HISTORECO: Historical Spanish transition database on climate, geography and economics of the 20th-21st century

1
Scien i ic Da a | (2025) 12:810 | h ps://doi.o g/10.1038/s41597-025-05055-z
www.na u e.com/scien i icda a
HIS ORECO: His o ical Spanish
ansi ion da abase on clima e,
geog aphy and economics o he
20 h-21s cen u y
Guille mo Rod íguez-López1, ana Se ano1, Miguel Ma ín-Re o illo2 & Ignacio Cazca o 3,4 ✉
One o he majo di icul ies in he social sciences is ob aining compa able da a om di e en sou ces.
Especially, o small-scale da a, such as municipali ies wi hin a coun y, and e en mo e when he
a iables belong o di e en ields o knowledge. To a oid hese p oblems, we p esen a da abase
con aining 45 geog aphic, clima ic, hyd ological and demo-economic a iables (64 columns) ha
co e s he 20 h and 21s cen u ies o all municipali ies in Spain. To achie e his, we me ged se en een
da abases/sou ces, using se e al me hods, so wa e and p og amming languages (QGIS, R, Py hon)
o homogenize and downscale he a iables o Spanish municipali ies. Technical alida ion esul s
a e included wi h agg ega ion and al e na i e sou ces checks. This da abase is a aluable esou ce o
esea che s om di e en ields o esea ch, as he e is no o he esou ce wi h his empo al b ead h
and spa ial disagg ega ion in he cu en li e a u e. The da ase can con ibu e o comp ehensi e
analyses wi h empo al and spa ial compa isons a ound se e al cu en deba es in he li e a u e and
policy (local e ec s o global and clima e change, adap a ion, demog aphy, land use change, e c.).
Backg ound & Summa y
One o he main p oblems in he social sciences is ob aining compa able da a om di e en sou ces in o de
o conduc compa a i e analysis, such as mul ime hodology1. This p oblem is e en g ea e when we wan o
compa e o e he long e m. Cu en socio-economic and en i onmen al issues, such as depopula ion in some
a eas, he e ec s o clima e change as, o example, dese i ica ion, o di e ences in he de elopmen be ween
socie ies equi e da abases ha a e compa able in ime and space. These p oblems a e deeply oo ed in economic
sys ems, he en i onmen , and socie ies, so i is impe a i e o analyse he causes aking in o accoun long- e m
pe spec i es bu also mul idisciplina y a iables. Mo eo e , i is e en mo e di icul o ob ain compa able da a
om di e en sou ces when hey belong o di e en ields o knowledge. Resea che s need hese da a o ca y
ou his mul idisciplina y analysis.
All his is e en mo e di icul when we y o analyse municipali ies. This e i o ial le el o s a is ics, which
is mo e and mo e common nowadays, is adi ionally no o e ed by he S a is ics O ices, o a leas h ough
su ey me hods. Fo example, in he Eu opean Union, he mos common le el o e i o ial s a is ics in he main
a iables is o NUTS 2 (Au onomous Communi ies in he case o Spain) o NUTS 3 (P o inces in he case o
Spain). No e en he Eu opean Union has used a mo e disagg ega ed le el beyond he NUTS 3. In his same case,
Eu os a o e s a su ey wi h he main a iables o ci ies wi h da a om 2013, omi ing in o ma ion on smalle
ones (less han 50,000 inhabi an s)2.
The cons uc ion o municipal da ase s has a g owing p esence in he li e a u e co e ing se e al hemes. One
o he mos impo an e o s o p o ide geospa ial da a in mul iple opics is he wo k o UNECA3. O he exam-
ples o he cons uc ion o municipal da a ela ed o socio-economic deba es a e he cases o was e collec ion in
1Depa men o Economic Analysis, Facul y o Economics and Business S udies, Uni e si y o Za agoza, Ag i ood
Ins i u e o A agon (IA2), 50005, Za agoza, Spain. 2Depa men o economics, acul y o economics, Business and
Tou ism, Uni e sidad de Alcalá, Plaza de la Vic o ia 2, 28802, Alcalá de Hena es, Spain. 3ARAiD (A agonese Agency
o Resea ch and De elopmen ), Ag i ood Ins i u e o A agon (IA2), Depa men o Economic Analysis, Facul y
o Economics and Business S udies, Uni e si y o Za agoza, 50005, Za agoza, Spain. 4Basque cen e o clima e
Change, Pa que Cien í ico de UPV/EHU, 48940, Leioa, Spain. ✉e-mail: icazca @uniza .es
DaTa DESCRIPTOR
OPEN
2
Scien i ic Da a | (2025) 12:810 | h ps://doi.o g/10.1038/s41597-025-05055-z
www.na u e.com/scien i icda a
www.na u e.com/scien i icda a/
Po ugal, he implica ions o belonging o Special Economic Zones in China o he d inking wa e da abase in
US municipali ies4–6. Despi e hei impo ance, all o hese examples do no co e a long pe iod o ime. Ano he
example is he No wegian da ase o analysing he local go e nmen , which o e s da a o 50 yea s (1972–2022)
wi h demog aphic, poli ical o economic measu es7.
Taking all hese in o accoun , he case o Spain is s iking. Spanish s a is ics make i possible, no wi h-
ou di icul y, o join se e al da abases and ob ain compa able a iables in a long- e m pe spec i e a a highly
disagg ega ed le el. Spain is included in some da abases analysing he global clima e and geog aphical condi-
ions. Fu he mo e, Spanish s a is ics make i possible, o a leas in e , some a iables o he analysis o he
socio-economic cha ac e is ics wi h a pe spec i e o se e al decades.
Ou da ase can include mul idisciplina y a iables o he socio-economic analyses, using da a om wen y
di e en da abase sou ces a a ha monised le el o disagg ega ion (INE8, MITECO9–12, GDW13, SEPREM14,
HYDE15,HID16, ESYRCE17, GIA18, CRU TS e sion 4.0519 IPE-CISC20,21, IGN22, Goe lich, 201923, Albe us
202324, Bel án Tapia e al.25, Es eban-Oli e and Ma í-Hennebe g, 202326,27) as well as o he ex ensi e da a
sou ces (i.e. e en along his o ical census da a, yea books, e c. on pape o /and digi alized as pd s) and li e a-
u e. The Supplemen a y In o ma ion ile p o ides a lis o da a sou ces, wi h he spa ial agg ega ion and em-
po al co e age indica ed (TableS1), and a summa y o he main limi a ions and unce ain ies o he o iginal
sou ces om which he key a iables o he s udy we e calcula ed, oge he wi h he a iables a ec ed by hem
(TableS2). Ob iously o some o he a iables, such as clima e da a, he e a e o he la ge in e na ional da abases
ha p o ide b oad and equen ele an da a (e.g. Menne e al.28; Sil insky e al.29). One o he mos impo an
added alue ea u es o he da abase is ha he e is no o he da ase ha o e s such b ead h in ime and space
wi h di e en socio-economic and geog aphical a iables. Some p ojec s in he Spanish economic his o y ha e
measu ed, in a long iew, some a iables ha a e included in ou da ase , such as he SPAREL30, essays o ine-
quali y among Spanish egions31,32 o he publica ions o Goe lich33,34 hey ocus on he popula ion o illages,
owns o municipali ies, o on egional inequali y. O he analyses wi h mo e socio-economic a iables a e he
wo ks o Ca e as and Ta unell35 bu hey do no p o ide a disagg ega ion a he municipal le el. The e a e also
o he academic wo ks ha y o p o ide o analyse da a o Spain. A good example is he wo k by Goe lich36,
which is ele an o he ea men o clima ic da a in ela ion o he Spanish p o inces. Ano he in e es ing a i-
cle and da a by his au ho , which we use ex ensi ely, is he Municipal Da abase o he 2011 Census23. Cazca o
e al.37 analysed he si ua ion o he main hyd aulic in as uc u es and wa e esou ces p o iding a con ex o
u al de elopmen ela ed o ag icul u e, and pa icula ly o highly wa e dependen i iga ed ag icul u e. Fo
he p esen decades, ob iously one inds a my iad o na ional and in e na ional da abases on hese opics o
blocks o a iables, a he municipal le el o a a esolu ion ha would allow an adequa e app oxima ion o his
le el. Recen examples include he In eg a ed Municipal Da a Sys em (SIDAMUN) o he Spanish Minis y o
Ecological T ansi ion and he Demog aphic Challenge38, o he new demog aphic da abase (DEMOSPA0521)
wi h almos 900 million demog aphic eco ds om Spain o he las wo decades (Lledó and Pa ía)39. Bu he
limi a ions o his o ical da a emain o many up- o-da e da abases, abula o GIS da a, pa icula ly o demo-
g aphic and o he socio-economic a iables.
He e, we p esen a da abase o Spanish municipali ies (depa ing om a 8,205 objec s/polygons shape-
ile on municipali ies o he Spanish Na ional Geog aphical Ins i u e40, hey we e homogenised wi h he 8,116
Goe lich’s23 homogeneous municipal popula ion se ies o each ou 8122 homogeneous municipali ies o he
2016–2017 en i ies) o decennial da a o 45 di e en a iables (apa om he 7 iden i ica ion columns, 64
independen columns wi h a iables, hence some o which a e p esen ed oge he in he same box below, lis -
ing he names in he da ase o se e al columns, when hey e e o he same a iable concep ). The ime span
includes da a om 1900 o he p esen wi h he ollowing c oss sec ions: 1900, 1910, 1920, 1930, 1940, 1950,
1960, 1970, 1981, 1991, 2001, 2011 and 2021. The ollowing sec ions desc ibe he a iables and how we ha e
ha monised hem in ou da abase.
Table1 shows he clima ic a iables o ou da ase . The o al p ecipi a ion a iable is he a e age o each dec-
ade o o al annual ain all in millime es o he decadal equency. Fo he annual equency, i is he mon hly
o al p ecipi a ion in he e e ence yea . The mean empe a u e is he a e age o each decade o he annual
a e age empe a u e in deg ees Celsius o he decadal a iable, and he mon hly a e age empe a u e in he
e e ence yea in case o he annual equency. SPEI “is based on a mon hly clima ic wa e balance (p ecipi a ion
minus PET), and i is exp essed as a s anda dised Gaussian a ia e wi h a mean o ze o and a s anda d de ia ion
o one”41. In his way, he SPEI da a is he a e age o each decade o he annual mean o he SPEI d ough index
in millime es. In case o he yea ly equency, he alue co esponds o he mon hly a e age in he e e ence
yea . The ollowing a iable is he p ecipi a ion in he ege a ion g ow h pe iod, namely, he a e age o each
decade (o yea depending he equency) o accumula ed ain all in he g ow h ege a ion pe iod ( om Ap il
o Oc obe ). The os days a iable includes he a e age o each decade o he numbe o days wi h a minimum
empe a u e below 0°C. The yea ly equency ep esen s he annual numbe o days wi h a minimum empe a-
u e below 0°C. The dummies o Köppen’s clima e classi ica ion ake he alue 1 i he municipali y has a speci ic
clima e acco ding o he Köppen’s clima e classi ica ion, and 0 o he wise. We ake in o accoun i he munici-
pali y belongs o he d y and ho clima e, oceanic clima e, Medi e anean clima e, Medi e anean clima e wi h
cool summe s, bo eal clima e, humid and ho clima e, con inen al clima e wi h cool summe s and con inen al
bo eal clima e.
Table2 p esen s he main in o ma ion on he geog aphical a iables. The dis ance a iables measu e he
dis ance in kilome es om he cen oid o he municipali y o he nea es coas , o he cen oid o Mad id and
o he cen oid o he p o incial capi al in a s aigh line. The dis ance o he nea es municipali y wi h mo e
han 10,000 and 5,000 inhabi an s measu e he dis ance in kilome es o he cen oid o bo h municipali ies. The
la i ude and longi ude coo dina es co espond o he cen oid o he municipali y in decimal deg ees (DD). The
3
Scien i ic Da a | (2025) 12:810 | h ps://doi.o g/10.1038/s41597-025-05055-z
www.na u e.com/scien i icda a
www.na u e.com/scien i icda a/
al i ude a iable is he a e age al i ude o he municipali y in me es abo e sea le el. The uggedness is he s and-
a d de ia ion o he al i ude o he municipali y. The a ea is he su ace o he municipali y in squa e kilome es.
All he land use a iables included in Table3 a e measu ed in hec a es. The i iga ed su ace o neighbou ing
municipali ies includes all he i iga ed su ace o neighbou ing municipali ies, i.e., he municipali ies sha ing a
common bounda y wi h he e e ence municipali y.
Table4 p esen s he main in o ma ion on he hyd ological a iables. The i e basin indica es he i e basin
o which each municipali y belongs. The ese oi wa e olume capaci y o he municipali y is measu ed in
cubic hec ome es. The usable olume capaci y o dammed wa e in cubic hec ome es. The nea es main i e
p o ides he name o he nea es i e basin bigge han 500 squa e kilome es. The dis ance o his main i e
is in kilome es. The nea es wa e cou se shows he name o he nea es wa e cou se, and he dis ance o his
wa e cou se is in kilome es, ega dless o i s size.
Finally, Table5 lis s he socioeconomic-demog aphic a iables. The popula ion o he municipali ies is he
numbe o inhabi an s. The a iables o he popula ion esiding in municipali ies wi h mo e han
x
inhabi an s
(in housands) loca ed wi hin a adius be ween
i
and
j
km a e measu ed in numbe o inhabi an s. These a ia-
bles include all he popula ion who li ed in municipali ies wi h mo e han
x
inhabi an s and hey a e wi hin
speci ic dis ance adio, de ined by
i
and
j
kms. The denomina ion o he a iables ollows he pa e n:
Pxij,
, being
p
he popula ion (in housands),
x
he h eshold o he popula ion (in housands) esiding in municipali ies wi h
mo e han his popula ion han he h eshold
x
loca ed in he adius be ween he
i
and
j
kilome es. Fo example,
p50_25_50km shows he popula ion o municipali ies wi h mo e han 50,000 inhabi an s which a e si ua ed in
a adius o he e e enced municipali y be ween 25 and 50 kilome es (inspi ed also in Bel án Tapia e al.25). The
ollowing a iables a e he dis ance in kilome es o he nea es municipali y wi h mo e han 5,000 and 10,000
inhabi an s. The Simpson42 egional classi ica ion a iables indica e he Spanish ag a ian egion o which each
municipali y belongs, depending on whe he i is di ided in o 5 o 11 a eas. This classi ica ion can be use ul o
he ea men o di e en a iables. The Popula ion_class ini ia es a classi ica ion, in line wi h INE, o ype o he
municipali y only wi h espec o he popula ion size (u ban, in e media e, u al). I is based on he classi ica ion
o he Spanish S a is ics Ins i u e in he 1950 Popula ion Census43. This classi ica ion indica es ha a municipal-
i y is u al, i i has less han 2,000 inhabi an s. An u ban municipali y has mo e han 10,000 inhabi an s. The
Va iables Name in he da ase Da a sou ce Timespan F equency Uni s
To al p ecipi a ion pp CRU TS. Ve sion 4.05 1900–2020
(1950–2021) Decadal (& Yea ly) Decadal mean o o al annual
p ecipi a ion (mm)
Mean empe a u e _a e age CRU TS. Ve sion 4.05 1900–2020
(1950–2021) Decadal (& Yea ly) Decadal mean o mean annual
empe a u e (C°)
SPEI spei IPE – CSIC 1900–2020
(1950–2021) Decadal (& Yea ly) Decadal mean annual mean SPEI
d ough index (mm)
P ecipi a ion in
ege a ion g ow h
pe iod g ow_pe iod_pp CRU TS. Ve sion 4.05 1900–2020
(1950–2021) Decadal (& Yea ly) Decadal mean o he accumula ed
p ecipi a ion in he mon hs om Ap il
o Oc obe , bo h included (mm)
F os days os _days CRU TS. Ve sion 4.05 1900–2020
(1950–2021) Decadal (& Yea ly) Decadal mean o numbe o
days pe yea wi h a minimum
empe a u e < 0 °C
Dummies o
Köppen’s clima e
classi ica ion
d y_ho _clima e
d y_cold_clima e
oceanic
medi e anean
medi e anean_ esh_summe
humid_ho
con inen al_ esh_summe
con inen al_bo eal
IGN 1900 - 2020 Decadal 0 – 1: Dummy a iable ha akes
he alue 1 i he municipali y has a
speci ic clima e ype, and 0 o he wise
Table 1. Lis , sou ces, imespan and uni s o clima ic a iables.
Va iables Name in he da ase Da a sou ce Timespan Uni s
Dis ance o coas o_coas _km IGN S a ic a iable Kilome es
Dis ance o Mad id o_mad_km IGN S a ic a iable Kilome es
Dis ance o p o ince capi al ci y o_p o _cap_km IGN S a ic a iable Kilome es
Coo dina es (X,Y) longi ude
la i ude IGN S a ic a iable La i ude and longi ude coo dina es o he
cen oid o he municipali y (DD)
Al i ude al i ude IGN S a ic a iable A e age al i ude in me es abo e he sea
Ruggedness uggedness IGN S a ic a iable S anda d de ia ion o he al i ude o
municipali y
A ea a ea IGN S a ic a iable A ea o he municipali y (km2)
Table 2. Lis , sou ces, imespan and uni s o geog aphical a iables. No e: Values o se e al s a is ics we e
also consis en ly checked wi h da a om F ancisco Bel án-Tapia (e.g. used in Bel án Tapia e al., 2021)25. The
au ho s wan o g ea ly acknowledge his help in p o iding his da a, e ealing a e y high consis ency o bo h
da abases.
4
Scien i ic Da a | (2025) 12:810 | h ps://doi.o g/10.1038/s41597-025-05055-z
www.na u e.com/scien i icda a
www.na u e.com/scien i icda a/
in e media e municipali ies ha e be ween 2,000 and 10,000 inhabi an s. Ou da ase shows he denomina ion o
his classi ica ion in each decade. Mainly ollowing Albe us24 (also Monclús and Oyón, 198844, Villanue a and
Leal45) a dummy a iable is elabo a ed on whe he a municipali y has coloniza ion owns in he municipal a ea,
adding also a column on he decade in which i had i / hem and he own popula ion associa ed o i . A dummy
a iable is a bina y a iable (which akes alues 0 o 1) used in s a is ical models o ep esen ca ego ical da a
wi h wo le els, such as he p esence o absence o a cha ac e is ic. I allows quali a i e in o ma ion o be
included in quan i a i e analyses, helping assess he impac o ca ego ical ac o s on ou comes.
Based on Es eban-Oli e and Ma í-Hennebe g26,27 we compu e ( o each decade and yea ) dis ances o he
municipali y (cen oid) o High-Speed Railway (HSR), o Ibe ian gauge and o na ow gauge s a ions. HSR lines
a e sec ions o ecen ly buil ack o high-speed ail se ices, since 1992, co esponding o he Eu opean s and-
a d gauge (1435 mm). Ibe ian gauge ne wo k is he la ges in Spain ( o med by lines which opened a e 1848
wi h he Ibe ian gauge wid h, 1668 mm) and he na ow-gauge ne wo k includes a ious ypes (wi h wid hs o
1435 mm, 1062 mm, 1000 mm, 915 mm and 750 mm). Simila ly, dis ances o ai po s in each decade/yea a e
compu ed using as e e ence he yea in which he passenge s’ ai po s ini ia ed hei ac i i y (AENA, 2025)46.
Fo all hose a iables, i should be no ed ha i e u ned an “in ” alue, i means ha he e was no such a s a-
ion o ai po in he decade wi hin Spain (hence he algo i hm e u ns “an in ini e” dis ance).
Ou e e ence da ase could be in e es ing o he en i onmen al, his o ical, social and economics li e a u e,
due o he ans e sali y be ween di e en ypes o a iables and he long- ime span pe iod i co e s (mo e han
a cen u y), as well as da a wi h a high le el o disagg ega ion, conside ing he municipali ies. The di e si y o
he da ase allows o de elop esea ch wi h se e al a iables included in i o combined wi h o he a iables
ob ained by o he esea che s. Besides, his da ase allows o e alua e and con ibu e o se e al deba es in he
cu en li e a u e. One o he po en ial uses o he da a is on he causes, e ec s and consequences o clima e
change in a Medi e anean coun y, allowing he analysis o he long- e m ends o hese a iables and he
e alua ion o causal e ec s. Fo example, Spain has one o he highes le els o hyd ic s ess due o clima e
change in he EU, mainly due o he po en ial dese i ica ion o a high pe cen age o i s e i o y47–49. Indeed,
one o he key u u e wo ks wi h he da abase is o u he p ojec and unde s and, based on pas ends, how
a iables such as empe a u e o d ough indica o s a ec ed by clima e and en i onmen al change, a e expec ed
Va iables Name in he da ase Da a sou ce Timespan F equency Uni s
D yland su ace d yland HYDE 3.2, HID, GIA and ESYRCE
(Minis y o Ag icul u e) 1900–2020
(1950–2021) Decadal (& Yea ly) Hec a es
Pas u es su ace pas u es HYDE 3.2, HID, GIA and ESYRCE
(Minis y o Ag icul u e) 1900–2020
(1950–2021) Decadal (& Yea ly) Hec a es
I iga ed su ace i iga ed HID, GIA, ESYRCE (Minis y o
Ag icul u e) and HYDE 3.2 1900–2020
(1950–2021) Decadal (& Yea ly) Hec a es
Cul i a ed a ea cul i a ed_a ea HYDE 3.2, HID, GIA and ESYRCE
(Minis y o Ag icul u e) 1900–2020
(1950–2021) Decadal (& Yea ly) Hec a es
I iga ed su ace o bo de ing
municipali ies i ig_neighb HID, GIA, ESYRCE (Minis y o
Ag icul u e) and HYDE 3.2 1900–2020
(1950–2021) Decadal (& Yea ly) Hec a es
Cul i a ed a ea o bo de ing
municipali ies cul i a ed_a ea_neighb HID, GIA, ESYRCE (Minis y o
Ag icul u e) and HYDE 3.2 1900–2020
(1950–2021) Decadal (& Yea ly) Hec a es
Table 3. Lis , sou ces, imespan and uni s o land use a iables.
Va iables Name in he da ase Da a sou ce Timespan F equency Uni s
Ri e basin Basin MITECO (2022) S a ic a iable Ca ego ical a iable
Rese oi a ea in he
municipali y Rese oi _a ea MITECO (2011, 2025) 1900–2020
(1950–2021) Decadal (& Yea ly) Squa e kilome e s
Rese oi wa e olume
capaci y Rese oi _ olume Global Dam Wa ch (GDW), MITECO
(2011, 2025) and SEPREM 1900–2020
(1950–2021) Decadal (& Yea ly) Hm3
Usable olume capaci y
o dammed wa e Usable_ ese oi _ olume Global Dam Wa ch (GDW), MITECO
(2011, 2025) and SEPREM 1900–2020
(1950–2021) Decadal (& Yea ly) Hm3
Main use o he
Rese oi Rese oi _main_use Global Dam Wa ch (GDW) 1900–2020
(1950–2021) Decadal (& Yea ly) Ca ego ical a iable
Rese oi wa e olume
capaci y by use
Vol_I iga ion
Vol_Elec ici y
Vol_Supply
Vol_O he
Global Dam Wa ch (GDW), MITECO
(2011, 2025) and SEPREM 1900–2020
(1950–2021) Decadal (& Yea ly) Hm3
The nea es main
i e (basin su ace
>500 km2)nea es _main_ i e MITECO (2018) S a ic a iable Ca ego ical a iable
Dis ance o he nea es
main i e Dis _main_ i e s MITECO (2018) S a ic a iable Km
The nea es wa e cou se nea es _all_ i e MITECO (2018) S a ic a iable Ca ego ical a iable
Dis ance o he nea es
wa e cou se Dis _all_ i e s MITECO (2018) S a ic a iable Km
Table 4. Lis , sou ces, imespan and uni s o hyd ological a iables.
5
Scien i ic Da a | (2025) 12:810 | h ps://doi.o g/10.1038/s41597-025-05055-z
www.na u e.com/scien i icda a
www.na u e.com/scien i icda a/
o a ec socio-economic a iables and pe o mance. O he ypes o po en ial s udies can be de eloped ela ed
o he en i onmen al impac s o economic ac i i ies, and ice e sa50–53. Al e na i e u u e applica ions a e he
analysis o he causal ela ionship be ween some geog aphical o hyd ological cha ac e is ics o he e i o y
and economic ac i i ies54 and/o demog aphic/se lemen decisions. In his sense, he da abase has al eady been
used in Cazca o e al.54 o s udy he ex en o which i iga ed ag icul u e has con ibu ed o popula ion change
in Spanish municipali ies. Ano he example o po en ial use in his line is he analysis o he de e minan s o
popula ion loca ion, agglome a ion p oblems and ma ke po en ial31,55,56.
Figu e1 summa ises and classi ies he p ocess o ob aining each o he a iables co esponding o he di e -
en hema ic blocks o he da abase.
Me hods
The echniques used o he elabo a ion o he da abase p esen ed in his a icle a e based on a a ie y o s a is i-
cal app oaches, GIS and da a analysis echniques o ien ed especially o he homogenisa ion and spa io- empo al
e-scaling o da a om p ima y sou ces. These echniques a y widely, depending la gely on he na u e o he
o iginal da a and he o ma in which hey a e o iginally ound. In his sec ion, he me hodological p ocess ol-
lowed o in eg a e each o he a iables in o he da abase is de ailed by hema ic g oup o da a. Despi e he di e -
si y o applied echniques, o ensu e spa ial- empo al cohe ence in he analysis, we ha e used a s a ic da abase
o municipali ies wi h homogeneous adminis a i e bounda ies om 2016 (a ailable a Geog aphical Na ional
Ins i u e INE web po al) h oughou he s udy pe iod. This app oach is essen ial o a oid dis o ions a ising
om changes in municipal bounda ies, which could a ec he in e p e a ion o key a iables.
Clima ic a iables. As can be seen in Table1, he clima e da a a e om CRU TS. Ve sion 4.05. The only
excep ion is he SPEI d ough index, which comes om he po al gene a ed by he Py enean Ins i u e o Ecology,
pa o he Spanish Na ional Resea ch Council (IPE-CSIC). The SPEI used is he 1-mon h ime esolu ion SPEI.
F om hese da a, he en i e his o ical se ies o SPEI o each o he Spanish municipali ies has been cons uc ed
and agg ega ed by decades o homogenise i wi h he es o he da abase.
Va iables Name in he da ase Da a sou ce Timespan F equency Uni s
Numbe o inhabi an s popula ion Goe lich (2019) and INE 1900–2010 Decadal Inhabi an s
Popula ion esiding in municipali ies
wi h mo e han x popula ion (in
housands) ha a e loca ed wi hin a
adius be ween i and j km away (i < j);
Px_i_j_km
p10_25_50km
p50_25_50km
p100_25_50km
p500_25_50km
p10_0_25km
p50_0_25km
p100_0_25km
p500_0_25km
Bel án Tapia e al.25 and INE 1900–2010 Decadal Inhabi an s
Dis ance o he nea es municipali y
o mo e han 10,000 inhabi an s dis ance_pop_10000 INE 1900–2010 Decadal Kilome es
Dis ance o he nea es municipali y
o mo e han 5,000 inhabi an s dis ance_pop_5000 INE 1900–2010 Decadal Kilome es
Classi ica ion o Simpson’s egion Simpson_A eas_5
Simpson_A eas_11 Simpson (1995) S a ic a iable Ca ego ical a iable
Type o he municipali y wi h espec
o he popula ion size (u ban,
in e media e, u al) Popula ion_class INE 1900–2020 Decadal Ca ego ical a iable
Dummy o whe he a municipali y
has coloniza ion owns in he
municipal a ea municip_coloniz_in_municip Monclús and Oyón (1988)44, Villanue a
and Leal (1991)45 and Albe us (2023) 1900–2020
(1950–2021) Decadal (& Yea ly) Boolean (0–1)
Decade o c ea ion o he coloniza ion
own in he municipal a ea decade_coloniz_in_municip Monclús and Oyón (1988)44, Villanue a
and Leal (1991)45 and Albe us (2023) 1900–2020
(1950–2021) Decadal (& Yea ly) Decade
Popula ion o coloniza ion owns in
he municipali y ( e e ence) pop_coloniz_in_municip Albe us (2023), Goe lich (2019) and INE 1900–2020
(1950–2021) Decadal (& Yea ly) Inhabi an s
Dis ance o he closes Ibe ian Gauge
ailway s a ions ailways a ion_dis ance Es eban-Oli e and Ma í-Hennebe g 2023 1900–2021
(1950–2023) Decadal (& Yea ly) Kilome es
Name o he closes Ibe ian Gauge
ailway s a ions nea es _s a ion_name Es eban-Oli e and Ma í-Hennebe g, 2023 1900–2021
(1950–2023) Decadal (& Yea ly) Ca ego ical a iable
Dis ance o he closes High-Speed
Railway (HSR) s a ions highspeeds a ion_dis ance Es eban-Oli e and Ma í-Hennebe g, 2023 1900–2021
(1950–2023) Decadal (& Yea ly) Kilome es
Name o he closes High-Speed
Railway (HSR) s a ions nea es _highspeed_name Es eban-Oli e and Ma í-Hennebe g, 2023 1900–2021
(1950–2023) Decadal (& Yea ly) Ca ego ical a iable
Dis ance o he closes Na ow Gauge
ailway s a ions na ow ail_line_dis ance Es eban-Oli e and Ma í-Hennebe g, 2023 1900–2021
(1950–2023) Decadal (& Yea ly) Kilome es
Name o he closes Na ow Gauge
ailway s a ions na ow ail_line_name Es eban-Oli e and Ma í-Hennebe g, 2023 1900–2021
(1950–2023) Decadal (& Yea ly) Ca ego ical a iable
Dis ance o he closes (o 48) na ional
ai po ai po _dis ance Based on AENA46 1900–2021
(1950–2023) Decadal (& Yea ly) Kilome es
Table 5. Lis , sou ces, imespan and uni s o socioeconomic-demog aphic a iables.

6
Scien i ic Da a | (2025) 12:810 | h ps://doi.o g/10.1038/s41597-025-05055-z
www.na u e.com/scien i icda a
www.na u e.com/scien i icda a/
In any case, he me hods used o he da a om bo h sou ces a e iden ical, as is he dis ibu ion o he da a.
Bo h sou ces dis ibu e he da a in a as e o ma wi h a cell size o 0.5° × 0.5° wi h a mon hly ime esolu ion.
The GIS ope a ion known as zonal s a is ics was used o e-scale he da a o he municipal le el, s a ing om he
as e o ma wi h he clima ic a iable in ques ion and a shape ile o he municipali ies o Spain. This ope a ion
applies, o each municipali y, an a e age o he alue o he cells o he as e ile con ained in he municipal
polygon weigh ed by he p opo ion o he su ace a ea o he cell included wi hin he polygon. Thus, h ough
Eq.1, we can app oxima e he alue o each o he clima e a iables a he municipal le el.
=
∑



⋅





∑
∈
∈
C
(1)
m
ij R
A
Acij
ij R
A
A
,,,
,
c
ij
m
ij
c
ij
m
ij
,
,
,
,
In Eq.1,
∈ijR,l
means all cell
i j,
o he clima ic a iable
C
as e ma ix
;
Aij
, e e s o he o al a ea o cell
i j,
;
Aij
m
,
is he a ea o he cell con ained in municipali y
m
and
cij,,
is he alue o he cell (clima ic a iable
c
). The
applica ion o zonal s a is ics has been implemen ed h ough he R package exac ex ac 57. In Fig.2, we can see
a schema ic o how he zonal s a is ics ope a ion wo ks as e lec ed in equa ion numbe 1. This igu e is also
ep esen a i e o equa ion numbe 2, in which we will simply compu e he weigh ed s anda d de ia ion ins ead
o he weigh ed mean.
Once he downscaling o he municipal le el has been ca ied ou in his way, we ob ain a mon hly alue.
Since we need o agg ega e he da a a a decadal le el in o de o homogenise he da a a a empo al le el, we use
he mos app op ia e s a is ic (a e age, sum, e c.) o agg ega e hem a an annual le el, a e which he decadal
a e age is ca ied ou .
Al hough his was he common me hodological p ocess, some o he a iables ha e equi ed complemen a y
and/o al e na i e me hodological p ocedu es. The a iable g ow_pe iod_pp, is one o hem (see Fig.3). This
a iable e e s o he o al p ecipi a ion ecei ed du ing he mon hs o plan g ow h, i.e. he mon hs om Ap il
o Oc obe (bo h included). The empo al esolu ion o he o iginal da a (CRU TS. Ve sion 4.05) is mon hly.
The e o e, a e calcula ing he o al mon hly p ecipi a ion ecei ed in he municipali y, he g owing mon hs o
each o he yea s we e il e ed ou and he annual o al added. To ob ain he decadal alue, a simple a e age was
aken be ween he yea s o he decade. The o he a iables ha equi ed special ea men due o he na u e o
he o iginal da a we e hose ela ed o he Köppen clima e classi ica ion. Gi en ha he o iginal da a (Na ional
Geog aphic Ins i u e o Spain) a e in a polygonal shape ile o ma , whe e each polygon ep esen s a speci ic ype o
clima e, o he ypes o p ocesses we e needed o adap hem o a municipal amewo k. Fo his pu pose, he GIS
ope a ion known as in e sec was used om he shape iles o he Köppen clima e classi ica ion and he municipal
one. This has c ea ed a polygon o each ype o clima e in ela ion o each municipali y. In his way, we ob ain he
clima e ypes ound in each o he Spanish municipali ies and gene a e a dummy a iable o each o hem.
Fig. 1 S uc u e o he de elopmen o a iables h ough each o he 5 blocks o a iable ypes.
7
Scien i ic Da a | (2025) 12:810 | h ps://doi.o g/10.1038/s41597-025-05055-z
www.na u e.com/scien i icda a
www.na u e.com/scien i icda a/
Geog aphical a iables. The geog aphical a iables a e basically aspec s o dis ance, loca ion and elie .
As we can see in Table2, we ha e he dis ance o Mad id (capi al o Spain), he dis ance o he capi al o he
p o ince and he dis ance o he coas . We also ha e he la i ude (y) and longi ude (x) coo dina es. The i s s ep
in calcula ing hese a iables is o ex ac he cen oids o each municipali y om he shape ile o he Spanish
municipali ies o he Na ional Geog aphic Ins i u e58, using he QGIS algo i hm Gene a e poin s (pixel cen oids)
inside polygons59. F om he cen oids o he municipali ies, he X and Y coo dina es we e es ima ed om he ad
hoc algo i hm o he ield calcula o in QGIS. Once hese had been calcula ed, he dis ance a iables we e also
calcula ed om he municipal cen oids. The dis ance a iable o Mad id and o he p o incial capi al was pe -
o med by gene a ing 2 independen laye s o each o hem h ough a selec ion by a ibu es, using he R package
dply 60 on he shape ile a ibu e able loaded in an s objec 61. An s objec (sho o simple ea u es objec ) in R
is a da a s uc u e designed o handling spa ial da a e icien ly while adhe ing o in e na ional s anda ds. I com-
bines geome ic in o ma ion (such as poin s, lines, and polygons) wi h a ibu e da a in a o ma simila o a da a
ame. This makes i easy o in eg a e spa ial analysis wi h common R wo k lows. Addi ionally, s objec s suppo
coo dina e e e ence sys ems (CRS), enabling p ecise spa ial analysis and isualiza ion.
A e ha , he dis ance o all he municipali ies cen oids o each o he cen oids o he laye s was calcula ed
and il e ed o ma ch, in he case o he p o incial capi al, he en i y o he same p o ince as he municipali y
in ques ion. The QGIS algo i hm Dis ance o nea es hub (poin s) implemen ed in he R package qgisp ocess62
was used o his pu pose. Fo he coas line, he p ocedu e was simila , excep ha he dis ance be ween he
cen oids o he municipali ies and he coas line was calcula ed using he Dis ance o nea es hub (line o hub)
algo i hm, also implemen ed in he R package qgisp ocess. The municipal a ea was calcula ed using he $a ea
algo i hm in he QGIS ield calcula o applied o he municipali ies’ shape ile. Finally, al i ude and uggedness
we e again calcula ed using he zonal s a is ics ope a ion implemen ed in he exac ex ac R package om he
shape ile o municipali ies (zones) and a digi al ele a ion model. In he case o al i ude, he weigh ed mean is
used in he same way as in Eq.1. Ruggedness, as we know, can be es ima ed as he s anda d de ia ion o he al i-
ude o each municipali y, so he zonal s anda d de ia ion is applied acco ding o Eq.2 (see also Fig.4).
H()
(2)
m
ij R
A
Acijm
ij R
A
A
,,,
,
c
ij
m
ij
c
ij
m
ij
,
,
,
,
σ=
∑


⋅−





∑
∈
∈
The nomencla u e is he same as in Eq.1, adding
Hm
, which is he a e age al i ude in each municipali y.
Land use a iables. The land use a iables a e ela ed o ag icul u al uses. Among hem, we can dis inguish
be ween he a ea o ain ed c ops, he a ea o i iga ed c ops and he a ea o pas u e. The a ea o i iga ed c ops
(equipped o i iga ion, o be speci ic) comes om he His o ical I iga ion da ase (HID hence o h) (Siebe e al.)16
and om he Global I iga ion A ea (GIA) da abase (Meie e al.)18 combined wi h C op a ea and yield su ey
(ESYRCE he ea e 17) o ex end he se ies o he decade o he 2010s and 2020 s (see Fig.5). This a iable includes
he i iga ed land unde plas ic g eenhouses. Since he o ma o he main clima ic a iables is iden ical (da a in
as e o ma wi h a spa ial esolu ion o 0.5°, he alue o each cell e e s o he numbe o hec a es con ained in
i ), he p ocedu e is simila . Fo he o he land uses, he o ma is he same, bu he sou ce o he da a is he
His o y Da abase o he Global En i onmen (HYDE e sion 3.2)15. In his case, in o de o know he o al munic-
ipal hec a es o each land use, a sum weigh ed by he a ea o he cell loca ed in he municipal polygon o he hec-
a e alue o each cell is made. The calcula ion o he municipal hec a es o land use can be exp essed acco ding
o Eq.3. The nomencla u e o Eq.3 is e y simila o Eq.1, only he sub-index
c
is eplaced by he sub-indice
l
,
which e e s o each land use. This me hod has been used o he a iables D yland (d yland), Pas u es (pas u e)
and I iga ed (i iga ed_ha) a ea.
7543
8752
7661
6543
0.15 0.70.7 0.2
0.65 110.7
0.61 10.3
0.05 0.30.2 0.1
X
Pixel aluesPixel aoco e ed by
p o ince a ea


P o ince Value
5.3
Fig. 2 G aphic scheme o zonal s a is ics p ocessing ope a ion.
8
Scien i ic Da a | (2025) 12:810 | h ps://doi.o g/10.1038/s41597-025-05055-z
www.na u e.com/scien i icda a
www.na u e.com/scien i icda a/
∑
=⋅
∈
L
A
A
(3)
mij R
ij
m
ij lij
,
,
,,,
l
The a iable Cul i a ed a ea is de i ed om he sum o D yland Su ace and I iga ed Su ace. The a iables
I iga ed su ace o adjacen municipali ies and Cul i a ed su ace o adjacen municipali ies, on he o he hand,
ha e unde gone a di e en ans o ma ion. In bo h cases, he i s s ep has been o gene a e an adjacency ma ix,
i.e., o each o he municipali ies, he municipali ies ha a e in con ac wi h i ake a alue o 1, and 0 o he wise.
F om his ma ix, he alue o he I iga ed o Cul i a ed su ace o hose municipali ies ha ha e a alue o 1,
i.e. he municipali ies ha a e adjacen o he municipali y in ques ion, is added. This p ocess is ep esen ed by
Eq.4. In i ,
C
is a con igui y ma ix o size
nxn
, whe e
n
is he numbe o municipali ies. The e o e,
Cmj,
will be
an elemen o he ma ix ha is 1 i municipali y
m
is adjacen o municipali y
j
, and 0 o he wise. On he o he
hand,
Vj
will be a ec o o size
n
con aining he land use alues in hec a es o each municipali y.
∑
=⋅
=
SCV(4)
mj
n
mj j
1
Hyd ological a iables. Hyd ological a iables a e essen ially linked o he access o di e en municipal-
i ies o wa e esou ces, wa e cou ses o a iables ela ed o wa e in as uc u e such as ese oi s (see Fig.6).
These a iables a e de ailed in Table4. To calcula e he basin a iable, he GIS Spa ial join ope a ion was used,
implemen ed in QGIS using he municipali y shape ile and he hyd og aphic basin shape ile (Minis y o he
Ecological T ansi ion and he Demog aphic Challenge)10. In his ope a ion, each o he municipali ies akes he
alue o he hyd og aphic basin in which i is loca ed. I a municipali y is loca ed be ween wo basins, i akes
he alue o he basin ha co e s he la ges a ea o he municipali y. Then, we ha e he a iables ela ed o he
ese oi s, i.e., he ese oi a ea, he ese oi olume and he usable ese oi olume. The di e ence be ween
o al ese oi olume and usable ese oi olume lies in hei speci ic de ini ions and unc ions. To al ese oi
olume e e s o he en i e amoun o wa e s o ed in he ese oi a a gi en ime, encompassing all he wa e
om he bo om o he su ace, including po ions ha may no be accessible o p ac ical use. In con as , usable
ese oi olume ep esen s he po ion o wa e ha can be ac i ely u ilized o he ese oi ’s in ended pu poses,
such as i iga ion, wa e supply, hyd oelec ic powe gene a ion o lood con ol. This excludes he so-called “dead
olume”, which emains below he ou le le el o is e ained o p e en sedimen esuspension and ensu e he
ese oi ’s s uc u al s abili y. As a esul , usable ese oi olume is always less han o equal o he o al ese oi
olume, wi h he di e ence de e mined by he ese oi ’s design and ope a ional pa ame e s.
Fig. 3 E olu ion o g owing season p ecipi a ion.
9
Scien i ic Da a | (2025) 12:810 | h ps://doi.o g/10.1038/s41597-025-05055-z
www.na u e.com/scien i icda a
www.na u e.com/scien i icda a/
To calcula e hese a iables, we ha e used he ese oi shape ile om he in en o y o dams and ese oi s
o he Minis y o he Ecological T ansi ion and he Demog aphic Challenge (MITECO)9 and he in en o y o
dams o he Spanish Socie y o Dams and Rese oi s (SEPREM)14. The GIS in e sec ope a ion, implemen ed in
QGIS, was used o calcula e he ese oi a ea o each municipali y. As p e iously men ioned, his ope a ion has
ob ained a polygon o each pa o he ese oi loca ed in each municipali y. Once we ha e he polygons o he
ese oi s di ided by municipali ies, we calcula e he a ea, using he s _a ea unc ion o he R s package61. We
mus ake in o accoun ha municipali ies do no always ha e he same ese oi capaci y o e ime, since dams
and ese oi s a e buil a di e en imes by di e en hyd ological p ojec s. Fo his eason, he SEPREM in en-
o y o ese oi s was used o de e mine he yea in which he cons uc ion o each ese oi was comple ed. In
o de o ob ain he decadal alue o he ese oi a ea by municipali y, we il e o each decade by he yea o
cons uc ion, g oup by municipali y and add up he o al ese oi a ea (i is possible ha a municipali y has
mo e han one ese oi ). These il e ing and summing p ocesses by municipali y we e execu ed in R om he
dply package. Conside ing ha he a ea o each ese oi in a gi en municipali y is exp essed in Eq.5 whe e
Pm
is he polygon o in e sec ion be ween he municipali y
m
and he ese oi
:
AA eaP() (5)
m m ,,
=
The o al ese oi a ea in he municipali y m o he decade y is equal o:
∑
=
∈
AA (6)
my y m ,,
Once we ha e calcula ed he ese oi a ea, we ob ain he alues o he olume and use ul olume dammed pe
municipali y. MITECO’s (20119, 202512) in en o y o dams and ese oi s con ains he da a on he olume and use ul
olume o each ese oi . The e o e, all ha emains is o dis ibu e he olume o he ese oi among he di e en
municipali ies be ween which he ese oi is loca ed. This dis ibu ion is done in p opo ion o he p e iously cal-
cula ed a ea o he ese oi ha is loca ed in each municipali y. Thus, he calcula ion o he ese oi olume and he
municipal use ul ese oi olume pe ese oi
and municipali y
m
is cha ac e ised acco ding o Eq.7:
=⋅V
A
AV
(7)
m m
,,
Again, we ha e o il e by yea and make a sum g ouped by municipali y. The e o e, he calcula ion o he
ese oi olume and he o al use ul ese oi olume by municipali y is as ollows:
Fig. 4 Ruggedness by municipali y.
16
Scien i ic Da a | (2025) 12:810 | h ps://doi.o g/10.1038/s41597-025-05055-z
www.na u e.com/scien i icda a
www.na u e.com/scien i icda a/
alues compa ed o FAO-AQUASTAT in he p o ince o Za agoza (13,962 ha). The highes pe cen age disc ep-
ancy ela i e o he i iga ed a ea i sel we e ound in p o inces wi h smalle i iga ed a eas like Guipuzcoa,
which is one o he mos humid Spanish p o inces. Absolu e disc epancies logically occu ed in p o inces wi h
la ge i iga ed a eas. The WAPE a p o incial le el is 6.47%, showing a high deg ee o conco dance be ween he
es ima ed HID and FAO-AQUSTAT alues.
O e all, he co ela ion be ween he es ima ed and FAO-AQUASTAT i iga ion alues is high, wi h coe i-
cien s o 1 a bo h he egional and p o incial le els, indica ing he obus ness o he GIS-based es ima es used
in his s udy. In his espec , Fig.12 shows he compa ison by p o ince o he es ima ed i iga ion alues om
HID in ou da abase ( o he yea 2000) and hose eco ded by FAO-AQUASTAT es ima es.
I should be no ed ha he HID i iga ion se ies only up o he yea 2005. The e o e, in o de o achie e a
high deg ee o accu acy, di e en me hods we e used. Fi s ly, he In o ma ion Sys em on Land Use in Spain
(hence o h, SIOSE, Sis ema de In o mación de Ocupación del Suelo en España)68 was ini ially used as a da a
sou ce. This da abase p o ides comp ehensi e geospa ial da a on land use and land co e h oughou Spain. This
da ase o e s de ailed and egula ly upda ed in o ma ion, which makes i an in e es ing esou ce o e i o ial
planning and land use change s udies. Howe e , when app oaching h ough i and using GIS ope a ions such as
In e sec , a ea calcula ions, e c., he di e ences in he o iginal da a ( ec o o ma s. as e o ma , me hodolog-
ical di e ences in i s es ima ion, e c.) led us o conside o he app oaches. In his con ex , he Global I iga ion
Da ase (GID) men ioned abo e had an iden ical o ma , and i s me hodology was e y simila . In ac , i is
based on he HID se ies i sel , ex ended wi h di e en emo e sensing me hods. In his way, i was also possible
o apply exac ly he same me hodology (based on he zonal s a is ics GIS ope a ion) o es ima e he municipal
i iga ed hec a es o he whole se ies. Compa ing he esul s ob ained wi h he wo da a sou ces, i was possible
o app ecia e a mo e homogeneous se ies wi h much less quan i a i e jumps wi h he decades es ima ed om
he HID sou ce (1900 – 2000). Thus, he app oxima ion using he GID sou ce was chosen o ob ain homogenei y
and o educe he e o .
Besides, o achie e he highes possible deg ee o p ecision in he adjus men , especially in ecen dec-
ades, o icial s a is ics we e used. In his case, i is he ESYRCE. Fo his alida ion, a ela ionship is es ablished
be ween he o icial s a is ics and he es ima ed i iga ed a ea. The o icial ESYRCE17 s a is ics p o ide da a
on ag icul u al land, including i iga ed a eas, a he NUTS 3 le el. Ou goal is o align he g ow h a e o ou
es ima ed a iable a he NUTS 3 le el wi h he ESYRCE s a is ics. Fi s , we adjus ou es ima ed a iable o
he yea 2000 o ma ch he NUTS 3 o als p o ided by ESYRCE. Nex , we calcula e he g ow h a e be ween he
adjus ed i iga ion alues o 2000 and he ESYRCE da a o 2010. Then, we calcula e he g ow h a e be ween
he adjus ed i iga ion alues o 2000 and he ESYRCE da a o 201017. We hen sub ac he 2000 es ima e om
his new 2010 alue, o he absolu e change in hec a es a he NUTS 3 le el, which needs o be dis ibu ed ac oss
municipali ies. To dis ibu e his change, we calcula e each municipali y’s sha e o he change in i iga ion based
on he es ima ed change wi hin i s espec i e Au onomous Communi y. This p opo ion is used o alloca e he
NUTS 3 le el change among he municipali ies. Finally, we add he municipal-le el change (2000–2010) o he
i iga ion es ima e o 2000 de i ed om he HID se ies, allowing us o app oxima e he 2000–2010 change
based on o icial o als while main aining a consis en s uc u e ac oss all decades. Thus, we ind ha he g ow h
a es es ima ed om GIA do no de ia e om he o icial g ow h a es. To upda e i iga ion o he 2020 s in he
Fig. 12 Accu acy assessmen o es ima ed i iga ed a ea. No e: In his igu e we can see he compa ison be ween
he FAO-AQUASTAT alues o 1999 a NUTS 3 le el ( ed iangle) and he es ima ed agg ega ed municipal
i iga ion in ou s udy a he same le el (blue ci cle) in o de o analyse he simila i y o ou es ima es.

17
Scien i ic Da a | (2025) 12:810 | h ps://doi.o g/10.1038/s41597-025-05055-z
www.na u e.com/scien i icda a
www.na u e.com/scien i icda a/
da abase, he same p ocedu e has been used, bu aking as a e e ence he alues o 2011 and 2021 by es ima ing
he changes be ween bo h yea s.
As can be seen om he o icial s a is ics, he me hodology is e y close o he o als o he mos ecen
decades. Howe e , hese s a is ics do no p o ide us wi h a e y long ime se ies, so ha he e i ica ion o he
p e ious ones is no co e ed, and u he mo e no o als ha e been e i ied, nei he o he cul i a ed a ea no o
he ain ed c ops. Fo his eason, we ha e used o he his o ical li e a u e sou ces, such as Ca e as & Ta unell35,
pa icula ly on he p ima y sec o . Chap e 4 o his wo k, on he ag icul u al and ishing sec o 69, con ains
his o ical s a is ics on he cul i a ed a ea, wi h discon inui ies, also disagg ega ed be ween ain ed and i iga ed
c ops a he na ional le el. Using he p e iously men ioned me ic, WAPE, we can es ima e he deg ee o adjus -
men o ou se ies in compa ison wi h he his o ical s a is ics. Be o e analysing he esul s o his compa ison,
we ha e o ake in o accoun ha we a e compa ing sligh ly di e en concep s. The es ima ed i iga ion da a col-
lec he a ea equipped o i iga ion, ega dless o whe he o no i was i iga ed ha yea . Ca e as & Ta unell
(2005), on he o he hand, collec he a ea ha was e ec i ely i iga ed based on he censuses. These concep ual
di e ences can cause signi ican disc epancies be ween he igu es o one and he o he .
In Eq.9,
yi
is he alue es ima ed in ou se ies and
yi
ˆ
is he alue o he his o ical se ies o 35. Using his me ic,
o he whole se ies we ha e ob ained a di e ence o 4.3% o he o al cul i a ed a ea, 2.8% o he ain ed cul i-
a ed a ea and 11.6% in he case o i iga ion. Table6 shows he absolu e di e ence by decade in de ail wi h
alues exp essed in housands o hec a es.
Wi h espec o clima e a iables, he CRU TS Ve sion 4.0519 da abase was used o his s udy due o i s ex en-
si e empo al co e age (since 1901) and he wide a ailabili y o clima e a iables a he global le el, which allows
a de ailed his o ical analysis o he ela ionship be ween clima e and he demog aphic and economic de elop-
men o municipali ies. I s spa ial esolu ion o 0.5 deg ees acili a es i s use in la ge-scale s udies, and i s consis -
ency in da a collec ion and p ocessing me hodology makes i a eliable sou ce o analysing clima e pa e ns o e
ime. Fu he mo e, he as e dis ibu ion acili a es downscaling h ough ope a ions such as zonal s a is ics,
as p e iously explained. Howe e , his same spa ial esolu ion o app oxima ely 50 km pe cell can be seen as
a weakness when applied a he municipal le el, especially in a eas wi h a iable opog aphy o mic oclima es,
whe e he da a may no adequa ely cap u e local clima ic a ia ions. Fu he mo e, as a global da abase, i may
p esen limi a ions in e ms o da a accu acy in a eas wi h spa se o poo ly de eloped me eo ological ne wo ks.
Gi en he p oblems discussed abo e, an app oxima ion o he accu acy o ou da a a di e en scales and a -
iables was made by compa ing hem wi h di e en da abases. We ha e o be cau ious when making compa isons
wi h clima e da a, as da a om di e en sou ces come om di e en measu emen echniques/me hodologies,
each wi h i s own ma gin o e o . A he na ional le el, he epo “Analysis o empe a u es in Spain in he pe iod
1961–201870 om he S a e Me eo ological Agency was used. In his epo we ind an a e age empe a u e
se ies o Spain since 1961 on a na ional scale, es ima ed based on da a om 42 e e ence s a ions. Using ou
da abase and agg ega ing he da a o he epo by decades (simple a e age), we ob ain a WAPE o 10%.
Fo he accu acy analysis o he p ecipi a ion a iable, a alida ion a he municipal le el was applied. Fo his
pu pose, he example o Ba celona was used, gi en he leng h o i s se ies (since 1940 wi h consis en da a). Daily
da a om he Eu opean Clima e Assessmen da ase (ECAD)71 o he El P a Ai po s a ion we e u ilised. These
daily da a, as in he es ima ion o ou p ecipi a ion a iable, we e summed annually, and he decadal mean was
calcula ed om he annual da a. Again, he WAPE alue eaches 10%. The same p ocess has been ca ied ou ,
excep ha he annual agg ega ion is done by means o he a e age and no he sum, wi h he se ies o a e age
empe a u e also ela i e o he Ba celona Ai po wea he s a ion. In his case, he WAPE eaches 8% o he
whole se ies. Taking in o accoun he unce ain ies associa ed wi h he da abases compa ed, we conside ha
his le el o accu acy gua an ees p ecise and consis en analyses when using clima e da a.
Rega ding he hyd ological a iables, gi en he me hodology and he sou ces used (o icial go e nmen s a-
is ics), he echnical checks ca ied ou we e ha , a e he alloca ion o he a eas and olume o ese oi s, he
o als o su ace a ea and olume o he ese oi s eco ded in he MITECO ese oi in en o y and he na ional
o al we e me , and in ac , hey we e me .
By way o summa y, he Table7 gi es an o e iew o he accu acy alues ob ained in he echnical alida ion
p ocess.
I is wo h no ing ha he e a e many a iables (in addi ion o he non-nume ical ones) ha a e no explici ly
alida ed h ough WAPE. Va iables such as he calcula ion o dis ances om he cen oid o he municipali y
o di e en en i ies ha e no been alida ed using his me hodology. As i is a calcula ion o Euclidean dis ance
Yea
Es ima ed Ca e as & Ta unell (2005) Absolu e e o (%)
I iga ed Rain ed To al c op a ea I iga ed Rain ed To al c op a ea I iga ed Rain ed To al c op a ea
1900 — — 16060 — — 17822 — — 9,9
1930 — — 19120 — — 21964 — — 12,9
1960 2226 18361 20587 1828 18694 20522 21,7 1,8 0,3
1970 2770 17763 20533 2198 18321 20520 26,0 3,0 0,1
1980 2932 17145 20077 2822 17677 20499 3,9 3,0 2,1
1990 2978 16494 19472 3199 16973 20172 6,9 2,8 3,5
2000 3667 14421 18088 3408 14897 18304 7,6 3,2 1,2
Table 6. C op a ea accu acy assessmen by decade.
18
Scien i ic Da a | (2025) 12:810 | h ps://doi.o g/10.1038/s41597-025-05055-z
www.na u e.com/scien i icda a
www.na u e.com/scien i icda a/
be ween wo a bi a ily chosen en i ies, he e is no o icial sou ce wi h which o sys ema ically check hem. The
alues o hese a iables ha e been checked o consis ency wi h speci ic examples.
Usage No es
The da a is also eely a ailable on he po al websi e h ps://his o eco.uniza .es, whe e addi ional ea u es and
possibili ies o il e ing, selec ing and downloading a e a ailable (e.g. que ies on he me ics o he di e en
municipali ies, compa isons, analysis o ends, e c.).
Code a ailabili y
The Code is a ailable on Gi Hub.
Recei ed: 8 No embe 2024; Accep ed: 23 Ap il 2025;
Published: xx xx xxxx
Re e ences
1. Seaw igh , J. Mul i-Me hod in Social Science: Combining Quali a i e and Quan i a i e Tools. (Camb idge Uni e si y P ess,
Camb idge, 2016).
2. Eu os a . Regions and ci ies. h ps://ec.eu opa.eu/eu os a /web/ci ies/da abase (2024).
3. UNECA. Geoin o ma ion in Socio-Economic De elopmen : De e mina ion o Fundamen al Da ase s o A ica. (2007).
4. Oli ei a, V., Sousa, V. & Dias-Fe ei a, C. Da ase o socio-economic and was e collec ion indica o s o Po ugal a municipal le el.
Da a B ie 22, 658–661 (2019).
5. Wang, J. The economic impac o Special Economic Zones: E idence om Chinese municipali ies. J De Econ 101, 133–147
(2013).
6. Hughes, S., Ki chho , C. J., Conede a, K. & F iedman, M. The Municipal D inking Wa e Da abase. PLOS Wa e 2, e0000081 (2023).
7. Fi a, J. H., Halse, A. H. & Na ik, G. J. Local go e nmen Da ase . www.jon. i a.no/da a.h m (2023).
8. INE. Demog a ía y población = Demog aphics and popula ion. P ep in a h ps://www.ine.es/dyngs/INEbase/es/ca ego ia.
h m?c=Es adis ica_P&cid=1254734710984 (2025).
9. MITECO. Embalses = Dams. Las Accessed 9-9-2024. h ps://wms.mapama.gob.es/sig/agua/Embalses/wms.aspx (2011).
10. MITECO. Cuencas hid og á icas de íos p incipales según P a s e e modi icado = Wa e sheds o majo i e s acco ding o
modi ied P a s e e . Las Accessed 8-11-2024. h ps://wms.mapama.gob.es/sig/Agua/CuencasRiosP a s/wms.aspx (2022).
11. MITECO. Ríos comple os clasi icados según P a s e e modi icado = Comple e i e s classi ied acco ding o modi ied P a s e e .
Las Accessed 9-9-2024. h ps://wms.mapama.gob.es/sig/Agua/RiosCompP a s/wms.aspx? (2018).
12. MITECO. Home -> Wa e -> Topics -> Wa e esou ces -> assessmen -> Hyd ological Bulle in -> Download o his o ical
ese oi da a since 1988. 02-11-2025. Las Accessed 3-3-2025. h ps://www.mi eco.gob.es/es/agua/ emas/e aluacion-de-los-
ecu sos-hid icos/bole in-hid ologico.h ml (2025).
13. Lehne , B. e al. The Global Dam Wa ch da abase o i e ba ie and ese oi in o ma ion o la ge-scale applica ions. Sci Da a 11,
1–18 (2024).
14. Sociedad Española de P esas y Embalses (SEPREM). In en a io de P esas Españolas. h ps://www.sep em.es/p esases.php?p=26
(2021).
15. Goldewijk, K. K., Beusen, A., Doelman, J. & S eh es , E. An h opogenic land use es ima es o he Holocene - HYDE 3.2. Ea h Sys
Sci Da a 9, 927–953 (2017).
16. Siebe , S. e al. A global da a se o he ex en o i iga ed land om 1900 o 2005. Hyd ol Ea h Sys Sci 19, 1521–1545 (2015).
17. Minis e io de Medio Ambien e y Medio Ru al y Ma ino. Encues a sob e Supe icies y Rendimien os Cul i os (ESYRCE). in Anua io
de Es adís ica (ed. Sec e a ía gene al écnica) ol. 1, 520–910 (Subdi ección Gene al de Es adís ica, Mad id, 2010).
18. Meie , J., Zabel, F. & Mause , W. A global app oach o es ima e i iga ed a eas - A compa ison be ween di e en da a and s a is ics.
Hyd ol Ea h Sys Sci 22, 1119–1133 (2018).
19. Ha is, I., Osbo n, T. J., Jones, P. & Lis e , D. Ve sion 4 o he CRU TS mon hly high- esolu ion g idded mul i a ia e clima e da ase .
Sci Da a 7, 109 (2020).
20. Begue ía, S., Se ano, S. M. V., Reig-G acia, F. & Ga cés, B. L. SPEIbase .2.10 [Da ase ]: A Comp ehensi e Tool o Global D ough
Analysis. P ep in a h ps://doi.o g/10.20350/digi alCSIC/16497 (2024).
21. T ullenque-Blanco, V., Begue ía, S., Vicen e-Se ano, S. M., Peña-Angulo, D. & González-Hidalgo, C. Ca alogue o d ough e en s
in peninsula Spanish along 1916–2020 pe iod. Sci Da a 11, 703 (2024).
22. Ins i u o Geog á ico Nacional (IGN). Clasi icación climá ica según Köppen. A las Nacional de España h ps://www.ign.es/web/
esou ces/docs/IGNCnig/ANE/Capi ulos/04_Climayagua.pd 103–104 (2010).
23. Goe lich, F. J. A municipal da abase om he 2011 Spanish census. Applied Economic Analysis 27, 226–238 (2019).
24. Albe us, M. The Poli ical P ice o Au ho i a ian Con ol: E idence om F ancois Land Se lemen s in Spain. Jou nal o Poli ics 85,
1258–1274 (2023).
Va iable Re e ence Scale o check WAPE
i iga ed FAO’s Aquas a Na ional 5.27%
i iga ed FAO’s Aquas a P o ince 6.47%
i iga ed His o ical I iga ion Da ase (HID) Na ional 3,2%
d yland Ca e as & Ta unell (2005) Na ional 2,8%
cul i a ed_a ea Ca e as & Ta unell (2005) Na ional 4,3%
_a e age S a e Me eo ological Agency Na ional 10%
_a e age Eu opean Clima e Assessmen da ase (ECAD) Municipal 8%
pp Eu opean Clima e Assessmen da ase (ECAD) Municipal 10%
Usable_ ese oi _ olume MITECO & SEPREM Dam in en o y Na ional 0%
Rese oi _ olume MITECO & SEPREM Dam in en o y Na ional 0%
Table 7. Accu acy alues ob ained in he echnical alida ion p ocess.
19
Scien i ic Da a | (2025) 12:810 | h ps://doi.o g/10.1038/s41597-025-05055-z
www.na u e.com/scien i icda a
www.na u e.com/scien i icda a/
25. Bel án Tapia, F. J., Díez-Minguela, A. & Ma inez-Gala aga, J. The shadow o ci ies: size, loca ion and he spa ial dis ibu ion o
popula ion. Annals o Regional Science 66, 729–753 (2021).
26. Es eban-Oli e , G. & Ma í-Hennebe g, J. The expansion o he Spanish ailway ne wo k (1848–1941): an analysis h ough he
e olu ion o i s companies. Re is a de His o ia Indus ial 31, 87–144 (2022).
27. Es eban-Oli e , G. & Ma í-Hennebe g, J. Railways o Spain GIS (1848–2023). P ep in a h ps://doi.o g/10.34810/da a917.
28. Menne, M. J., Du e, I., Vose, R. S., Gleason, B. E. & Hous on, T. G. An o e iew o he global his o ical clima ology ne wo k-daily
da abase. J A mos Ocean Technol 29, 897–910 (2012).
29. Sli inski, L. C. e al. Towa ds a mo e eliable his o ical eanalysis: Imp o emen s o e sion 3 o he Twen ie h Cen u y Reanalysis
sys em. Qua e ly Jou nal o he Royal Me eo ological Socie y 145, 2876–2908 (2019).
30. Ti ado-Fab ega , D. e al. SPAREL. spa el.com (2022).
31. Bel án Tapia, F. J. & Ma inez-Gala aga, J. Inequali y and G ow h in a De eloping Economy: E idence om Regional Da a (Spain,
1860–1930). Social Science His o y44, 169–192, h ps://doi.o g/10.1017/ssh.2019.44 (2020).
32. Díez-Minguela, A., Ma ínez-Gala aga, J. & Ti ado-Fab ega , D. Regional Inequali y in Spain, 1860-2015, (Palg a e Macmillan,
Lond es, 2018).
33. Goe lich, F. J., Maudos, J. & Mollá, S. Dis ibución de La Población y Accesibilidad a Los Se icios En España. (Ins i u o Valenciano de
In es igaciones Económicas (I ie) - Fundación Ramón A eces, 2021).
34. Goe lich, F. J. & Mollá, S. Desequilib ios demog á icos en España: e olución his ó ica y si uación ac ual. P esupues o y Gas o Público
102, 31–54 (2021).
35. Ca e as, A. & Ta unell, X. Es adis icas His o icas de España. Siglo XIX-XX. Fundación BBVA. h ps://doi.o g/10.1017/
CBO9781107415324.004 (2005).
36. Goe lich Gisbe , F. J. Da os climá icos his ó icos pa a las egiones españolas. CRU TS 2.1. In es igaciones de His o ia Economica 8,
29–40 (2012).
37. Cazca o, I., Dua e, R., Ma ín-Re o illo, M., Pinilla, V. & Se ano, A. How Sus ainable is he Inc ease in he Wa e Foo p in o he
Spanish Ag icul u al Sec o ? A P o incial Analysis o he Yea s 1955 and 2005. Sus ainabili y 7, 5094–5119 (2015).
38. MITECO. In eg a ed Municipal Da a Sys em (SIDAMUN). P ep in a h ps://www.mi eco.gob.es/en/ e o-demog a ico/ emas/
analisis-ca og a ia.h ml (2024).
39. Lledó, J. & Pa ía, J. M. A de ailed da abase o sub-annual Spanish demog aphic s a is ics: 2005–2021. Sci Da a 11, 79 (2024).
40. IGN. Base de Da os de Di isiones Adminis a i as de España: Recin os municipales de España = Da abase o Adminis a i e
Di isions o Spain: Municipal p ecinc s o Spain. Las accessed 7-2-2024. h ps://cen odedesca gas.cnig.es/Cen oDesca gas/
ca alogo.do; h ps://cen odedesca gas.cnig.es/Cen oDesca gas/busquedaSe ie.do?codSe ie=LILIM; h ps://da os.gob.es/es/
ca alogo/e00125901-spaignllm (2024).
41. Begue ía, S., Vicen e-Se ano, S. M. & Angulo-Ma ínez, M. A Mul iscala Global D ough Da ase : The SPEIbase: A New G idded
P oduc o he Analysis o D ough Va iabili y and Impac s. Bull Am Me eo ol Soc 91, 1351–1356 (2010).
42. Simps on, J. Spanish Ag icul u e: The Long Sies a, 1765-1965. (Camb idge Uni e si y P ess, 1995).
43. INE. Censo de Población - Popula ion Census. (1950).
44. Monclús F aga, F. J. & Oyón, J. L. His o ia y E olución de La Colonización Ag a ia En España. (Ins i u o de Es udios de Adminis ación
Local: Minis e io de Ag icul u a, Pesca y Alimen ación: Minis e io de Ob as Públicas y U banismo, 1988).
45. Villanue a Pa edes, A. & Leal Maldonado, J. His o ia y E olucion de La Colonizacion Ag a ia En España. La Plani icacion Del
Regadio y Los Pueblos de Colonizacion. (Minis e io de Ag icul u a, Pesca y Alimen acion (MAPA), Mad id (Spain), 1991).
46. AENA. Ae opue os y des inos de AENA = AENA ai po s and des ina ions. Las Accessed 3-3-2025. h ps://www.aena.es/es/
ae olineas/ae opue os-y-des inos/nues os-ae opue os.h ml#ae opue osnacionales (2025).
47. Va gas, E. E ec s o Clima e Change in Medi e anean Wa e Resou ces and Thei Economic Implica ions. (2016).
48. Va gas-Amelin, E. & Pindado, P. The challenge o clima e change in Spain: Wa e esou ces, ag icul u e and land. J Hyd ol (Ams )
518, 243–249 (2014).
49. Es ela, T., Pé ez-Ma in, M. A. & Va gas, E. Impac s o clima e change on wa e esou ces in Spain. Hyd ological Sciences Jou nal 57,
1154–1167 (2012).
50. O iz-Bobea, A. Chap e 76 - The empi ical analysis o clima e change impac s and adap a ion in ag icul u e. in Handbook o
Ag icul u al Economics (eds. Ba e , C. B. & Jus , D. R. B. T.-H. o A. E.) ol. 5, 3981–4073 (Else ie , 2021).
51. Millne , A. & Die z, S. Adap a ion o clima e change and economic g ow h in de eloping coun ies. En i on De Econ 20, 380–406
(2015).
52. Tol, R. S. J. The Economic Impac s o Clima e Change. Re En i on Econ Policy 12, 4–25 (2018).
53. Cas ells-Quin ana, D., Lopez-U ibe, M., del, P. & McDe mo , T. K. J. Adap a ion o clima e change: A e iew h ough a de elopmen
economics lens. Wo ld De 104, 183–196 (2018).
54. Cazca o, I., Ma ín-Re o illo, M., Rod íguez-López, G., Se ano, A. & Sil es e Rod íguez, J. Re aining popula ion wi h wa e ?
I iga ion policies and depopula ion in Spain. ol. 256 P ep in a h ps://hdl.handle.ne /10419/298604 (2024).
55. K ugman, P. The Role o Geog aphy in De elopmen . In Reg Sci Re 22, 142–161 (1999).
56. Head, K. & Maye , T. G a i y, ma ke po en ial and economic de elopmen . J Econ Geog 11, 281–294 (2011).
57. Bas on, D. Fas Ex ac ion om Ras e Da ase s using Polygons. Ve sion 0.10.0. h ps://doi.o g/10.32614/CRAN.package.exac ex ac ,
h ps://isciences.gi lab.io/exac ex ac /, h ps://gi hub.com/isciences/exac ex ac (2023).
58. Ins i u o Geog á ico Nacional. Base de Da os de Di isiones Adminis a i as de España: Recin os municipales de España. P ep in a
h ps://cen odedesca gas.cnig.es/Cen oDesca gas/busquedaSe ie.do?codSe ie=LILIM (2024).
59. QGIS De elopmen Team. QGIS Geog aphic In o ma ion Sys em. P ep in a (2024).
60. Wickham, H., F ançois, R., Hen y, L., Mülle , K. & Vaughan, D. dply : A G amma o Da a Manipula ion. P ep in a h ps://dply .
idy e se.o g (2023).
61. Pebesma, E. Simple Fea u es o R: S anda dized Suppo o Spa ial Vec o Da a. R J 10, 439–446 (2018).
62. Dunning on, D., Vande haeghe, F., Caha, J. & Muenchow, J. R package qgisp ocess: use QGIS p ocessing algo i hms. Ve sion 0.2.0
P ep in a h ps://doi.o g/10.5281/zenodo.10418745 (2023).
63. Ba olomé- od íguez, M. I., Rubio- a as, M. & Sesma-ma ín, D. Wa e o Whom? Un a elling he Alloca ion o Wa e S o age
Capaci y be ween I iga ion and Elec ici y Uses in Spain du ing he 20 h Cen u y. His Ag a 94, 165–201 (2024).
64. INE. In oducción, Tomo III. Volúmenes p o inciales. Censo de 1960 = In oduc ion, Volume III. P o incial olumes. 1960 Census.
Fondo documen al. Ins i u o Nacional de Es adís ica (INE). h ps://www.ine.es/inebaseweb/pd Dispache .do;jsessionid=
CF3C7F2070AA86B3ABC52A83BB5BFBEB.inebaseweb02? d=126937&ex =.pd (2024).
65. Rod íguez-López, G., Se ano González, A., Ma ín-Re o illo, M. & Cazca o, I. HISTORECO: His o ical Spanish T ansi ion
Da abase on Clima e, Geog aphy, and Economics o he 20 h-21s Cen u y. igsha e. Da ase 2. h ps://doi.o g/10.6084/m9.
igsha e.27262032. 2 (2024).
66. Meie , J., Zabel, F. & Mause , W. Global I iga ed A eas. P ep in a h ps://doi.o g/10.1594/PANGAEA.884744 (2018).
67. FAO. AQUASTAT - FAO’s Global In o ma ion Sys em on Wa e and Ag icul u e -> Spain. Las accessed 3-3-2025. h ps://www. ao.
o g/aquas a /en/geospa ial-in o ma ion/global-maps-i iga ed-a eas/i iga ion-by-coun y/coun y/ESP (2025).
68. IGN. Sis ema de In o mación de Ocupación del Suelo de España (SIOSE) [Ca og a ía Digi al]. 1:25.000. P ep in a h ps://www.
siose.es/p esen acion (2011).
20
Scien i ic Da a | (2025) 12:810 | h ps://doi.o g/10.1038/s41597-025-05055-z
www.na u e.com/scien i icda a
www.na u e.com/scien i icda a/
69. Ba ciela, C., Gi áldez, J. & López, I. Sec o ag a io y pesca. in Es adís icas his ó icas de España: siglos XIX - XX (eds. Ca e as, A. &
Ta unell, X. (coo dinado es)) ol. 3, 257–356 (2006).
70. Minis e io de Ag icul u a, P. y A. Anua io de Es adís ica. ol. 1 (Mad id, 2000).
71. Squin u, A. A., an de Sch ie , G., B ugna a, Y. & Klein Tank, A. Homogeniza ion o daily ECA&D empe a u e se ies. In e na ional
jou nal o clima ology 39, 1243–1261 (2018).
acknowledgemen s
Au ho s acknowledge he inancial suppo p o ided by Fundación Ramón A eces (CISP20A6661), o
he Spanish Minis y o Science and Inno a ion esea ch p ojec s (PID2021-123220NB-I00, PID2022-
138886NB-I00, and PID2022-140010OB-I00) and o he Go e nmen o A agon–Eu opean Fund o Economic
and Regional De elopmen ( esea ch p ojec s S55_23R and S40_23R). Also, au ho s wan o acknowledge he
commen s ecei ed by he Edi o and anonymous e iewe s, as well as om pa icipan s a he XLVII Mee ing
o Regional S udies (RER-In e na ional Con e ence on Regional Science), IV In e na ional Con e ence/XVIII
Cong eso de His o ia Ag a ia/XI Encon o Ru al RePo , I Semina o New Academics in Regional Science
(SNAR), X Wo kshop de jó enes in es igado es en Economía y Emp esa and Talle de His o ia Económica o he
Uni e sidad de la República (U uguay).
au ho con ibu ions
Guille mo Rod íguez-López: Da a cu a ion, Fo mal analysis, In es iga ion, Me hodology, So wa e, Visualiza ion,
W i ing – o iginal d a , W i ing – e ised d a . Ana Se ano: Concep ualiza ion, Da a cu a ion, Fo mal
analysis, In es iga ion, Me hodology, W i ing – o iginal d a , W i ing – e ised d a . Miguel Ma ín-Re o illo:
Concep ualiza ion, Da a cu a ion, In es iga ion, Me hodology, W i ing – o iginal d a , W i ing – e ised d a .
Ignacio Cazca o: Concep ualiza ion, Da a cu a ion, Funding acquisi ion, In es iga ion, Me hodology, P ojec
adminis a ion, W i ing – o iginal d a , W i ing – e ised d a .
Compe ing in e es s
The au ho s decla e ha hey ha e no known compe ing inancial in e es s o pe sonal ela ionships ha could
ha e appea ed o in luence he wo k epo ed in his pape .
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/s41597-025-05055-z.
Co espondence and eques s o ma e ials should be add essed o I.C.
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 ea 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