This paper presented a methodology to fuse Landsat 30 m spectralreflec translation - This paper presented a methodology to fuse Landsat 30 m spectralreflec Indonesian how to say

This paper presented a methodology

This paper presented a methodology to fuse Landsat 30 m spectral
reflectance time series with contemporaneous MODIS 1 km active fire
detections to generate large area 30 m burned area maps. The methodology
attempts to overcome the limitations of the 16 day Landsat temporal
resolution by incorporating daily MODIS active fire detections.
It was demonstrated by processing a large annual dataset centered on
the forested ecoregions of the Western United States that covered predominantly
shrubland (36%), evergreen forest (30%), and grassland
(15%). Comparison with independently derived MTBS burned area perimeters
showed the absence of systematic commission errors due to
spectral changes unrelated to fire, and a good agreement in the identification
of regional burning patterns (Fig. 12). Further research to validate
the product will be required, ideally using independent burned area
reference data which complies with the CEOS Cal Val burned area validation
protocol (Boschetti et al., 2009). The comparison with MTBS perimeters
quantified a significant discrepancy in areal estimates between
the two datasets; while this discrepancy can be partially attributed to
the inclusion of unburned islands within the MTBS burned area perimeters
(Sparks et al., 2015), omission errors may also be due to other factors,
which are discussed below.
The precision in the temporal detection of the day of burning is limited
by the 16 day temporal resolution of the Landsat data. Over the
Western United States the observed uncertainty (median 23 days,
75th percentile 48 days) was significantly higher than the uncertainty
of current coarse resolution products, for example, estimated for the
MCD45 MODIS global burned area product as 1 day (median) and
4 days (75th percentile) (Boschetti, Roy, Justice, & Giglio, 2010). This
is of particular concern, because areas where the vegetation recovery
ismore rapid than the interval between consecutive cloud-free observations
may be missed by the present method. Use of contemporaneous
Landsat 5 and Landsat 7 data, or of contemporaneous Landsat 7 and
Landsat 8 data, would provide an 8-day repeat coverage and so enhanced
probability of cloud-free surface observation (Kovalskyy &
Roy, 2013; Roy et al., 2014). This combination would also reduce the
reporting uncertainty and, by providing a denser time series, enable
Landsat remote sensing closer to the fire event and therefore provide
a stronger spectral category change, likely resulting in improved burned
area detection.
A further limitation of the presented methodology is the requirement
that at least one MODIS active fire detection partially overlaps
with one ormore temporally and spatial adjacent burned area candidate
objects; analysis of the results showed that the omission of entire
MTBS polygons can be largely attributed to the absence of
coincident active fire detections. This has the potential of affecting
not only the presented methodology, but all hybrid algorithms
which use active fire detections combined with reflectance changes.
While the test over the Western United States showed that the majority
of fire events could be detected, previous global coverage research
illustrates that in low tree cover ecosystems MODIS fire
products significantly undersample the burning activity (Roy et al.,
2008) and that small burned area patches are less likely than large
patches to have any coincident active fire detections (Hantson,
Padilla, Corti, & Chuvieco, 2013). New research will be undertaken
to minimize this issue. First, incorporation of MODIS Aqua active
fire detections will be investigated as the Aqua and Terra overpass
times are different and so the two MODIS sensors combined provide
more active fire detections per day. Incorporation of active fire detections
from geostationary satellites that only detect very large
and hot fires but every 15 to 30 min (Roberts & Wooster 1992;
Zhang, Kondragunta, & Roy, 2014) will also be investigated.
In addition, research to refine the semantic rule-based change detection
will be conducted, for instance by further characterizing the preliminary
candidate burned area detections according to the
magnitude of spectral changes of suitable spectral band indices
(Bastarrika et al., 2011; Stroppiana, Bordogna, Boschetti et al.,
2012; Stroppiana, Bordogna, Carrara, et al., 2012).
The WELD version 1.5 dataset, used for the input Landsat data, does
not include any radiometric correction for the effects of topographic
variations. This limitation may be among the causes of the omission
errors observed in the results of the present study. The importance
of topographic correction on land cover and burned area classification
accuracy has been studied in previous work (Hantson & Chuvieco,
2011; Vanonckelen, Lhermitte, & Van Rompaey, 2013). However,
systematic topographic correction, requiring both high resolution
digital elevation models, and spatially and temporally accurate characterization
of the atmosphere (Tanre, Herman,
0/5000
From: -
To: -
Results (Indonesian) 1: [Copy]
Copied!
This paper presented a methodology to fuse Landsat 30 m spectralreflectance time series with contemporaneous MODIS 1 km active firedetections to generate large area 30 m burned area maps. The methodologyattempts to overcome the limitations of the 16 day Landsat temporalresolution by incorporating daily MODIS active fire detections.It was demonstrated by processing a large annual dataset centered onthe forested ecoregions of the Western United States that covered predominantlyshrubland (36%), evergreen forest (30%), and grassland(15%). Comparison with independently derived MTBS burned area perimetersshowed the absence of systematic commission errors due tospectral changes unrelated to fire, and a good agreement in the identificationof regional burning patterns (Fig. 12). Further research to validatethe product will be required, ideally using independent burned areareference data which complies with the CEOS Cal Val burned area validationprotocol (Boschetti et al., 2009). The comparison with MTBS perimetersquantified a significant discrepancy in areal estimates betweenthe two datasets; while this discrepancy can be partially attributed tothe inclusion of unburned islands within the MTBS burned area perimeters(Sparks et al., 2015), omission errors may also be due to other factors,which are discussed below.The precision in the temporal detection of the day of burning is limitedby the 16 day temporal resolution of the Landsat data. Over theWestern United States the observed uncertainty (median 23 days,75th percentile 48 days) was significantly higher than the uncertaintyof current coarse resolution products, for example, estimated for theMCD45 MODIS global burned area product as 1 day (median) and4 days (75th percentile) (Boschetti, Roy, Justice, & Giglio, 2010). Thisis of particular concern, because areas where the vegetation recoveryismore rapid than the interval between consecutive cloud-free observationsmay be missed by the present method. Use of contemporaneousLandsat 5 and Landsat 7 data, or of contemporaneous Landsat 7 andLandsat 8 data, would provide an 8-day repeat coverage and so enhancedprobability of cloud-free surface observation (Kovalskyy &Roy, 2013; Roy et al., 2014). This combination would also reduce thereporting uncertainty and, by providing a denser time series, enableLandsat remote sensing closer to the fire event and therefore providea stronger spectral category change, likely resulting in improved burnedarea detection.A further limitation of the presented methodology is the requirementthat at least one MODIS active fire detection partially overlapswith one ormore temporally and spatial adjacent burned area candidateobjects; analysis of the results showed that the omission of entireMTBS polygons can be largely attributed to the absence ofcoincident active fire detections. This has the potential of affectingnot only the presented methodology, but all hybrid algorithmswhich use active fire detections combined with reflectance changes.While the test over the Western United States showed that the majorityof fire events could be detected, previous global coverage researchillustrates that in low tree cover ecosystems MODIS fireproducts significantly undersample the burning activity (Roy et al.,2008) and that small burned area patches are less likely than largepatches to have any coincident active fire detections (Hantson,Padilla, Corti, & Chuvieco, 2013). New research will be undertakento minimize this issue. First, incorporation of MODIS Aqua activefire detections will be investigated as the Aqua and Terra overpasstimes are different and so the two MODIS sensors combined providemore active fire detections per day. Incorporation of active fire detectionsfrom geostationary satellites that only detect very largeand hot fires but every 15 to 30 min (Roberts & Wooster 1992;Zhang, Kondragunta, & Roy, 2014) will also be investigated.In addition, research to refine the semantic rule-based change detectionwill be conducted, for instance by further characterizing the preliminarycandidate burned area detections according to themagnitude of spectral changes of suitable spectral band indices(Bastarrika et al., 2011; Stroppiana, Bordogna, Boschetti et al.,2012; Stroppiana, Bordogna, Carrara, et al., 2012).The WELD version 1.5 dataset, used for the input Landsat data, doesnot include any radiometric correction for the effects of topographicvariations. This limitation may be among the causes of the omissionerrors observed in the results of the present study. The importanceof topographic correction on land cover and burned area classificationaccuracy has been studied in previous work (Hantson & Chuvieco,2011; Vanonckelen, Lhermitte, & Van Rompaey, 2013). However,systematic topographic correction, requiring both high resolutiondigital elevation models, and spatially and temporally accurate characterizationof the atmosphere (Tanre, Herman,
Being translated, please wait..
Results (Indonesian) 2:[Copy]
Copied!
Makalah ini disajikan metodologi untuk sekering Landsat 30 m spektral
reflektansi time series dengan kontemporer MODIS 1 km aktif api
deteksi untuk menghasilkan area yang luas peta wilayah 30 m terbakar. Metodologi
mencoba untuk mengatasi keterbatasan dari 16 hari sementara Landsat
resolusi dengan memasukkan harian MODIS deteksi api yang aktif.
Ini ditunjukkan dengan mengolah dataset tahunan yang besar berpusat pada
ekoregion hutan dari Amerika Serikat Barat yang menutupi sebagian besar
semak (36%), evergreen hutan (30%), dan padang rumput
(15%). Perbandingan dengan berasal independen MTBS dibakar sekeliling daerah
menunjukkan tidak adanya kesalahan komisi sistematis karena
perubahan spektral yang tidak terkait dengan api, dan kesepakatan yang baik dalam identifikasi
pola pembakaran daerah (Gambar. 12). Penelitian lebih lanjut untuk memvalidasi
produk akan diperlukan, idealnya menggunakan wilayah dibakar independen
data referensi yang sesuai dengan CEO Cal Val dibakar daerah validasi
protokol (Boschetti et al., 2009). Perbandingan dengan perimeter MTBS
diukur perbedaan yang signifikan dalam perkiraan areal antara
dua dataset; sementara perbedaan ini sebagian dapat dikaitkan dengan
masuknya pulau terbakar dalam MTBS dibakar sekeliling daerah
(Sparks et al., 2015), kesalahan kelalaian mungkin juga karena faktor lain,
yang dibahas di bawah.
Ketepatan dalam deteksi temporal hari pembakaran dibatasi
oleh 16 hari resolusi temporal dari data Landsat. Selama
Amerika Serikat Barat ketidakpastian diamati (median 23 hari,
75 persentil 48 hari) secara signifikan lebih tinggi dari ketidakpastian
produk resolusi kasar saat ini, misalnya, diperkirakan untuk
produk daerah MCD45 MODIS global yang dibakar sebagai 1 hari (median) dan
4 hari (75 persentil) (Boschetti, Roy, Keadilan, & Giglio, 2010). Ini
menjadi perhatian khusus, karena daerah di mana pemulihan vegetasi
ismore cepat dari interval antara pengamatan bebas awan berturut-turut
mungkin terlewatkan oleh metode ini. Penggunaan kontemporer
Landsat 5 dan Landsat 7 data, atau kontemporer Landsat 7 dan
Landsat 8 data, akan memberikan cakupan ulangi 8-hari dan ditingkatkan
probabilitas observasi permukaan bebas awan (Kovalskyy &
Roy, 2013;. Roy et al, 2014). Kombinasi ini juga akan mengurangi
ketidakpastian pelaporan dan, dengan menyediakan lebih padat time series, memungkinkan
Landsat penginderaan jauh lebih dekat dengan peristiwa kebakaran dan karena itu memberikan
perubahan kategori spektral kuat, mungkin mengakibatkan peningkatan terbakar
deteksi daerah.
Keterbatasan lebih lanjut dari metodologi yang disajikan adalah persyaratan
bahwa setidaknya satu MODIS aktif deteksi kebakaran sebagian tumpang tindih
dengan satu ormore temporal dan spasial yang berdekatan terbakar calon daerah
objek; analisis hasil penelitian menunjukkan bahwa kelalaian seluruh
poligon MTBS dapat sebagian besar disebabkan tidak adanya
bertepatan deteksi api yang aktif. Ini memiliki potensi mempengaruhi
tidak hanya metodologi disajikan, tetapi semua algoritma hybrid
yang menggunakan deteksi api yang aktif dikombinasikan dengan perubahan reflektansi.
