Boosted decision tree classifications of land cover over Turkey integrating MODIS, climate and topographic data

2011-01-01
EVRENDİLEK, FATİH
Gulbeyaz, O.
This study investigates the impact of using different combinations of Moderate Resolution Imaging Spectroradiometer (MODIS) and ancillary datasets on overall and per-class classification accuracies for nine land cover types modified from the classification system of the International Geosphere Biosphere Programme (IGBP). Twelve land cover maps were generated for Turkey using boosted decision trees (BDTs) based on the stepwise addition of 14 explanatory variables derived from a time series of 16-day MODIS composites between 2000 and 2006 (Normalized Difference Vegetation Index (NDVI), Enhanced Vegetation Index (EVI) and four spectral bands) and ancillary climate and topographic data (minimum and maximum air temperature, precipitation, potential evapotranspiration, aspect, elevation, distance to sea and slope) at 500-m resolution. Evaluation of the 12 BDTs indicated that the BDT built as a function of all the MODIS and climate variables, aspect and elevation produced the highest degree of overall classification accuracy (79.8%) and kappa statistic (0.76) followed by the BDTs that additionally included distance to sea (DtS), and both DtS and slope. Based on an independent validation dataset derived from a pre-existing national forest map and Landsat images of Turkey, the highest overall accuracy (64.7%) and kappa coefficient (0.58) among the 12 land cover maps was achieved by using MODIS-derived NDVI time series only, followed by NDVI and EVI time series combined; NDVI, EVI and four MODIS spectral bands; and the combination of all MODIS and climate data, aspect, elevation and distance to sea, respectively. The largest improvements in producer's accuracies were observed for grasslands (+50%), barrenlands (+46%) and mixed forests (+39%) and in user's accuracies for grasslands (+53%), shrublands (+30%) and mixed forests (+28%), in relation to the lowest producer's accuracy. The results of this study indicate that BDTs can increase the accuracy of land cover classifications at the national scale.
INTERNATIONAL JOURNAL OF REMOTE SENSING

Suggestions

Application of artificial neural network and logistic regression methods to landslide susceptibility mapping and comparison of the results for the ulus district, Bartin Bartin, ulus ilçesi için yapay sinir aǧi ve lojistik regresyon yöntemlerinin heyelan duyarlilik çalişmasina uygulanmasi ve karşilaştirilmasi
Eker, Arif Mert; Dikmen, Mehmet; Cambazoǧlu, Selim; Düzgün, Şebnem H.s.b.; Akgün, Haluk (2012-03-01)
Bu çalışma, Coğrafi Bilgi Sistemlerine (CBS) dayalı lojistik regresyon (LR) ve yapay sinir ağı (YSA) analizlerini kullanarak, Karadeniz bölgesindeki Bartın ilinin Ulus ilçesi için bir heyelan duyarlılık haritası hazırlamayı amaçlamaktadır. Bu araştırma kapsamında, Maden Tetkik ve Araştırma Genel Müdürlüğü tarafından hazırlanan heyelan envanter haritası, heyelan sınıflandırma haritasına temel olarak alınmıştır. Çalışma alanındaki analizlerin tamamı aktif heyelanlara istinaden gerçekleştirilmiştir. Bununla bi...
A new approach to evaluate the MODIS annual NPP product (MOD17A3) using forest field data from Turkey
Gulbeyaz, Onder; Bond-Lamberty, Ben; Akyürek, Sevda Zuhal; West, Tristram O. (Informa UK Limited, 2018-01-01)
In this study, we present the first evaluation of the MODIS (Moderate Resolution Imaging Spectroradiometer) annual net primary production (NPP) for Turkey's forest ecosystems using field measurements. Due to lack of country scale field measurements (i.e. flux tower for forest ecosystems), tree DBH (diameter at breast height) data set provided by Ministry of Forest and Water Affairs (MFWA) of Turkey is used to calculate NPP of Turkey's forest ecosystems. The lack of a reliable NPP data set leads the research...
New change detection method using double segmentation and its application on remotely sensed images
Gedik, Ekin; Yarman Vural, Fatoş Tunay; Department of Computer Engineering (2013)
Change detection research, a branch of statistical data analysis, focuses on detecting changed samples between di erent observations of the same dataset. The proposed study presents a novel change detection procedure and its application as a complete framework which is designed to work on remotely sensed images. The scope of the study is defined as detecting man-made change objects between satellite images of the same region, acquired at di erent times. Proposed framework has three main steps as preprocessi...
Standard survey methods for estimating colony losses and explanatory risk factors in Apis mellifera
van der Zee, Romee; Gray, Alison; Holzmann, Celine; Pisa, Lennard; Brodschneider, Robert; Chlebo, Robert; Coffey, Mary F.; Kence, Aykut; Kristiansen, Preben; Mutinelli, Franco; Bach Kim Nguyen, Bach Kim Nguyen; Noureddine, Adjlane; Peterson, Magnus; Soroker, Victoria; Topolska, Grazyna; Vejsnaes, Flemming; Wilkins, Selwyn (Informa UK Limited, 2013-01-01)
This chapter addresses survey methodology and questionnaire design for the collection of data pertaining to estimation of honey bee colony loss rates and identification of risk factors for colony loss. Sources of error in surveys are described. Advantages and disadvantages of different random and non-random sampling strategies and different modes of data collection are presented to enable the researcher to make an informed choice. We discuss survey and questionnaire methodology in some detail, for the purpo...
An approach for landslide risk assesment by using geographic information systems (gis) and remote sensing (rs)
Erener, Arzu; Düzgün, H. Şebnem; Department of Geodetic and Geographical Information Technologies (2009)
This study aims to develop a Geographic Information Systems (GIS) and Remote Sensing (RS) Based systematic quantitative landslide risk assessment methodology for regional and local scales. Each component of risk, i.e., hazard assessment, vulnerability, and consequence analysis, is quantitatively assessed for both scales. The developed landslide risk assessment methodology is tested at Kumluca watershed, which covers an area of 330 km2, in Bartın province of the Western Black Sea Region, Turkey. GIS and RS t...
Citation Formats
F. EVRENDİLEK and O. Gulbeyaz, “Boosted decision tree classifications of land cover over Turkey integrating MODIS, climate and topographic data,” INTERNATIONAL JOURNAL OF REMOTE SENSING, pp. 3461–3483, 2011, Accessed: 00, 2020. [Online]. Available: https://hdl.handle.net/11511/65225.