1
Scien i ic Da a | (2025) 12:1672 | h ps://doi.o g/10.1038/s41597-025-05949-y
www.na u e.com/scien i icda a
Full-scope ca bon dioxide emission
da ase o Chinese ci ies in 2023
Fanxin Meng 1 ✉ , Hanbo Hu1, Yu ong Sun1, Li Zhang 2 ✉ , Jiaqi Hou1, Zhe Zhang3 ✉ ,
Lingyun Pang3, Bo eng Cai3 & Yuli Shan 4,5 ✉
Ci ies play a c ucial ole in implemen ing ca bon educ ion s a egies and a e essen ial adminis a i e
in his e o . A ull-scope ci y CO2 emission in en o y is impo an o designing e ec i e emission
con ol s a egies and is hus c ucial o achie ing China’s ca bon peaking and neu ali y goals.
Howe e , ecen esea ch has ocused on Scope 1 and Scope 2 emissions, wi h insu icien a en ion
paid o Scope 3 emissions. In his s udy, we cons uc a me hodological model o ull-scope ca bon
emission accoun ing a he ci y le el and es ablish a da ase ha includes Scope 1, 2 and 3 emissions
o Chinese ci ies in 2023. This da ase p o ides aluable da a suppo o in e ci y compa isons and
in aci y managemen o ci y-wide ca bon emissions. The o al ca bon emissions ac oss Chinese ci ies
in 2023 show signi ican quan i a i e and spa ial di e ences. No ably, he emissions in he op 10 ci ies
a e almos 90 o 160 imes highe han hose in he bo om 10 ci ies. Ci ies wi h highe Scope 1 and
2 ca bon emissions a e p edominan ly loca ed in he sou heas e n coas al a eas, whe eas ci ies wi h
highe Scope 3 ca bon emissions a e concen a ed in apidly de eloping a eas o cen al China.
Backg ound & Summa y
As he la ges g eenhouse gas (GHG) emi e globally, China plays an essen ial ole in comba ing clima e change.
In alignmen wi h in e na ional clima e commi men s, China has in oduced ca bon peaking and neu ali y
s a egies, which aim o peak ca bon emissions by 2030 and achie e ca bon neu ali y by 2060, espec i ely.
U ban a eas a e associa ed wi h he consump ion o mo e han 66% o ene gy and he p oduc ion o mo e han
70% o ca bon dioxide (CO2), which is a majo con ibu o o emissions and global clima e impac s and is pi -
o al in ealizing hese s a egies1. Gi en ha ci ies a e esponsible o a signi ican p opo ion o global ca bon
emissions, comp ehensi e ca bon emission disclosu e is c i ical o in o m and suppo he o mula ion and
execu ion o ne -ze o ca bon clima e ac ion plans2.
Comp ehensi e and consis en ca bon emission epo ing is c ucial, as i p o ides ci ies wi h a eliable basis
o o mula ing policies o educe hei ca bon oo p in . Wang e al.3 examined he challenges and oppo uni ies
o Chinese ci ies in implemen ing he coun y’s ca bon peaking and ca bon neu ali y s a egies, he eby p o-
posing ha accu a e u ban emission accoun ing is cen al o achie ing ca bon neu ali y. Zhang e al.4 demon-
s a e ha by he end o 2020 nea ly one- hi d o Chinese ci ies ha e al eady eached ca bon emission peaks;
howe e , a sys ema ic quan i ica ion o o al u ban emissions emains indispensable o elucida ing he empo al
e olu ion o agg ega e ci y-le el ca bon ajec o ies. S anda ds such as he Global P o ocol o Communi y-Scale
G eenhouse Gas Emission In en o ies5 and he US Communi y P o ocol o Accoun ing and Repo ing o
G eenhouse Gas Emissions6 p o ide s uc u ed me hodologies o u ban GHG accoun ing. These p o ocols
p o ide de ini ions o he key emission scope ca ego ies ha mus be conside ed: di ec emissions om sou ces
loca ed wi hin he ci y bounda y (Scope 1), emissions om pu chased elec ici y and hea , s eam and/o cooling
(Scope 2), and all o he emissions occu ing ou side he ci y bounda y (Scope 3)5. By o e ing clea and comp e-
hensi e guidelines, hese s anda ds enable ci ies o assess hei emissions holis ically and de elop a ge ed s a -
egies o educe hei clima e impac s7. Mo eo e , ecen s udies ha e highligh ed he impo ance o imp o ing
he da a quali y and enhancing anspa ency in u ban emissions epo ing. Fo example, Akpuokwe e al.8 and
Bake e al.9 emphasized he need o mo e localized adap a ion o global amewo ks o mee speci ic egional
1School o En i onmen , Beijing No mal Uni e si y, Beijing, 100875, China. 2School o he En i onmen and Sa e y
Enginee ing, Jiangsu Uni e si y, Zhenjiang, 212013, China. 3Cen e o Ca bon Neu ali y, Chinese Academy o
En i onmen al Planning, Beijing, 100043, China. 4School o Geog aphy, Ea h, and En i onmen al Sciences,
Uni e si y o Bi mingham, Bi mingham, B15 2TT, UK. 5Bi mingham Ins i u e o Sus ainabili y and Clima e Ac ion
(BISCA), Uni e si y o Bi mingham, Bi mingham, B15 2TT, Uni ed Kingdom. ✉e-mail: meng [email protected];
[email p o ec ed]; [email p o ec ed]; y[email p o ec ed]
DATA DeSC IpTO
OpeN
2
Scien i ic Da a | (2025) 12:1672 | h ps://doi.o g/10.1038/s41597-025-05949-y
www.na u e.com/scien i icda a
www.na u e.com/scien i icda a/
o na ional policy needs; Lin on e al.10 no ed he lack o in eg a ion be ween u ban emission da a and b oade
clima e policies, and hey ad oca ed o mo e coo dina ed app oaches o emission go e nance. Addi ionally, he
Uni ed Na ions Sus ainable De elopmen Goals (SDGs) epo explici ly desc ibes he equi emen o accu a e
u ban ca bon emission accoun ing h ough wo a ge ed objec i es (i.e., SDG 11 on sus ainable ci ies and com-
muni ies and SDG 13 on clima e ac ion)11; while compelling na ions o ins i u ionalize clima e- esponsi e go -
e nance amewo ks o u ban sys ems and o de elop ad anced ci y-scale ca bon accoun ing me hods is u gen ,
and many e o s ha e been made in ecen yea s12, Huo ila, e al.13 emphasized he need o s anda dized assess-
men me hods o align and guide ci ies owa ds ca bon neu ali y, he eby educing p oblems caused by he lack
o consis ency in u ban ca bon accoun ing me hods and emission anges. These s udies collec i ely demons a e
he inc easing ecogni ion o he need o s anda dized ye locally adap able amewo ks, which ein o ces he
ole o u ban ca bon accoun ing in achie ing clima e change mi iga ion and u ban sus ainable de elopmen .
Howe e , gi en he inhe en ly open-sys em na u e o u ban en i onmen s, he delinea ion o ansbounda y ca -
bon accoun ing bounda ies emains a pe sis en challenge14, necessi a ing u he schola ly esea ch on me hod-
ological amewo ks and da a collec ion p o ocols o es ablishing ci y-scale emission in en o ies15.
Al hough exis ing esea ch highligh s he impo ance o ca bon emissions om u ban supply chains, cu en
ca bon accoun ing s udies ocus p ima ily on Scope 1 and 2 emissions; Shan e al.16 de eloped an in en o y o
di ec emissions (Scope 1) o e a long ime scale o 287 ci ies o suppo he o mula ion o ca bon mi iga ion
policies. Adop ing Shanghai as a case s udy, Wei e al.17 quan i ied Scope 1 elec ici y- ela ed ca bon emissions
and comp ehensi ely analysed he unde lying d i ing ac o s, Liu e al.18 examine me hodologies o quan i ying
he ca bon oo p in s o megaci ies, unde sco ing he impe a i e o conduc ing comp ehensi e g eenhouse-gas
emission in en o ies in densely popula ed, economically dynamic u ban cen e s. Zhang e al.19 conduc ed a
comp ehensi e assessmen o Shanxi P o ince’s pa hways owa d ca bon-peak and ca bon-neu ali y, igo ously
quan i ying he in luences o egional cha ac e is ics—including esou ce endowmen s, ene gy-consump ion
pa e ns, and indus ial s uc u es—on p o incial emissions. Shan e al.20 in es iga ed Scope 1 ca bon emissions
a ibu able o bo h p oduc ion p ocesses and ossil uel consump ion in he Chinese lime manu ac u ing sec o ,
p o iding a sys ema ic assessmen o emission sou ces in his ene gy-in ensi e indus y. Addi ionally, ou p e-
ious esea ch aimed o p o ide a high- esolu ion ca bon emission in en o y o China, co e ing bo h Scope 1
and 2 emissions21. Howe e , less a en ion has been gi en o Scope 3 emissions. Zhu e al.22 epo ed ha Scope
3 emissions gene ally exceed Scope 1 o 2 ca bon emissions, e en su passing he o al o bo h ypes o emissions.
Kucuk a e al.23 e ealed ha supply chain- ela ed indi ec emissions ( ep esen ed by Scope 3) a e esponsible
o nea ly 56.5% o he o al ca bon emissions o a ious sec o s in Tu key. Wi hou a comple e and accu a e
accoun ing o ci y-le el ca bon emissions, p omo ing an equi able dis ibu ion o emission educ ion espon-
sibili ies among ci ies becomes challenging. This ine iciency may disp opo iona ely bu den ups eam ci ies
wi hin he supply chain wi h he esponsibili y o educing emissions. A comp ehensi e ci y-le el accoun ing
o Scope 3 emissions has been implemen ed o 79 membe s o he C40 Ci ies Clima e Leade ship G oup24, and
he esul s e ealed ha igno ing Scope 3 ca bon emissions could esul in unde es ima ion o he global GHG
emissions by 4%. Zhang e al.25 p o ide a comp ehensi e quan i ica ion o Wuyishan’s g eenhouse-gas emis-
sion p o ile, e ealing ha 42% o he ci y’s o al in en o y o igina es beyond i s adminis a i e bounda y. This
indings unde sco e he impe a i e o ci ies o explici ly inco po a e ex a- e i o ial sou ces when compiling
emission accoun s and o mula ing mi iga ion s a egies. Zheng e al.26 in es iga ed ca bon leakage induced by
in e ci y supply chains ac oss 309 Chinese ci ies and e ealed ha u ban supply chain ne wo ks a e associa ed
wi h app oxima ely 80% o he o al ca bon emissions. Mo eo e , Xia e al.27 epo ed ha in 2017, he emission
educ ion achie emen s o nea ly 186 Chinese ci ies we e signi ican ly in luenced by he mi iga ion e o s o
ups eam ci ies wi hin hei supply chains. Hence, he e is a lack o a uni ied easibili y amewo k o consid-
e ing comp ehensi e Scope 1–3 emissions a he ci y le el, posing challenges in es ablishing a uni ied u ban
ca bon emissions da ase .
To b idge hese esea ch gaps, we compiled ci y-le el ull-scope CO2 emission da a o all 339 adminis a i e
ci ies in China. Ou da ase includes Scope 1 emissions om indus ial ene gy use, indus ial p ocesses, build-
ings, anspo a ion, and ag icul u e; Scope 2 emissions om ex e nal powe ans e s; and Scope 3 emissions,
which encompass emissions om supply chain ac i i ies associa ed wi h 10 key ma e ials impo ed in o ci ies.
In his s udy, we employed he Mon e Ca lo simula ion me hod o quan i y he unce ain ies in ca bon emis-
sion da a, wi h con as alida ion agains p io esea ch indings o ensu e he c edibili y o he da ase . Ou
ull-scope emission da ase o e s a no el, holis ic pe spec i e o e alua ing he o al ca bon emissions in ci ies.
By p o iding up- o-da e, consis en , and anspa en da a ac oss Scopes 1, 2, and 3, his da ase plays an impo -
an ole in ad ancing he de elopmen o ca bon da a in China. I no only p o ides a mo e accu a e assessmen
o he ca bon oo p in bu also se es as a c ucial ool o achie ing an equi able dis ibu ion o ca bon espon-
sibili y among ci ies, suppo ing in o med policy decisions and e ec i e clima e ac ion a he local le el.
Me hods
We adop ed he IPCC- ecommended baseline me hodology28 (i.e., he emission ac o app oach) o accoun o
ca bon emissions in ci ies ac oss all h ee scopes. The ull-scope ca bon emission accoun ing and e i ica ion
model wo k low comp ised he ollowing i e p incipal phases:
(1) Compila ion o ac i i y da a, he p ima y da a o Scope 1 and Scope 2 we e ob ained om he China
High-Resolu ion Emission Da abase(CHRED)4,29 and China Ca bon Wa ch (CCW)30, whe eas he p ima-
y da a o Scope 3 we e acqui ed h ough he China Economic and Social Big Da a Resea ch Pla o m31.
Following da a compila ion and cleaning, he equisi e ac i i y-le el da a we e de i ed. De ailed sou ces
o hese da a a e p o ided in he “Ac i i y Da a” subsec ion o his chap e .Table 1 delinea ed, in de ail, he
p ima y da a sou ces employed o he quan i ica ion o Scope 1, Scope 2, and Scope 3 emissions;
3
Scien i ic Da a | (2025) 12:1672 | h ps://doi.o g/10.1038/s41597-025-05949-y
www.na u e.com/scien i icda a
www.na u e.com/scien i icda a/
(2) Fo he acquisi ion o emission ac o s, please e e o Supplemen a yTable A2 o sec o -speci ic emission
ac o s and hei sou ces ca ego ized by scope;
(3) Calcula ion o ca bon emissions by Scope, he ca bon emissions o each scope we e calcula ed by subs i-
u ing he co esponding ac i i y-le el da a and emission ac o da a in o he espec i e o mulas p esen ed
below, he eby de i ing he emissions o each scope.
(4) Implemen a ion o unce ain y analysis and alida ion p o ocols, whe ein Mon e Ca lo simula ions we e
employed o quan i y da a unce ain ies, complemen ed by sys ema ic con as e i ica ion agains es ab-
lished da abases o ensu e me hodological eliabili y.
(5) Da a isualiza ion, he spa ial dis ibu ion o ca bon emissions was isualized using A cGIS ( e sion 10.8)
o gene a e ca bon emission maps, acili a ing he p esen a ion o his s udy’s da ase .
Sys em bounda y o ull-scope emissions. As shown in Fig.1, in his da ase , Scope 1 ca bon emis-
sions e e ed o he di ec emissions esul ing om he combus ion o ossil uels ac oss key sec o s, including
indus ial ene gy, buildings, ag icul u e, and anspo a ion, as well as emissions om indus ial p ocesses and
ac i i ies. Ca bon emissions o abso p ion associa ed wi h de o es a ion o land-use changes we e no included.
Scope 2 emissions e e ed o he ca bon emissions caused by impo ing elec ici y om ou side p e ec u e-le el
ci ies. As demons a ed by Hillman e al.32, ood, wa e , and cemen (shel e ) cons i u ed essen ial ma e ial pilla s
o u ban ope a ions. These c i ical ma e ials we e p ima ily p oduced beyond u ban adminis a i e bounda-
ies, hus necessi a ing c oss-bo de anspo a ion ne wo ks o sus ain me opoli an unc ions. On his basis,
we employed hyb id analysis33 o iden i y en key ma e ials o u ban cons uc ion and ope a ion p ocesses as
speci ic p oduc s o Scope 3 ca bon emission accoun ing, including se en majo ag icul u al p oduc s— ice,
whea , co n, po k, bee , mu on, and poul y—and wo key cons uc ion ma e ials—s eel and cemen —as well
as wa e , which was classi ied as a esou ce p oduc . The a ionale o selec ing hese ma e ials in his accoun ing
amewo k was based on he ollowing conside a ions: he ag icul u al p oduc ion sec o , encompassing bo h
c op cul i a ion and li es ock a ming, ep esen ed he second-la ges sou ce o ca bon emissions in China34; he
s eel and cemen indus ies collec i ely accoun ed o app oxima ely 15% o he global an h opogenic CO2 emis-
sions35; as he wo ld’s la ges de eloping na ion, Chinese ci ies we e p ojec ed o expe ience sus ained g ow h
in i s demand o hese ma e ials36,37; and wa e -induced ca bon emissions we e mani es ed p ima ily h ough
ene gy consump ion associa ed wi h wa e - ela ed ac i i ies, which cu en ly accoun ed o app oxima ely 2%
o he na ion’s o al elec ici y usage. This p opo ion was p ojec ed o inc ease signi ican ly o e he coming
decades38.
Scope 1 me hodology. Scope 1 emissions we e calcula ed ia Eq. (1).
∑
=×+
CE QEFCE(1)
ScopeFueluse ii Indus ial p ocess1,
P o ince Ci y Scope 1 Da a
Sou ce Scope 2 Da a
Sou ce
Scope 3 Da a Sou ce
Ag icul u al
p oduc s S eel Cemen Wa e
Beijing Beijing CHRED 3.04,29 Beijing S a is ical
Yea book 202461
Beijing
S a is ical
Yea book
2024
S a is ical Bulle in
on Na ional
Economic
and Social
De elopmen in
202362
Beijing 2023 Wa e
Resou ces Bulle in63
Tianjin Tianjin CHRED 3.0 Tianjin S a is ical
Yea book 202464 Tianjin S a is ical Yea book
2024 Tianjin 2023 Wa e
Resou ces Bulle in65
Hebei Shijiazhuang CHRED 3.0 Hebei S a is ical
Yea book 202466 Hebei S a is ical Yea book
2024
Bulle in o Wa e
Resou ces o Hebei
P o ince in 202367
Hebei Tangshan CHRED 3.0 Hebei S a is ical
Yea book 2024 Hebei S a is ical Yea book
2023
Bulle in o Wa e
Resou ces o Hebei
P o ince in 2023
Hebei Qinhuangdao CHRED 3.0 Hebei S a is ical
Yea book 2024 Hebei S a is ical Yea book
2023
Bulle in o Wa e
Resou ces o Hebei
P o ince in 2023
Hebei Handan CHRED 3.0 Hebei S a is ical
Yea book 2024 Hebei S a is ical Yea book
2023
Bulle in o Wa e
Resou ces o Hebei
P o ince in 2023
Hebei Xing ai CHRED 3.0 Hebei S a is ical
Yea book 2024 Hebei S a is ical Yea book
2023
Bulle in o Wa e
Resou ces o Hebei
P o ince in 2023
Table 1. Table o P ima y Da a Sou ces (Selec ed Ci ies). No e: Fo he P ima y da a sou ces o he emaining
ci ies, please e e o Supplemen a y TableA1.
4
Scien i ic Da a | (2025) 12:1672 | h ps://doi.o g/10.1038/s41597-025-05949-y
www.na u e.com/scien i icda a
www.na u e.com/scien i icda a/
whe e QFuel use,i deno es he di e en ypes o uels used in indus ial, building, anspo a ion and ag icul u al
sec o s; EFi deno es he emission ac o o a ious ypes o uels; and CEIndus ial p ocess deno es he ca bon emission
da a o indus ial p ocesses in local p oduc ion.
Scope 2 me hodology. The assump ions made in he Scope 2 me hodology we e as ollows:
1) The elec ici y p oduced in a gi en ci y was p io i ized o local supply.
Scope 2 emissions we e calcula ed ia Eqs. (2) and (3).
CE QEF(2)
ScopeCi yelec ici y impo Powe g id2
=×
QQ Q(3)
Ci y elec ici yimpo Ci ypowe consump ionCi ypowe gene a ion
=−
whe e QCi y elec ici y impo deno es he amoun o elec ici y impo ed; EFPowe g id deno es he emission ac o s
o egional powe g ids; QCi y powe consump ion deno es he amoun o elec ici y consumed; and QCi y powe gene a ion
deno es he amoun o elec ici y p oduced wi hin he ci y i sel . We assumed ha locally p oduced elec ici y is
p io i ized o local consump ion. I QCi y elec ici y impo is nega i e, his indica es ha his ci y expo s elec ici y
on a ne basis and does no impo ou side elec ici y.
Scope 3 me hodology. The assump ions made in he Scope 3 me hodology we e as ollows:
1) The impo olume o he ci y was eco ded as he di e ence be ween he ci y’s demand and supply.
2) P oduc s manu ac u ed wi hin he ci y we e p io i ized o local supply, whe eas c i ical ma e ial sho ages
esul ing om inal demand ac i i ies we e add essed h ough ex e nal sou cing o achie e a supply‒de-
mand balance wi hou conside ing he in en o y.
Scope 3 emissions we e calcula ed ia Eqs. (4) and (5).
∑
=×
=
CE QEF
(4)
Scope
i
n
impo ii3
1
,
=−QQ Q(5)
impo idemandi supplyi,,,
whe e Qimp o ,i deno es he impo olume o he i- h ma e ial, i deno es he ype o ma e ial being accoun ed
o , EFi is he ca bon emission ac o o he i- h p oduc ’s supply chain, Qdemand,i deno es he demand o he i- h
p oduc , and Qsupply,i deno es he supply olume o he i- h p oduc .
ac i i y da a
Scope 1 ac i i y da a. Ac i i y da a o Scope 1 ca bon emissions we e de i ed mainly om en e p ise-le el
da a o he China High-Resolu ion Emission Da abase4,29 (CHRED 3.0) and p e ec u e-le el ci y da a om China
Ca bon Wa ch30.
Speci ically, he accoun ing o emissions in he elec ici y sec o was based on accoun ing models de el-
oped om da a such as en e p ise GHG e i ica ion epo s, ci y s a is ical yea books, and indus ial elec ici y
consump ion.
Fig. 1 Mul i a ia e model o ull-scope ca bon emission accoun ing.
5
Scien i ic Da a | (2025) 12:1672 | h ps://doi.o g/10.1038/s41597-025-05949-y
www.na u e.com/scien i icda a
www.na u e.com/scien i icda a/
Fo he indus ial sec o , he model was based on da a om he CHRED da abase, ca bon emission e i i-
ca ion da a, annual epo s o en i onmen al s a is ics, qua e ly epo s o en e p ises in key indus ies, and
online moni o ing da a om en e p ises.
Road anspo da a we e ob ained ia an accoun ing model es ablished on he basis o he daily u ban oad
ne wo k speed and conges ion index.
Wa e anspo and ailway da a we e modelled on he basis o he CHRED da abase and mon hly u no e
da a.
A ia ion da a we e modelled on he basis o daily in o ma ion on he numbe o ligh s pe ai po and he
engine pa ame e s o a ious ai c a ypes.
Da a o he cons uc ion sec o we e modelled on he basis o Suomi-NPP VIIRS nigh ligh da a, land use
da a and CHRED da a.
Da a o he ag icul u e sec o we e modelled on he basis o he alue added o ag icul u e, o es y and
ishe ies and he CHRED da abase.
Scope 2 ac i i y da a. Ac i i y da a o Scope 2 emissions included powe gene a ion om ossil ene gy
sou ces and non ossil ene gy sou ces. Fossil powe gene a ion da a we e de i ed mainly om go e nmen doc-
umen s. Non ossil powe gene a ion da a included ou ca ego ies: wind powe , sola powe , hyd opowe , and
nuclea powe . Single-poin sou ce da a o wind powe , pho o ol aic powe and hyd opowe in China we e
ob ained ia emo e sensing and GIS echnologies. Combined wi h he ac ual p o incial ins alled powe gen-
e a ion capaci y in 2023, s a is ics and GIS echnology we e used o ob ain u ban wind powe , sola powe and
hyd opowe gene a ion da a. Nuclea powe gene a ion da a we e ob ained by using uni -le el gene a ion in o -
ma ion om s a is ics p o ided by he China Nuclea Ene gy Indus y Associa ion39.
Scope 3 ac i i y da a. The supply olume o p oduc s o Scope 3 emission accoun ing can be ela i ely
easily ob ained om go e nmen public da a o s a is ical yea books, whe eas es ima ing he demand olume
is mo e complex. The e o e, in his s udy, p oduc demand da a we e ob ained ia he hyb id analysis me hod.
As shown in Fig.2, on he basis o he di e en sou ces o he o iginal da a, h ee me hods we e designed in his
s udy o accoun o ac i i y le el da a. The inpu da a sou ces used o ca bon accoun ing in each ci y a e docu-
men ed in Table1 (No e: Fo ci ies whose 2024 s a is ical yea book has no ye been eleased, he co esponding
p o incial 2024 s a is ical yea book is used in conjunc ion wi h each ci y’s 2023 s a is ical bulle in o comple e
he calcula ions).
Me hod 1. Me hod 1 in ol ed ob aining he in low quan i ies h ough su eys conduc ed wi h ele an
depa men s. Speci ically, da a on ag icul u al p oduc in lows we e collec ed h ough su eys wi h he ci y’s
Depa men o Ag icul u e o o he depa men s esponsible o ag icul u e; da a on he indus ial p oduc
in lows needed o cons uc ion, oad pa ing, and indus ial p oduc ion we e ob ained om he ci y’s U ban and
Ru al Cons uc ion Bu eau, T a ic Bu eau, and Indus y and Comme ce Bu eau; and wa e in lows we e meas-
u ed h ough su eys wi h he ci y’s Wa e Resou ces Bu eau.
Me hod 2. When only municipal-le el s a is ical sou ces we e a ailable, such as hose documen ed in ci y s a-
is ical yea books and go e nmen bulle ins, Me hod 2 was employed. In Me hod 2, he p oduc ion and demand
quan i ies o each p oduc ype we e calcula ed sepa a ely, wi h he p oduc in low quan i y de i ed om he
di e ence be ween he wo. The speci ic calcula ion me hod was as ollows:
(1) Th ee ypes o ood c ops
CE QQ EF[( 80%)] (6)
Scopeig ainp oduc
i
demand ip oduc ioni i3, ,,
∑
=−××
∈
whe e Qp oduc ion,i deno es he ou pu o ood c ops in s a is ical yea books by ci y; Qdemand,i deno es he ood
c op demand da a by ci y; and EFi deno es he emission ac o s o ood c ops. P oduc ion da a we e di ec -
ly ob ained om annual s a is ical epo s. Acco ding o p e ious s udies, 38% o China’s g ain p oduc ion
was used o mee he ood demand o u ban and u al esiden s, while 42% was used o sa is y he demand
o animal eed. The ac ual g ain supply was es ima ed a 80% o he o al g ain p oduc ion40. The demand
da a we e di ided in o wo pa s: ood consump ion da a o u ban and u al esiden s and g ain demand
da a o li es ock eed. The demand was calcula ed as ollows:
=+
−−
QQ Q(7)
demand idemandg ain o peopledemandg ain o animal,
QPFC (8)
demand g ain o people ig ain p oduc ci y le elpe capi a g aini
,,
=×
−∈−
QPFC EF (9)
demand g ain o animal ig ain p oduc animalsi pe uni animalg ain i ypeo g ains
,,,
=× ×
−∈
whe e Qdemand-g ain o people, i∈g ain p oduc deno es he ood demand quan i y o he i- h g ain ype; Pci y-le el deno es
he u ban esiden popula ion; FCpe capi a g ain, i deno es he pe capi a annual consump ion o g ain ype i in
he ci y; Qdemand-g ain o animals, i∈g ain p oduc deno es he eed g ain demand quan i y o he i- h ype o li es ock;
Panimals, i deno es he annual li es ock popula ion da a; FCpe uni animal g ain, i deno es he amoun o g ain used pe
uni o animal p oduc p oduc ion; and EF ype o g ains deno es he alloca ion a io o g ains used as eed.
6
Scien i ic Da a | (2025) 12:1672 | h ps://doi.o g/10.1038/s41597-025-05949-y
www.na u e.com/scien i icda a
www.na u e.com/scien i icda a/
(2) Fou majo ypes o mea
CE QQ EF[( 90%) ](10)
Scopeimea p oduc
i
demand ip oduc ioni i3, ,,
∑
=−××
∈
whe e Qp oduc ion,i deno es he p oduc ion quan i y o ou majo mea p oduc s om he annual s a is ical
yea books o each ci y; Qdemand,i deno es he demand da a o each ype o mea p oduc ; and EFi deno es
he emission ac o o each ype o mea . P oduc ion da a we e di ec ly ob ained om ci y s a is ical yea -
books. On he basis o p e ious s udies and conside ing he consump ion o mea was e and o al, he ac u-
al mea supply was calcula ed as 90% o he o al mea p oduc ion40. The demand da a we e solely based on
he mea consump ion da a o u ban and u al esiden s. The demand was calcula ed as ollows:
=
−
QQ (11)
demand idemandmea ,
=×
−∈ −
QPFC (12)
demand mea imea p oduc ci y le elpe capi a mea i
,,
whe e Qdemand-mea ,i∈mea p oduc deno es he demand quan i y o he i- h ype o mea p oduc ; Pci y-le el de-
no es he u ban esiden popula ion; and FCpe capi a mea ,i deno es he pe capi a annual consump ion o he
i- h ype o mea a he ci y le el.
(3) S eel
CE QQ EF[( )] (13)
Scopes eelsdemandp oduc ions eels3
∑
=−×
−
whe e Qp oduc ion deno es he s eel p oduc ion quan i y om he annual s a is ical yea books o each ci y,
Qdemand deno es he s eel demand da a o each ci y, and EFs eels deno es he emission ac o o c ude s eel.
C ude s eel p oduc ion da a we e ob ained om ci y s a is ical yea books, whe eas s eel demand da a
we e calcula ed on he basis o h ee componen s, namely, he s eel demand o building cons uc ion, he
s eel demand o anspo a ion in as uc u e cons uc ion, and he s eel demand o consume goods
p oduc ion. The s eel demand o building cons uc ion was de i ed by es ima ing he s eel consump ion
o new buildings by cons uc ion ype in he ci y. The s eel demand o anspo a ion in as uc u e was
calcula ed by es ima ing he s eel consump ion o new ailways and highways in he ci y. The s eel demand
o consume goods p oduc ion was de e mined by measu ing he s eel con en in consump ion goods
pu chased and anspo ed in o he ci y om ex e nal egions o e he pas yea , such as anspo a ion,
Fig. 2 Decision ee o Scope 3 demand p oduc calcula ion.
7
Scien i ic Da a | (2025) 12:1672 | h ps://doi.o g/10.1038/s41597-025-05949-y
www.na u e.com/scien i icda a
www.na u e.com/scien i icda a/
machine y, and household appliances. The calcula ion was as ollows:
=+ +
−− −
QQ QQ (14)
demand demand s eel building demand s eel anspo a ion demand s eel goods,, ,
QASC (15)
demand s eel building jci y le elj uni a ea j,,,
∑
=×
−−
=× +×
−
QLSC LSC(16)
demand s eel anspo a ion ailwayuni leng h ailway high oad uni leng h high oad,
QPSC (17)
demand s eel goodsgguni good,
∑
=×
−
whe e Qdemand-s eel, building, Qdemand-s eel, anspo a ion and Qdemand-s eel,goods deno e he s eel demands o building
cons uc ion, anspo a ion in as uc u e cons uc ion, and consume goods p oduc ion, espec i ely; j
deno es he di e en building ypes; Aci y-le el,j deno es he annual new cons uc ion a ea o building ype
j; and SCuni a ea,j deno es he s eel consump ion pe uni a ea o new cons uc ion o each building ype.
Mo eo e , L ailway and Lhigh oad deno e he annual new cons uc ion mileages o ailways and highways,
espec i ely; SCuni leng h ailway and SCuni leng h high oad deno e he s eel consump ion alues pe uni leng h o
newly cons uc ed ailways and highways, espec i ely; I ailway and Ihigh oad deno e he annual in es men s in
new ailways and highways, espec i ely; SCuni leng h ailway and SCuni leng h high oad deno e he s eel consump-
ion alues pe uni in es men in new ailways and highways, espec i ely; g deno es he majo ypes o
consume goods, including anspo a ion, machine y, household appliances, and o he consume goods;
Pg deno es he quan i y o each majo consume good; and SCuni good deno es he s eel consump ion pe
uni o each majo consume good.
(4) Cemen
CE QQ EF[( )] (18)
Scopecemen demandp oduc ioncemen 3
∑
=−×
−
whe e Qp oduc ion deno es he cemen p oduc ion quan i y om he annual s a is ical yea books o each ci y,
Qdemand deno es he cemen demand da a o each ci y, and EFcemen deno es he emission ac o o cemen .
Cemen p oduc ion da a we e ob ained om s a is ical yea books. Cemen demand da a we e calcula ed
on he basis o h ee componen s, namely, he cemen demand o u ban building cons uc ion, which was
es ima ed by calcula ing he cemen consump ion o new buildings by building ype in he ci y; he cemen
demand o u ban anspo a ion in as uc u e cons uc ion, which was calcula ed by es ima ing he ce-
men consump ion o new ailways and highways in he ci y; and he cemen demand o u ban indus ial
in as uc u e cons uc ion, which was calcula ed by con e ing he indus ial ixed asse in es men in he
ci y in o he le el o cemen consump ion. The calcula ion was as ollows:
=+ +
−− −
QQ QQ (19)
demand demand cemen building demand cemen anspo a ion demand cemen indus y,, ,
∑
=×
−−
QACC
(
20)
demand cemen building jci y le elj uni a ea j,,,
QLCC LCC
(
21)
demand cemen anspo a ion ailway ailway high oad high oad,
=× +×
−
QICC (22)
demand cemen indus y indus y in es men uni alue,,
=×
−
whe e Qdemand-cemen , building, Qdemand-cemen , anspo a ion and Qdemand-cemen ,indus y deno e he cemen demands o
building cons uc ion, anspo a ion in as uc u e cons uc ion, and indus ial in as uc u e cons uc-
ion, espec i ely; Aci y-le el,j deno es he annual new cons uc ion a ea o building ype j, CCuni a ea,j deno es
he cemen consump ion pe uni a ea o new cons uc ion o each building ype; L ailway and Lhigh oad
deno e he annual new cons uc ion mileages o ailways and highways, espec i ely; CC ailway and CC
high oad deno e he cemen consump ion le els pe uni leng h o newly cons uc ed ailways and highways,
espec i ely; Iindus y,in es men deno es he indus ial ixed asse in es men a he o iginal p ice; and CCuni alue
deno es he cemen ma e ial s eng h pe -uni alue.
(5) Wa e
∑
=−×
−
CE QQ EF[( )] (23)
Scopewa e demand p oduc ion wa e 3
whe e Qp oduc ion deno es he wa e esou ce p oduc ion quan i y om he wa e esou ce bulle ins o each
ci y, Qdemand deno es he wa e demand da a o each ci y, and EFwa e deno es he wa e esou ce emission
ac o . Wa e p oduc ion da awe e ob ained om he wa e esou ce bulle ins published by each ci y. The
wa e demand o each ci y was calcula ed as ollows:
=+++
QQ QQQ(24)
demand ag icul u al wa e indus ialwa e domes ic wa e ecologicalwa e
8
Scien i ic Da a | (2025) 12:1672 | h ps://doi.o g/10.1038/s41597-025-05949-y
www.na u e.com/scien i icda a
www.na u e.com/scien i icda a/
whe e Qag icul u al wa e deno es he annual ag icul u al wa e consump ion o each ci y; Qindus ial wa e deno es
he annual indus ial wa e consump ion o each ci y; Qdomes ic wa e deno es he annual domes ic wa e con-
sump ion o each ci y; and Qecological wa e deno es he annual ecological wa e consump ion o each ci y.
When da a a ailabili y was limi ed o p o incial-le el s a is ical sou ces, such as hose ob ained om p o-
incial s a is ical yea books and go e nmen bulle ins, Me hod 3 was applied. In cases o inaccessible p o incial
da a, na ionally s a is ics da a se ed as he al e na i e da a sou ce. In Me hod 3, a downscaling app oach was
adop ed o calcula e he p oduc demand, he eby using p o incial o na ional da a. Me hod 3 was applied only
o es ima e he demand o indus ial p oduc s.
Me hod 3.
(1) S eel
The s eel demand o building cons uc ion was es ima ed by calcula ing he s eel consump ion o new
buildings by building ype a he p o incial o na ional le el. This demand was hen alloca ed on he basis
o he p opo ion o he o al new cons uc ion a ea in he a ge ci y ela i e o he o e all new cons uc-
ion a ea. The s eel demand o anspo a ion in as uc u e cons uc ion was calcula ed by de e mining
he s eel consump ion o new ailways and highways a he p o incial o na ional le el, which was hen
alloca ed on he basis o he p opo ion o he anspo a ion in as uc u e in es men in he a ge ci y
ela i e o he o e all in es men in anspo a ion in as uc u e. The s eel demand o consume goods
p oduc ion was es ima ed by calcula ing he s eel consump ion o consume goods a he p o incial o
na ional le el, which was hen alloca ed on he basis o he p opo ion o he household consump ion
expendi u e o co esponding consume goods ca ego ies in he a ge ci y ela i e o he o al household
consump ion expendi u e in hose ca ego ies. The calcula ion was as ollows:
()
QASC
A
A
(25)
demand s eel building jp o ince na ion le elj uni a ea jci y le el
p o ince na ion le el
,(), ,()
∑
=××
−−
−
−
QQ
I
I
(26)
demand s eel ansdemands eel ansp o incena ionci y le el ans
p o ince na ion le el ans
,,,(),
(),
=×
−−
−
−
∑
=
×
−−
−∈
−∈
QQ
I
I
(27)
demand s eel goodsgdemand s eel goodsp o incena iong ci y le elg G
p o ince na ion le elg G
,,,(), ,
(),
whe e Ap o ince(na ion)-le el,j deno es he annual new cons uc ion a ea o building ype j a he p o incial o
na ional le el; Aci y-le el deno es he annual new cons uc ion a ea in he ci y; SCuni a ea,j deno es he s eel
consump ion pe uni a ea o new cons uc ion o each building ype; and Ap o ince(na ion)-le el deno es he
annual new cons uc ion a ea a he p o incial o na ional le el. Mo eo e , Ici y-le el, ans deno es he annual
anspo a ion in as uc u e in es men in he ci y; Qdemand-s eel, ans,p o ince(na ion) deno es he amoun o
s eel consumed in anspo a ion in as uc u e cons uc ion a he na ional o p o incial le el; Ip o ince(na-
ion)-le el, ans deno es he annual anspo a ion in as uc u e in es men a he p o incial o na ional le el;
Qdemand-s eel,goods,p o ince(na ion) deno es he appa en c ude s eel consump ion in he indus ial consume
goods manu ac u ing sec o a he na ional o p o incial le el; Ici y-le el,g∈G deno es he annual household
consump ion expendi u e o consume goods; and Ip o ince(na ion)-le el,g∈G deno es he annual household con-
sump ion expendi u e o consume goods. I he abo e p o incial-le el s a is ical da a we e una ailable o
could no be ully ob ained, he na ional annual s eel consump ion was used and alloca ed o he ci y on he
basis o he u ban esiden popula ion. The calcula ion was as ollows:
QSC
P
P
(28)
demand s eel na ion ci y le el
na ion le el
=×
−
−
−
whe e SCna ion deno es he o al na ional s eel demand; Pci y-le el deno es he u ban esiden popula ion o he
a ge ci y; and Pna ion-le el deno es he na ional esiden popula ion.
(2) Cemen
The na ional annual cemen consump ion was ob ained and alloca ed o a gi en ci y on he basis o he
u ban esiden popula ion as ollows:
QSC
P
P
(29)
demand cemen na ion ci y le el
na ion le el
=×
−
−
−
whe e SCna ion deno es he o al na ional s eel demand; Pci y deno es he u ban esiden popula ion o he
a ge ci y; and Pna ion deno es he na ional esiden popula ion.
emission ac o s. All emission ac o s equi ed o he calcula ions a e p o ided in Supplemen a y TableA2.
9
Scien i ic Da a | (2025) 12:1672 | h ps://doi.o g/10.1038/s41597-025-05949-y
www.na u e.com/scien i icda a
www.na u e.com/scien i icda a/
Da a alida ion and unce ain y analysis. C oss-checking was widely ecognized as a c i ical me hod
o ensu ing da a in eg i y and consis ency ac oss mul iple esea ch eams41. This app oach, oo ed in he p in-
ciples o igo ous scien i ic alida ion, allowed o he iden i ica ion and co ec ion o biases ha migh ha e
esul ed om indi idual analy ical p ocesses o eam-speci ic me hodologies. By engaging mul iple eams in
independen ye s anda dized analyses o di e en da a subse s, c oss-g oup alida ion acili a ed he e i ica ion
o esul s and ensu ed he obus ness and ep oducibili y o he indings ac oss a ying expe imen al condi ions.
Fu he mo e, i enhanced anspa ency and p omo ed collabo a ion, which was essen ial when di e se da ase s
a e in eg a ed o when la ge-scale s udies a e conduc ed ha equi ed in e disciplina y inpu . In he con ex o
scien i ic esea ch and he quali y assu ance o la ge‐scale da ase s, he c i ical impo ance o c oss-checking
was a i med by mul iple in e na ional s anda ds and bes p ac ices. Fo example, he GHG P o ocol’s “Tools &
Guidance” module42 ad ises ha when deploying i s sui e o c oss-sec o al, depa men al, and ci y-le el me h-
odologies, in en o y esul s p oduced by di e en ools o eams should ha e been subjec ed o ho izon al com-
pa ison and c oss- alida ion o unco e anomalies and con adic ions, he eby ensu ing ha he inal epo
was comp ehensi e, anspa en , and consis en . Likewise, he Be keley Ini ia i e o T anspa ency in he Social
Sciences (BITSS)43 cau ions ha la ge eam sizes can ampli y biases and ecommended ha sub- eams inde-
penden ly ca ied ou —and hen mu ually e iewed—each s age o da a collec ion, cleaning, analysis, and a chi -
ing. This sub- eam c oss- e iew p ocess gua an eed ha dispa a e esea che s ollowing he same p ocedu es
achie ed conco dan ou comes, he eby sa egua ding he o e all cohe ence and anspa ency o he s udy44. The
adop ion o c oss-g oup alida ion hus se es no only o enhance he scien i ic igo o he analysis bu also
o p omo e consensus among esea che s, hus ensu ing ha he conclusions we e g ounded in consis en and
ep oducible da a.
The p ocedu e o his me hod is was ollows: (1) Da a di ision ( he da ase was spli in o mul iple subse s on
he basis o speci ic c i e ia, wi h each g oup analysing hei espec i e subse independen ly), (2) independen
analysis (each eam applied s anda dized analy ical me hods and models o hei designa ed da a subse ), (3)
c oss- alida ion ( he eams alida ed each o he ’s esul s o ensu e consis ency and accu acy ac oss he di e en
g oups), (4) compa ison and adjus men (di e ences in he esul s we e iden i ied, and necessa y adjus men s
o me hods o da a p ocessing we e made), and (5) inal consensus (a e econciling inconsis encies, a uni ied
epo was gene a ed, in which he indings om all g oups we e consolida ed).
In his s udy, 45 esea che s om 31 ins i u ions in he China Ci y G eenhouse45 (CCG) wo king g oup
pa icipa ed in he basic da a collec ion and ca bon emission calcula ion p ocesses. P io o ini ia ing da a
collec ion and c oss-g oup alida ion, we con ened a one-week in ensi e wo kshop on he ca bon emission
accoun ing me hodology guidelines o his s udy o all 45 esea che s. This aining ensu ed a uni o m unde -
s anding o he p ocedu es, unde lying o mulas, pa ame e selec ions, and epo ing o ma s. Following he
wo kshop, each esea che independen ly calcula ed he scope-speci ic ca bon emissions using he same guid-
ance manual, and subsequen c oss-checking u he con i med he me hodological obus ness and he absence
o any sys ema ic compu a ional e o s.
Assessing unce ain y was indispensable o obus g eenhouse-gas in en o y de elopmen , as i bo h high-
ligh ed he eliabili y o he cu en da ase and guided subsequen me hodological enhancemen s. Incomple e
o inconsis en ac i i y eco ds and he in insic a iabili y o emission ac o s we e he p incipal con ibu o s
o o e all unce ain y in emissions accoun ing. The IPCC46 ecommends wo p incipal app oaches o quan-
i ying hese unce ain ies: classical e o ‐p opaga ion echniques and Mon e Ca lo simula ion. While e o
p opaga ion o e ed a s aigh o wa d es ima e by combining a iances analy ically, Mon e Ca lo me hods could
accommoda e complex, non‐Gaussian dis ibu ions and co ela ions among inpu pa ame e s, yielding a mo e
comp ehensi e unce ain y p o ile.
P e ious wo k had applied Mon e Ca lo simula ion ex ensi ely o cha ac e ize unce ain ies in g eenhouse‐
gas and o he pollu an in en o ies a he na ional scale47–49. D awing on hose applica ions, along wi h he
IPCC’s unce ain y‐analysis guidance, na ional GHG in en o y epo s50, plus pee ‐ e iewed e alua ions o
u ban ca bon da a quali y and expe judgmen , we de i ed unce ain y bounds o ou ci y‐le el ac i i y da a-
se s. Mo eo e , a numbe o s udies had al eady implemen ed simila s ochas ic app oaches o quan i y he
unce ain ies associa ed wi h Scope 1 and Scope 2 CO2 emissions in indi idual ci ies. Fo he analysis p esen ed
in his pape , we employed he Mon e Ca lo simula ion me hod ecommended by he IPCC o es ima e he
unce ain y in he GHG accoun ing esul s o Chinese ci ies. This p ocess in ol ed h ee s eps: (1) de ined he
p obabili y dis ibu ion as a no mal dis ibu ion o he es ima ion pa ame e s such as ac i i y da a and emis-
sion ac o s ac oss he di e en sec o s and ci ies, along wi h co esponding unce ain y analysis s a is ics; (2)
calcula ed he co esponding emissions o each ca ego y acco ding o he me hod o calcula ing he ca bon
emissions esul ing om p oducing he p oduc s in his s udy; and (3) epea ed he simula ion 10,000 imes.
P obabili y dis ibu ions o he di e en ca ego ies o he o al emissions we e subsequen ly ob ained, as we e
co esponding unce ain y analysis s a is ics.
Fo de ailed me hodological app oaches and unce ain y quan i ica ion pe aining o Scope 1 and 2 ca bon
emission da a, eade s we e e e ed o ou p io s udy21. In quan i ying he unce ain y associa ed wi h Scope 3
ca bon emissions, we adhe ed o he IPCC’s46 ie ed guidance by assigning dis inc unce ain y bounds acco d-
ing o he p o enance o ac i i y da a. Speci ically, Me hod 1 pa alleled a Tie 3 app oach, elied on high‐ ideli y,
ield‐su eyed da a collec ed di ec ly by municipal au ho i ies; he IPCC ecommended an unce ain y band
o 2–5%51, and we ha e adop ed a alue o 4%. Me hod 2 co esponded o Tie 2, d ew upon igu es published
in municipal s a is ical yea books and o he publicly a ailable da ase s; consis en wi h he IPCC’s sugges ed
5–10% ange51, we applied an unce ain y o 8%. Me hod 3 was equi alen o Tie 1, whe ein ci y‐le el es i-
ma es we e de i ed by p opo ional down‐scaling o b oade agg ega es; gi en he IPCC’s 30–100% guidance
o Tie 151, we imposed a 40% unce ain y when alloca ing p o incial da a o ci ies and 50% when alloca ing
16
Scien i ic Da a | (2025) 12:1672 | h ps://doi.o g/10.1038/s41597-025-05949-y
www.na u e.com/scien i icda a
www.na u e.com/scien i icda a/
46. In e go e nmen al Panel on Clima e Change (IPCC). IPCC Guidelines o na ional g eenhouse gas in en o ies. Ins i u e o Global
En i onmen al S a egies (IGES) (2006).
47. Sha n , Y. e al. Me hodology and applica ions o ci y le el CO2 emission accoun s in China. Jou nal o Cleane P oduc ion 161,
1215–1225, h ps://doi.o g/10.1016/j.jclep o.2017.06.075 (2017).
48. Jama u u, S. A. e al. Quan i ying u u e ca bon emissions unce ain ies unde s ochas ic modeling and Mon e Ca lo simula ion:
Insigh s o en i onmen al policy conside a ion o he Bel and Road Ini ia i e Region. Jou nal o En i onmen al Managemen 370,
122463, h ps://doi.o g/10.1016/j.jen man.2024.122463 (2024).
49. Eo y, V., Topp, C. F., Bu le , A. & Mo an, D. Add essing unce ain y in e icien mi iga ion o ag icul u al g eenhouse gas emissions.
Jou nal o Ag icul u al Economics 69(3), 627–645, h ps://doi.o g/10.1111/1477-9552.12269 (2018).
50. The Fou h Na ional Communica ion on Clima e Change o he People’s Republic o China.
51. Pulles, T., Meije , J. & an Aa denne, J. Es ima ing Unce ain ies in GHG Emissions om Fuel Combus ion. In e go e nmen al Panel
on Clima e Change (IPCC) (2006).
52. Meng, F. e al. China Ci y-Le el Full-scope Ca bon Emissions Da ase 2023. Figsha e. h ps://doi.o g/10.6084/m9. igsha e.28645499
(2025).
53. Xu, J., Guan, Y., Old ield, J., Guan, D. & Shan, Y. China ca bon emission accoun s 2020-2021. Applied Ene gy 360, 122837, h ps://
doi.o g/10.1016/j.apene gy.2024.122837 (2024).
54. Wang, H., Zhang, R., Liu, M. & Bi, J. The ca bon emissions o Chinese ci ies. A mosphe ic Chemis y and Physics 12(14), 6197–6206,
h ps://doi.o g/10.5194/acp-12-6197-2012 (2012).
55. Ene gy Ins i u e(EI). S a is ical Re iew o Wo ld Ene gy 2025 (74 h ed.). In collabo a ion wi h Kea ney and KPMG. h ps://www.
ene gyins .o g/s a is ical- e iew (2025).
56. In e na ional Ene gy Agency (IEA). CO2 Emissions in 2023. IEA, Pa is. h ps://www.iea.o g/ epo s/co2-emissions-in-2023 (2024).
57. Mickey, F. How has U.S. ene gy use changed since 1776? U.S. Ene gy In o ma ion Adminis a ion. h ps://www.eia.go (2025).
58. C ippa, M., Guizza di, D., Pagani, F., Pisoni, E. GHG Emissions a sub-na ional le el. Eu opean Commission, Join Resea ch Cen e.
h ps://da a.j c.ec.eu opa.eu/da ase /d67eeda8-c03e-4421-95d0-0adc460b9658 (2023).
59. Li, W. e al. Assessmen o g eenhouse gasses and ai pollu an emissions embodied in c oss-p o ince elec ici y ade in China.
Resou ces, Conse a ion and Recycling 171, 105623, h ps://doi.o g/10.1016/j. escon ec.2021.105623 (2021).
60. Tong, K. e al. G eenhouse gas emissions om key in as uc u e sec o s in la ge and smalle Chinese ci ies: me hod de elopmen
and benchma king. Ca bon Managemen 7(1–2), 27–39, h ps://doi.o g/10.1080/17583004.2016.1165354 (2016).
61. Beijing Municipal Bu eau o S a is ics, Na ional Bu eau o S a is ics Beijing Su ey Team. Beijing S a is ical Yea book 2024. Beijing:
China S a is ics P ess (2025).
62. Na ional Bu eau o S a is ics o he People’s Republic o China. S a is ical Bulle in on Na ional Economic and Social De elopmen in
2023 Na ional Economic and Social De elopmen . Beijing: China S a is ics P ess. (2024).
63. Beijing Municipal Wa e A ai s Bu eau. Beijing Wa e Resou ces Bulle in 2023. Beijing: Beijing Municipal Wa e A ai s Bu eau
(2024).
64. Tianjin Municipal Bu eau o S a is ics, Na ional Bu eau o S a is ics Tianjin Su ey Team. Tianjin S a is ical Yea book 2024. Beijing:
China S a is ics P ess (2025).
65. Tianjin Municipal Wa e A ai s Bu eau. Tianjin Wa e Resou ces Bulle in 2023. Tianjin: Tianjin Municipal Wa e A ai s Bu eau.
(2024).
66. Hebei Municipal Bu eau o S a is ics, Na ional Bu eau o S a is ics Hebei Su ey Team. Hebei S a is ical Yea book 2024. Beijing:
China S a is ics P ess (2025).
67. Hebei Municipal Wa e A ai s Bu eau. Hebei Wa e Resou ces Bulle in 2023. Hebei: Hebei Municipal Wa e A ai s Bu eau (2024).
acknowledgemen s
This esea ch was unded by Na ional Na u al Science Founda ion o China (72522012, 72174028, 72504111)and
he Ho izon Eu ope P ojec EU-CHINA-BRIDGE (101137971), which a e suppo ed by UKRI g an (10132630)
a he Uni e si y o Bi mingham. This s udy was pe o med by olun ee s om he China Ci y G eenhouse
Wo king G oup. The au ho s would like o hank he ollowing people o hei con ibu ions o he ull-
scope CO2 emissions da abase and his a icle: This s udy u ilized a collabo a i e wo king g oup model o
da a collec ion. The au ho s g a e ully acknowledge he 45 membe s o he China Ci y G eenhouse Wo king
G oup o hei signi ican con ibu ions o he collec ion, ex ac ion, and ini ial compila ion o he unde lying
ac i i y da a used in he ull-scope CO2 emissions da abase. These indi iduals se ed as da a con ibu o s and
a e no co-au ho s o his manusc ip . Scope 1 and Scope 2 ac i i y da a we e p ima ily sou ced om publicly
accessible na ional da abases. All Scope 3 ac i i y da a we e me iculously ga he ed om o icially published ci y-
le el s a is ical yea books and s a is ical bulle ins; de ailed sou ces o each ci y a e p o ided in Supplemen a y
TableA1. The au ho s would like o hank he ollowing people o hei da a collec ion e o s: Chen L , Chinese
Academy o En i onmen al Planning. Jing Guo, Chinese Academy o En i onmen al Planning. Lingyun Pang,
Chinese Academy o En i onmen al Planning. Wanyue Shan, Beijing Uni e si y o Chemical Technology. Xu
Zhang, School o En i onmen , Beijing No mal Uni e si y. Jiaqi Hou, School o En i onmen , Beijing No mal
Uni e si y. Jin ao Sun, School o New Ene gy and En i onmen , Jilin Uni e si y. Yimiao Wang, Beijing No mal
Uni e si y. Qiwen Ma, Beijing No mal Uni e si y. Fumin Gu, Suzhou Xie en En i onmen al P o ec ion
Technology Se ice Co., L d. Yanqing Wang, Dalian Ins i u e o Chemical Physics, Chinese Academy o Sciences.
Yiwen Sun, Public Tes ing E alua ion and Iden i ica ion Technology Cen e , Jiangsu Academy o Ag icul u al
Sciences. Shijie Li, Guangdong Uni e si y o Technology. Jianhui Cong, School o Economics and Managemen ,
Shanxi Uni e si y. Dandan Zhang, Depa men o he Buil En i onmen , Aal o Uni e si y, Finland. Buy ayem,
Xinjiang Ag icul u al Uni e si y. Shengnan Zhao, Chi eng Uni e si y. Yongliang Yang, Zhejiang Sci-Tech
Uni e si y. Zhiyuan Ning, Shanghai Hebang Ce i ica ion Co., L d. (NSF In e na ional). Biao Liu, School o
Land Science and Technology, China Uni e si y o Geosciences (Beijing). Zhihong Li, Hong Kong Huayi Design
Consul ing (Shenzhen) Co., L d. Yiwei Xiong, Foshan Ins i u e o Quali y and S anda diza ion. Mengli Wu,
Tianjin Youmei En i onmen al P o ec ion Technology Co., L d. Zhouye Zhao, Tongji Uni e si y. Nan Li, He ei
Uni e si y o Technology. Yue Hu, Dalian Ins i u e o Chemical Physics, Chinese Academy o Sciences. Yao Tang,
Wuhan Uni e si y. Lu Yang, Wuhan Tex ile Uni e si y. Mengbing Du, Wuhan Uni e si y. YunLong Wu, Capi al
Uni e si y o Economics and Business. YuShen Liu, Hongjie Ji, Shanxi Uni e si y. Wei He, Wuhan Uni e si y. Jian
Huang, Wuhan Uni e si y. Peng Li, Hangzhou Bingxin En i onmen al P o ec ion Packaging Co., L d. Mingji Lao,
Fangyuan Ma k Ce i ica ion G oup Co., L d.
17
Scien i ic Da a | (2025) 12:1672 | h ps://doi.o g/10.1038/s41597-025-05949-y
www.na u e.com/scien i icda a
www.na u e.com/scien i icda a/
au ho con ibu ions
Fanxin Meng and Bo eng Cai led he p ojec and designed he esea ch; Hanbo Hu, Yu ong Sun, Jiaqi Hou, Zhe
Zhang and Lingyun Pang collec ed he aw da a, assembled he da a and pa icipa ed in he da ase cons uc ion;
Fanxin Meng, Hanbo Hu and Li Zhang w o e he manusc ip ; Yu ong Sun, Li Zhang, Zhe Zhang and Yuli Shan
e ised he manusc ip .
Compe ing in e es s
The au ho s decla e no compe ing in e es s.
addi ional in o ma ion
Supplemen a y in o ma ion The online e sion con ains supplemen a y ma e ial a ailable a h ps://doi.o g/
10.1038/s41597-025-05949-y.
Co espondence and eques s o ma e ials should be add essed o F.M., L.Z., Z.Z. o Y.S.
Rep in s and pe missions in o ma ion is a ailable a www.na u e.com/ ep in s.
Publishe ’s no e Sp inge Na u e emains neu al wi h ega d o ju isdic ional claims in published maps and
ins i u ional a ilia ions.
Open Access This a icle is licensed unde a C ea i e Commons A ibu ion-NonComme cial-
NoDe i a i es 4.0 In e na ional License, which pe mi s any non-comme cial use, sha ing, dis ibu-
ion and ep oduc ion in any medium o o ma , as long as you gi e app op ia e c edi o he o iginal au ho (s)
and he sou ce, p o ide a link o he C ea i e Commons licence, and indica e i you modi ied he licensed ma e-
ial. You do no ha e pe mission unde his licence o sha e adap ed ma e ial de i ed om his a icle o pa s
o i . The images o o he hi d pa y ma e ial in his a icle a e included in he a icle’s C ea i e Commons
licence, unless indica ed o he wise in a c edi line o he ma e ial. I ma e ial is no included in he a icle’s C e-
a i e Commons licence and you in ended use is no pe mi ed by s a u o y egula ion o exceeds he pe mi ed
use, you will need o ob ain pe mission di ec ly om he copy igh holde . To iew a copy o his licence, isi
h p://c ea i ecommons.o g/licenses/by-nc-nd/4.0/.
© The Au ho (s) 2025