Sementara tes selama Amerika Serikat Barat menunjukkan bahwa mayoritas
dari peristiwa kebakaran dapat dideteksi, penelitian cakupan global sebelumnya
menggambarkan bahwa tutupan pohon yang rendah ekosistem api MODIS
produk secara signifikan undersample aktivitas pembakaran (Roy et al.,
2008) dan bahwa patch wilayah dibakar kecil kurang mungkin dibandingkan besar
patch untuk memiliki bertepatan deteksi api yang aktif (Hantson,
Padilla, Corti, & Chuvieco , 2013). Penelitian baru akan dilakukan
untuk meminimalkan masalah ini. Pertama, penggabungan MODIS Aqua aktif
deteksi api akan diselidiki sebagai Aqua dan Terra layang
kali berbeda dan dua sensor MODIS dikombinasikan menyediakan
deteksi api lebih aktif per hari. Penggabungan deteksi api yang aktif
dari satelit geostasioner yang hanya mendeteksi sangat besar
kebakaran dan panas tapi setiap 15 sampai 30 menit (Roberts & Wooster 1992;
. Zhang, Kondragunta, & Roy, 2014) juga akan diselidiki
Selain itu, penelitian untuk memperbaiki semantik deteksi perubahan berbasis aturan
akan dilakukan, misalnya dengan lebih mencirikan awal
calon dibakar pendeteksian daerah sesuai dengan
besarnya perubahan spektral indeks Band spektral cocok
(Bastarrika et al, 2011;.. Stroppiana, Bordogna, Boschetti et al,
2012 ;.. Stroppiana, Bordogna, Carrara, et al, 2012)
The Weld versi 1.5 dataset, digunakan untuk data Landsat masukan, tidak
tidak termasuk koreksi radiometrik untuk efek topografi
variasi. Keterbatasan ini dapat menjadi salah satu penyebab kelalaian
kesalahan diamati pada hasil penelitian ini. Pentingnya
koreksi topografi pada tutupan lahan dan membakar daerah klasifikasi
akurasi telah dipelajari dalam pekerjaan sebelumnya (Hantson & Chuvieco,
2011; Vanonckelen, Lhermitte, & Van Rompaey, 2013). Namun,
sistematis koreksi topografi, membutuhkan baik resolusi tinggi
model elevasi digital, dan spasial dan karakterisasi temporal akurat
dari atmosfer (Tanre, Herman,
Being translated, please wait..
 
Other languages
The translation tool support: Afrikaans, Albanian, Amharic, Arabic, Armenian, Azerbaijani, Basque, Belarusian, Bengali, Bosnian, Bulgarian, Catalan, Cebuano, Chichewa, Chinese, Chinese Traditional, Corsican, Croatian, Czech, Danish, Detect language, Dutch, English, Esperanto, Estonian, Filipino, Finnish, French, Frisian, Galician, Georgian, German, Greek, Gujarati, Haitian Creole, Hausa, Hawaiian, Hebrew, Hindi, Hmong, Hungarian, Icelandic, Igbo, Indonesian, Irish, Italian, Japanese, Javanese, Kannada, Kazakh, Khmer, Kinyarwanda, Klingon, Korean, Kurdish (Kurmanji), Kyrgyz, Lao, Latin, Latvian, Lithuanian, Luxembourgish, Macedonian, Malagasy, Malay, Malayalam, Maltese, Maori, Marathi, Mongolian, Myanmar (Burmese), Nepali, Norwegian, Odia (Oriya), Pashto, Persian, Polish, Portuguese, Punjabi, Romanian, Russian, Samoan, Scots Gaelic, Serbian, Sesotho, Shona, Sindhi, Sinhala, Slovak, Slovenian, Somali, Spanish, Sundanese, Swahili, Swedish, Tajik, Tamil, Tatar, Telugu, Thai, Turkish, Turkmen, Ukrainian, Urdu, Uyghur, Uzbek, Vietnamese, Welsh, Xhosa, Yiddish, Yoruba, Zulu, Language translation.

Copyright ©2025 I Love Translation. All reserved.

E-mail: