3. Methodology for mapping burnt area in the RussianFederationOur algo translation - 3. Methodology for mapping burnt area in the RussianFederationOur algo Indonesian how to say

3. Methodology for mapping burnt ar

3. Methodology for mapping burnt area in the Russian
Federation
Our algorithm first creates a single monthly composite
from the three S10 syntheses available each month. The
seven monthly composites covering April to October are
then used to map monthly burnt area in the 6-month May–
October ‘fire season.’ For fires occurring in each month,
the previous and current monthly composites are regarded
as the pre- and post-fire images, respectively. This approach
is clearly more data intensive than methods using
just start- and end-of-season imagery to provide annual
burned area totals (e.g., Fraser & Li, 2002; Fraser, Li, &
Cihlar, 2000; Kasischke & French, 1995). However, we
believe that our monthly methodology may well detect a
greater proportion of burnt areas, particularly in areas of
low-to-zero forest cover where regrowth of grassy vegetation
may mask some fires that occurred early in the season.
In any case, the provision of monthly data will improve
investigations between fire activity and local climate variables
and allows the derived emissions estimates to be
better compared to simultaneous ground-based or satellitederived
atmospheric chemistry observations (Drummond,
1992; Oberlander, Brenninkmeijerm, Crutzen, Lelieveld, &
Elansky, 2002).
Because NIR reflectance generally falls markedly on
burning, the remote detection of newly burnt areas is commonly
based on NIR reflectance thresholding, or observation
of a significant NIR reflectance change (Fraser, Li, & Cihlar,
2000; Stroppiana et al., 2002). Investigation of spectroradiometric
temporal changes occurring in the S10 data products,
in concert with analysis of the USGS Global Land
Cover Database (http://edcdaac.usgs.gov/glcc/glcc.html;
Anderson, Hardy, Roach, & Witmer, 1976; Brown, Loveland,
Ohlen, & Zhu, 1999) confirmed that a NIR reflectance
decrease on burning does indeed occur for fires in all types of
vegetation found in Russia. However, one important consideration
is the fact that snow, ice, and frost frequently cover
large parts of Russia, and the strong NIR reflectance of these
surfaces can cause significant problems for the accurate
identification of newly burnt areas. Specifically, major errors
of omission can occur if a newly burnt area is covered by ice,
frost, or snow between the specified pre-fire and post-fire
image dates. Conversely, major errors of commission can
occur when the snow, ice, or frost cover of an unburned
region melts, resulting in a fall in NIR reflectance that may
appear similar to that due to fire. To counter these effects, we
use a NIR-minima criterion to create the monthly composites
from the VGT S10 data. This criterion, which is similar to
that employed by Fraser & Li (2002), has the effect of
preferentially selecting composite pixels in their non-ice/
snow/cloud covered state, and also selecting them in their
burned state if they were subject to fire that month. A SWIR
threshold equivalent to a reflectance of 8% was used concurrently
to identify low NIR reflectance pixels that were due
to cloud shadow, and not to burning, and therefore excluded these from the compositing process. This SWIR threshold
was determined via a training set of burned/unburned/cloudy
regions.
The reasoning behind our methodology is demonstrated
further in Fig. 2, which presents a NIR and SWIR spectral
reflectance time series for two typical Russian forest areas,
one that burns between 21 July and 1 August and another
that remains unburned. The high NIR reflectance shown in
(a) and the relatively modest SWIR reflectance shown in
(b) for both areas in VGT data acquired before 1 May
2001 and after 11 September 2001 indicate heavy snow
cover during these periods. For the burned pixel, a sharp
fall in (a) NIR and (b) SWIR reflectance is observed
between the S10 data labelled as 21/07 and 01/08,
corresponding to the pre- and post-fire S10 composites,
respectively. For the burned area, the most significant NIR
and SWIR reflectance decrease outside of the April period
of snow cover melting occurs coincident with the fire event
(21 July to 1 August). The unburnt pixel shows no
reflectance decrease at that time but does show significant
decreases in NIR and SWIR reflectance at other times, but
these have nothing to do with burning. The first occurs
between 01/07 and 11/07, the second between 01/09 and
11/09, and both are caused by the appearance of surface
ice or frost in the 01/07 and 21/08 S10 products. This
increases the NIR and SWIR surface reflectance, and the
subsequent melting causes the observed reflectance decrease
in the S10 products. Fig. 2c and d indicates how the decreases in NIR and
SWIR reflectance observed in the July and September S10 data of the unburned area are absent when we move to our
derived monthly composites. This is because use of the
NIR-minima criterion used to form the monthly composite
excludes pixels with snow, ice, frost, or cloud cover if any
of the three input S10 products have the pixel unaffected
by these pheno
0/5000
From: -
To: -
Results (Indonesian) 1: [Copy]
Copied!
