for future monitoring. It is important to note that not all satellites are efficient in detecting fire scars due to different. spatial spectral and temporal resolutions Empirical studies have established that fire. comprises a major threat to protected areas and suggested that a special focus be placed in the. detection of fire scars and rapid alert of fires within forest reserves This study tested the. performance of two MODIS fire products in detecting burnt area scars in Baikiaea plurijuga. woodlands that is MODIS MOD14A1 and MODIS MCD14ML,1 1 Problem statement. By measuring thermal anomalies caused by active fires remote sensing has considerable. potential for mapping burnt areas and fire regimes However limited studies have focused in. evaluating the impact of sensor resolution in detecting burnt scars or active fires particularly in. African Savana woodlands Because of their hardwood nature African Savanna woodlands. have been designated as an important economic resource and the incidence and severity of. forest fires appears to have accelerated over the past decade reducing the commercial value of. this timber resource,1 2 Objectives, 1 To test the performance of MODIS MOD14A1 and MCD14ML fire products in the. detection of fire scars within African Savanna woodlands. 2 To assess the effects of spatial resolution and sensor characteristics on ability of remote. sensing products to detect fire in African Savanna woodlands. 1 3 Hypothesis, Increase in spatial resolution decreases the sensors ability to detect fire scars in African. Savanna woodlands,1 4 Justification of the study, Baikiaea plurijuga or teak woodlands confined to the western and north western parts of the.
country on Kalahari sands extending over one million hectares contain the most. commercially exploitable indigenous timber species Their average productivity ranges. between 150 200 m3 ha with a mean annual increment of 0 6 to 0 7 m3 ha Chihambakwe. 1987 Because of their economic value about 800 000 hectares of the natural baikiaea. woodlands were demarcated and gazetted as forest reserves However the incidence and. severity of forest fires appears to have accelerated over the past decade and reduces the. commercial value of timber CHIHAMBAKWE 1987 Hence accurate detection of fire in. these ecosystems using remote sensing and statistical indices of agreement may contribute. towards development of an operation fire detection system. Chapter 2 Materials and methods,2 1 Gwayi State Forest. In this study accuracy assessment of the two fire products was done in two different study sites. that is Gwayi State forest and Matusadona National park Gwayi forest is one of several. Accuracy assessment of MODIS fire products in African Savanna woodlands A Dissertation submitted to the Department of Geography and Environmental Science in Partial Fulfilment of the Requirements for the Master of Science Degree in Geographical Information Science and Remote Sensing June 2016 NDUMEZULU T MPOFU R156317D i P a g e Acknowledgements My greatest challenge was moving from a

Related Books