Arboriculture & Urban Forestry 42(6): November 2016 McPherson 2005; Huang et al. 2007; Shahidan et al. 2010; Ma and Ju 2011; Akamphon 2014). As urban forests are vital in ecology, their man- agement must include strategic and appropriate urban design and planning (Kong and Nakagoshi 2005; Huang et al. 2007; Iovan et al. 2008). Never- theless, rapid urbanization, which poses a threat to the safety of an ecosystem (Hepinstall-Cymerman et al. 2013), has compelled scholars to focus on urban greenspaces. Human society seems to have realized that living without nature is difficult and unsafe (Kong and Nakagoshi 2005; Li et al. 2010). With progressing urbanization, managing urban forests has become a significant con- cern. The growth of residential and commercial areas can negatively affect vegetation and ecol- ogy. Hence, one of the issues in urban forestry is determining the state and quantity of urban vegetation and buildings and controlling their growth and deterioration (Kong and Nakagoshi 2005; Iovan et al. 2008; Gillespie et al. 2012). Suf- ficient knowledge on urban forests, such as knowl- edge of tree location, size, and species, is essential for effective urban forestry (Ardila et al. 2012). Manual field measurement was the earliest method used to study urban forests (Francis 1987). In this method, the whole city or some parts of an area are randomly selected for sampling (Nowak et al. 2008). However, an urban space is a com- plex area, and obtaining information on all the trees and vegetation spaces via field surveying is difficult, time consuming, and may provide inac- curate results. These limitations can be overcome by using remote sensing to obtain accurate infor- mation by monitoring and controlling urban areas and vegetation (Ardila et al. 2012). The present article reviewed several studies that utilized remote- sensing techniques to investigate urban forests; evaluate the potential of remote sensors and differ- ent methods, such as pixel-based and object-based; and obtain accurate urban forest information. EVALUATION OF DIFFERENT SATELLITE IMAGING TECHNIQUES Field surveys and visual interpretations of aerial imagery are conventionally used to generate a vegetation cover map, but these methods are costly and time consuming. Therefore, scien- tists integrated traditional tools into new remote- 401 sensing systems (Huang et al. 2007; Iovan et al. 2008); such tools include different satellite im- ages, passive optical systems, and active sensors. Moderate resolution imaging spectroradiom- eter (MODIS) is medium-resolution imagery used to monitor urban forests with a spatial resolution of 250–500 m. MODIS and Landsat imageries are multi-temporal remote-sensing tools, and their most important characteristic is the ability to obtain seasonal and annual information on dif- ferent types of vegetation and land covers (Peijun et al. 2010; Zheng and Qui 2012; Qu et al. 2014). Vegetation covers have been shown to have dif- ferent patterns in time-series experiments under various conditions, such as humidity, because of their potential to combine various information or species compositions (Zheng and Qui 2012). Nev- ertheless, MODIS and Landsat exhibit temporal limitations (16 days of repeat cycles) (Shouse et al. 2013). Moderate-resolution imageries oſten have mixed pixels because of their low spatial resolu- tion (approximately 30 m), and thus they cannot be defined as a specific pure class (Peijun et al. 2010) and can only detect land-use types at the city level (Huang et al. 2007; Zheng and Qui 2012). The other medium-resolution satellite systems used to study urban forests are Landsat systems, which can provide a means to rapidly monitor urban forests (Huang et al. 2007; Zhang et al. 2007). The visual interpretation of Landsat TM shows that bands 2 and 4 provide sufficient information on land-cover types, and an image with a false color composite of band 4-5-3 (R-G-B) clearly differentiates vegetation types, particularly when the adaptive enhancement technique is used (Kamaruzaman and Haszuliana 1996; Ismail and Jusoff 2004; Cai et al. 2010). How- ever, when the study area is bound with a compact plantation because of spectral similarity, small urban and clearance areas are difficult to separate from tree species and mixed agriculture crops (Ismail and Jusoff 2004). Huang (2007) demonstrated that the Landsat ETM + imagery of spread urban trees is less coarse than other medium-resolution imageries. High-resolution satellites have been developed to overcome the limitation of moderate-resolution imageries, such as low spatial resolution. Hence, high-resolution imageries, such as those from QuickBird (Tooke et al. 2009; Hashiba et al. 2004; Ardila et al. 2012), IKONOS (Green- ©2016 International Society of Arboriculture
November 2016
| Title Name |
Pages |
Delete |
Url |
| Empty |
Search Text Block
Page #page_num
#doc_title
Hi $receivername|$receiveremail,
$sendername|$senderemail wrote these comments for you:
$message
$sendername|$senderemail would like for you to view the following digital edition.
Please click on the page below to be directed to the digital edition:
$thumbnail$pagenum
$link$pagenum
Your form submission was a success. You will be contacted by Washington Gas with follow-up information regarding your request.
This process might take longer please wait