3. Methodology for mapping burnt area in the RussianFederationOur algorithm first creates a single monthly compositefrom the three S10 syntheses available each month. Theseven monthly composites covering April to October arethen used to map monthly burnt area in the 6-month May–October ‘fire season.’ For fires occurring in each month,the previous and current monthly composites are regardedas the pre- and post-fire images, respectively. This approachis clearly more data intensive than methods usingjust start- and end-of-season imagery to provide annualburned area totals (e.g., Fraser & Li, 2002; Fraser, Li, &Cihlar, 2000; Kasischke & French, 1995). However, webelieve that our monthly methodology may well detect agreater proportion of burnt areas, particularly in areas oflow-to-zero forest cover where regrowth of grassy vegetationmay mask some fires that occurred early in the season.In any case, the provision of monthly data will improveinvestigations between fire activity and local climate variablesand allows the derived emissions estimates to bebetter compared to simultaneous ground-based or satellitederivedatmospheric chemistry observations (Drummond,1992; Oberlander, Brenninkmeijerm, Crutzen, Lelieveld, &Elansky, 2002).Because NIR reflectance generally falls markedly onburning, the remote detection of newly burnt areas is commonlybased on NIR reflectance thresholding, or observationof a significant NIR reflectance change (Fraser, Li, & Cihlar,2000; Stroppiana et al., 2002). Investigation of spectroradiometrictemporal changes occurring in the S10 data products,in concert with analysis of the USGS Global LandCover Database (http://edcdaac.usgs.gov/glcc/glcc.html;Anderson, Hardy, Roach, & Witmer, 1976; Brown, Loveland,Ohlen, & Zhu, 1999) confirmed that a NIR reflectancedecrease on burning does indeed occur for fires in all types ofvegetation found in Russia. However, one important considerationis the fact that snow, ice, and frost frequently coverlarge parts of Russia, and the strong NIR reflectance of thesesurfaces can cause significant problems for the accurateidentification of newly burnt areas. Specifically, major errorsof omission can occur if a newly burnt area is covered by ice,frost, or snow between the specified pre-fire and post-fireimage dates. Conversely, major errors of commission canoccur when the snow, ice, or frost cover of an unburnedregion melts, resulting in a fall in NIR reflectance that mayappear similar to that due to fire. To counter these effects, weuse a NIR-minima criterion to create the monthly compositesfrom the VGT S10 data. This criterion, which is similar tothat employed by Fraser & Li (2002), has the effect ofpreferentially selecting composite pixels in their non-ice/snow/cloud covered state, and also selecting them in theirburned state if they were subject to fire that month. A SWIRthreshold equivalent to a reflectance of 8% was used concurrentlyto identify low NIR reflectance pixels that were dueto cloud shadow, and not to burning, and therefore excluded these from the compositing process. This SWIR thresholdwas determined via a training set of burned/unburned/cloudyregions.The reasoning behind our methodology is demonstratedfurther in Fig. 2, which presents a NIR and SWIR spectralreflectance time series for two typical Russian forest areas,one that burns between 21 July and 1 August and anotherthat remains unburned. The high NIR reflectance shown in(a) and the relatively modest SWIR reflectance shown in(b) for both areas in VGT data acquired before 1 May2001 and after 11 September 2001 indicate heavy snowcover during these periods. For the burned pixel, a sharpfall in (a) NIR and (b) SWIR reflectance is observedbetween the S10 data labelled as 21/07 and 01/08,corresponding to the pre- and post-fire S10 composites,respectively. For the burned area, the most significant NIRand SWIR reflectance decrease outside of the April periodof snow cover melting occurs coincident with the fire event(21 July to 1 August). The unburnt pixel shows noreflectance decrease at that time but does show significantdecreases in NIR and SWIR reflectance at other times, butthese have nothing to do with burning. The first occursbetween 01/07 and 11/07, the second between 01/09 and11/09, and both are caused by the appearance of surfaceice or frost in the 01/07 and 21/08 S10 products. Thisincreases the NIR and SWIR surface reflectance, and thesubsequent melting causes the observed reflectance decreasein the S10 products. Fig. 2c and d indicates how the decreases in NIR andSWIR reflectance observed in the July and September S10 data of the unburned area are absent when we move to ourderived monthly composites. This is because use of theNIR-minima criterion used to form the monthly compositeexcludes pixels with snow, ice, frost, or cloud cover if anyof the three input S10 products have the pixel unaffectedby these pheno
Being translated, please wait..
Results (Indonesian) 2:[Copy]
Copied!
3. Metodologi untuk daerah terbakar pemetaan di Rusia
Federasi
algoritma kami pertama menciptakan komposit bulanan
dari tiga S10 sintesis tersedia setiap bulan. The
tujuh komposit bulanan meliputi April-Oktober yang
kemudian digunakan untuk memetakan daerah bulanan terbakar di 6-bulan Mei-
Oktober 'musim kebakaran.' Untuk kebakaran terjadi di setiap bulan,
komposit bulanan sebelumnya dan saat ini dianggap
sebagai gambar sebelum dan sesudah api, masing-masing. Pendekatan ini
jelas lebih banyak data yang intensif daripada metode menggunakan
hanya start- dan akhir-of-musim citra untuk memberikan tahunan
total area yang terbakar (misalnya, Fraser & Li, 2002; Fraser, Li, &
Cihlar, 2000; Kasischke & French, 1995) . Namun, kami
percaya bahwa metodologi bulanan kami mungkin mendeteksi
proporsi yang lebih besar dari kawasan yang terbakar, terutama di daerah-daerah
tutupan hutan rendah ke nol di mana pertumbuhan kembali vegetasi rumput
dapat menutupi beberapa kebakaran yang terjadi di awal musim.
Dalam kasus apapun, penyediaan data bulanan akan meningkatkan
penyelidikan antara aktivitas kebakaran dan variabel iklim lokal
dan memungkinkan perkiraan emisi yang berasal akan
lebih baik dibandingkan dengan simultan atau satellitederived tanah berbasis
pengamatan kimia atmosfer (Drummond,
1992; Oberlander, Brenninkmeijerm, Crutzen, Lelieveld, &
Elansky, 2002).
Karena NIR reflektansi umumnya jatuh tajam pada
pembakaran, deteksi terpencil daerah yang baru terbakar umumnya
berdasarkan NIR pemantulan thresholding, atau pengamatan
dari NIR signifikan perubahan reflektansi (Fraser, Li, & Cihlar,
2000;. Stroppiana et al, 2002). Investigasi spectroradiometric
perubahan sementara yang terjadi di data produk S10,
dalam konser dengan analisis USGS Global Land
Sampul Database (http://edcdaac.usgs.gov/glcc/glcc.html;
Anderson, Hardy, Roach, & Witmer, 1976 ; Brown, Loveland,
Ohlen, & Zhu, 1999) menegaskan bahwa NIR reflektansi
penurunan pada pembakaran memang terjadi karena kebakaran di semua jenis
vegetasi yang ditemukan di Rusia. Namun, salah satu pertimbangan penting
adalah kenyataan bahwa salju, es, dan embun beku sering menutupi
sebagian besar Rusia, dan reflektansi NIR kuat ini
permukaan dapat menyebabkan masalah yang signifikan untuk akurat
identifikasi daerah yang baru dibakar. Secara khusus, kesalahan utama
dari kelalaian dapat terjadi jika daerah yang baru terbakar ditutupi oleh es,
es, atau salju antara ditentukan pra-api dan pasca-kebakaran
tanggal image. Sebaliknya, kesalahan utama dari komisi dapat
terjadi ketika salju, es, atau penutup es dari sebuah terbakar
wilayah mencair, mengakibatkan penurunan NIR reflektansi yang mungkin
muncul mirip dengan yang disebabkan kebakaran. Untuk mengatasi efek ini, kami
menggunakan kriteria NIR-minima untuk membuat komposit bulanan
dari data VGT S10. Kriteria ini, yang mirip dengan
yang digunakan oleh Fraser & Li (2002), memiliki efek
istimewa memilih piksel komposit di non-es / mereka
salju awan menutupi negara /, dan juga memilih mereka di
negara terbakar jika mereka tunduk api bulan itu. Sebuah SWIR
ambang setara dengan pemantulan 8% digunakan secara bersamaan
untuk mengidentifikasi piksel NIR reflektansi rendah yang disebabkan
ke awan shadow, dan tidak terbakar, dan karena itu dikecualikan tersebut dari proses compositing. Ambang SWIR ini
ditentukan melalui serangkaian pelatihan dibakar / terbakar / berawan
daerah.
Alasan di balik metodologi kami ditunjukkan
lebih lanjut dalam Gambar. 2, yang menyajikan NIR dan SWIR spektral
reflektansi time series selama dua kawasan hutan Rusia khas,
salah satu yang membakar antara 21 Juli dan 1 Agustus dan lain
yang tetap tidak terbakar. Reflektansi NIR tinggi ditunjukkan pada
(a) dan pemantulan SWIR relatif sederhana ditunjukkan pada
(b) untuk kedua daerah dalam data VGT diperoleh sebelum 1 Mei
2001 dan setelah 11 September 2001 menunjukkan salju berat
penutup selama periode ini. Untuk pixel terbakar, tajam
penurunan (a) NIR dan (b) SWIR reflektansi diamati
antara data S10 dicap sebagai 21/07 dan 01/08,
sesuai dengan pra dan S10 pasca-api komposit,
masing-masing. Untuk area yang terbakar, yang paling signifikan NIR
dan SWIR reflektansi penurunan luar periode April
penutup salju mencair terjadi bertepatan dengan acara api
(21-01 Juli Agustus). Pixel yang tidak terbakar tidak menunjukkan
penurunan reflektansi pada waktu itu tapi tidak menunjukkan signifikan
penurunan NIR dan SWIR reflektansi pada waktu lain, tapi
ini tidak ada hubungannya dengan pembakaran. Yang pertama terjadi
antara 01/07 dan 07/11, kedua antara 01/09 dan
11/09, dan keduanya disebabkan oleh penampilan permukaan
es atau beku dalam produk 01/07 dan 21/08 S10. Ini
meningkatkan NIR dan SWIR permukaan reflektansi, dan
mencair selanjutnya menyebabkan pemantulan penurunan diamati
dalam produk S10. Ara. 2c dan d menunjukkan bagaimana penurunan NIR dan
SWIR reflektansi diamati pada Juli dan September S10 data daerah terbakar tidak hadir ketika kita pindah ke kami
komposit bulanan berasal. Hal ini karena penggunaan
kriteria NIR-minima digunakan untuk membentuk komposit bulanan
mengecualikan piksel dengan salju, es, es, atau awan jika salah
satu dari tiga masukan produk S10 memiliki pixel terpengaruh
oleh pheno ini
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 ©2024 I Love Translation. All reserved.

E-mail: