Roman viticulture analysis based on Latin agronomists and the application of a geographic information system in lower Guadalquivir

Vi ual A chaeology Re iew, 7(14): 53-60, 2016 h p://dx.doi.o g/ a .2016.4481 © UPV, SEAV, 2015 Recei ed: Janua y 4, 2016 Accep ed: Ma ch 31, 2016 * Co esponding Au ho : Ped o T ape o Fe nández, ped o. ape o [email protected] DOI: h p://dx.doi.o g/10.4995/ a .2016.4481 53 ROMAN VITICULTURE ANALYSIS BASED ON LATIN AGRONOMISTS AND THE APPLICATION OF A GEOGRAPHIC INFORMATION SYSTEM IN LOWER GUADALQUIVIR ANÁLISIS DE LA VITICULTURA ROMANA A TRAVÉS DE LOS AGRONOMOS LATINOS Y LA APLICACIÓN DE UN SISTEMA DE INFORMACIÓN GEOGRÁFICA EN EL BAJO GUADALQUIVIR Ped o T ape o Fe nández* Semina io Agus ín de Ho ozco, Uni e si y o Cádiz, Facul ad de Filoso ía y Le as, A . Doc o Gómez Ulla, 1, 11003 Cádiz, Spain. ped o. ape o [email protected] Abs ac : The ad ancemen in he his o ical esea ch o business ac i i ies, such as ancien ag icul u e and speci ically Roman i icul u e, equi es he knowledge o he p oduc i e si es. This esea ch will p opose a way o in e connec he in o ma ion con ained in La in ag onomic adi ion, in o de o choose he placemen o a p oduc ion uni , using a geog aphical in o ma ion sys em (GIS). The La in ag onomis in o ma ion, oge he wi h cu en ag icul u al knowledge, may lead o a se o measu able, geo e e encing c i e ia o u u e p ospec ing, such as land o ien a ion o he winds and insola ion, he soil ypes and he slope. Also, he p oximi y o e es ial, lu ial and ma i ime communica ion ou es, as well as ci ies and wa e supply will be assessed. This me hodology is applied in he e i o y o he lowe Guadalqui i (sou hwes Spain) achie ing measu able c i e ia on a geo e e enced map, being able o classi y he da a and c ea e ca og aphy ha shows he bes places o u u e a chaeological p ospec ing. Key wo ds: GIS, lowe Guadalqui i , i icul u e, Columella, Roman ag icul u e. Resumen: El a ance en la in es igación his ó ica de ac i idades económicas, como e a la ag icul u a an igua y más conc e amen e la i icul u a omana, equie e del conocimien o de los luga es p oduc i os. Es e a ículo p opone una mane a de ans e i los conocimien os que apo an los ag ónomos la inos a la ho a de elegi el emplazamien o de una unidad p oduc i a median e un sis ema de in o mación geog á ica (SIG). La in o mación ag onómica la ina, jun o con el conocimien o ag ícola ac ual, puede da luga a una se ie de c i e ios medibles y geo e e enciables pa a u u as p ospecciones, como son la o ien ación del e eno en e a los ien os y la insolación, el ipo de suelo y la pendien e. También se alo a á la ce canía a las ías de comunicación e es e, lu ial y ma í ima, así como a las ciudades y al abas ecimien o de agua. La me odología que se p esen a se aplica en el espacio del bajo Guadalqui i (su oes e de España) consiguiendo medi es os c i e ios sob e un mapa geo e e enciado, pudiendo clasi ica los da os y gene ando una ca og a ía que mues e los mejo es luga es de ca a a unas u u as p ospecciones a queológicas. Palab as cla e: SIG, bajo Guadalqui i , i icul u a, Columela, ag icul u a omana. 1. In oduc ion When in es iga ing Roman ag icul u e, i is necessa y o conside he en i onmen and he e i o y o he chosen si e i s . I is necessa y o ca y ou some p ospec ions i we wan o know whe e he mos p oduc i e places a e. Al e na i ely, nowadays wi h he use o a geog aphic in o ma ion sys em (GIS) such as A cGIS, i is possible o simpli y his p ospec ion wo k. GIS use s a e able o ind he co ec geo e e ence o si es and s udy hei en i onmen , as well as o look o po en ial si es whe e his kind o exploi a ion can be ca ied ou (Goodchild, 2013; Van Joolen, 2003). Fo his pu pose, i is necessa y o ha e he co ec ag icul u al in o ma ion included in ancien sou ces. Fo example, in he La in ag onomis s, we can ind quo es abou some en i onmen al ac o s o conside o an ag icul u al holding. The i s s ep is o choose he s udy a ea. One o he ideas o his a icle is o show ha his can be applied o any loca ion, and he only hing ha would change would be he c i e ion. The ame chosen o his expe imen is he ma shland o he Guadalqui i , whe e in Roman imes, he e was an es ua y. The e was also an impo an Roman colony, Has a Regia (Ma ín-A oyo, 2013), in he speci ic e i o y o he ci y o Je ez (Fig. 1). The Semina io Agus ín de Ho ozco, Uni e si y o Cádiz, has abundan a chaeological and geog aphical in o ma ion abou his a ea o land. TRAPERO, 2016 54 Vi ual A chaeology Re iew, 7(14): 53-60, 2016 Figu e 1: Spa ial s udy a ea: Has a Regia colony. Po us Gadi anus is loca ed in he Spain’s sou hwes . I co esponds o oday’s Pue o de San a Ma ía. The e a e ce ain limi a ions wi hin his s udy, as he eal e i o ia o Has a Regia is unknown. The land o he ancien Guadalqui i ’s es ua y was called Lacus Ligus inus by he Romans. This na igable space allowed hem o each places which nowadays a e no e en nea he coas . Fo he e i o y we conside he limi s o be he p esumed si e o lacus Ligus inus, i.e. he ac ual coas and Guadale e Ri e o he eas (Lagós ena, 2014). Columella’s quo es se he s udy a ea o his e i o y. He is he mos comple e La in ag onomis , who was bo n in Gades (Cádiz). Wi h some o his quo es, i is possible o ex apola e his ancien in o ma ion o a map, especially in he case o ineya ds. This ype o exploi a ion could be signi ican in he e i o y o supplying Rome o he a my, and p obably pa o he annonae sys em. The ineya d ag icul u e in he Ul e io Bae ica p o ince is one o he mos impo an ac i i ies, bu i is no well s udied. O he p oduc s such as oli es o oli e oil a e be e esea ched and in es iga ed. In he Be ic p o ince, he p oblem wi h ineya ds and wine is ha we do no know he p oduc ion loca ions. In he same way, he po e ies iglinae whe e he Be ic wine ampho as we e p oduced (Ca e as, 2001) a e unknown, and his his o ical p oblem will be a p io i y in he u u e esea ch o he colony o Has a (Lagós ena, 2001). To ad ance in he know-how o his wo k, we p opose a numbe o c i e ia o condi ions ha Roman a me s conside ed when choosing a place o a m. I is impo an o emembe ha his a ea was a colony, ounded in he imes o Caesa . The e o e, land would be dis ibu ed and di ided in o cen u iae. The condi ions used o know he po en ial si es mus be based on bo h ancien sou ces and na u al necessi ies. The e a e so many c i e ia ha can be conside ed o his wo k ha we selec ed he land o ien a ion, soil ype, land slope and he p oximi y, as sugges ed by Vii anen (2010). 2. Land o ien a ion 2.1. Winds Land o ien a ion is one o he mos impo an conce ns by Roman ag onomis s o choose he place o plan ing c ops, especially ineya ds (Ca o. Ag i. 8.2. Va o R.R. 1.4.4. Ve g. G. 2.185., G. 2.295.) Nobis in uni e sum p aecipe e op imum isum es u in locis igidis me idiano ine a subician u , epidis o ien i ad e an u ; si amen non in es abun u Aus is Eu isque, elu o ae ma i imae in Bae ica. Sin au em egiones p aedic is en is ue in obnoxiae, melius Aquiloni el Fa onio commi en u . “We belie e i bes o ineya ds o ace sou hwa ds in cold egions, and in wa m egions hey su e om he sou he ly and sou h-eas e ly winds, such as he coas al a eas o Bae ica. I , howe e , he land is subjec o he a o emen ioned winds, i will be be e o hem o ace no he ly o wes e ly winds” (Col. 3.12.6.). Columella in his passage w o e abou he bad winds in he a ea, in he Be ic p o ince and mo e p obably nea Gades. The na u al condi ions such as winds, os s o d ough s we e p oblems o sol e in o de o plan ines. In ou pa icula wo king a ea, he sou h winds a e ho and d y in summe . The eas wind, called ‘le an e’, is simila o he sou h wind only s onge , an a e age o 60- 70 km/h and is e y common o las o a numbe o days. I is necessa y o coun e his condi ion which is why Columella chose his no h-wes o ien a ion unlike in he es o he Roman Empi e. O ien a ion o p o ec om winds could be s udied using some A cGIS ools, because he bes place o g ow a ineya d is a shel e ed place, away and p o ec ed om he eas e n and sou he ly winds, meaning e ain acing no h and wes . Fo ha , i is necessa y o c ea e a digi al e ain model (DTM) in as e o ma , using he in o ma ion in h p://www.jun adeandalucia.es/medioambien e/si e/ edi am/ we can download a .xyz ile, ha mus be con e ed in o a TIN ile wi h he ool “c ea e a TIN” and ans o med o a as e o ma . Wi h his 3D map o eal ele a ions, A cGIS allowed us, h ough he ool in Spa ial Analysis Tools/Su ace/aspec , o gene a e an o ien a ion map wi h ega ds he compass poin s. Then he only condi ion is o selec he op ions, which, in his case a e no h and wes o ien a ion in g een, eas and sou h in ed and an in e media e zone in he land o ien a ed o no h-eas and sou h-wes in yellow (Fig. 2). Figu e 2: Te ain o ien a ion. ROMAN VITICULTURE ANALYSIS BASED ON LATIN AGRONOMISTS AND THE APPLICATION OF A GEOGRAPHIC INFORMATION SYSTEM IN LOWER GUADALQUIVIR Vi ual A chaeology Re iew, 7(14): 53-60, 2016 55 2.2. Insola ion Fu he mo e, he e a e o he condi ions o conside in he case o o ien a ion, he insola ion he e ain-g ound ecei es: Loci po o i ia sun , quae e e ad in e necionem ine a pe ducun , macies e s e ili as e ae, salsa el ama a uligo, p aeceps e p ae up a posi io, nimium opaca e soli a e sa allis… “Fu he mo e, he poo quali ies o a place ha usually b ing des uc ion o ineya ds a e ha shness and ba en soil, sal y o bi e ma shes, s eep, ugged e ain, da k alleys ha a e no exposed o he sun” (Col. 4.22.8.) The places a ec ed by shadow o shade a e bad o ag icul u e. This is logical, as we can p edic ha no h acing moun ain hillside would be mo e exposed o da kness han a hillside acing sou h. The ex says ha he ineya ds ha we e plan ed ei he in a eas con aining plen y o sal y wa e o in sloping a eas would su e . These a e some decisi e plan ing c i e ia ha needed o be aken in o accoun when choosing a loca ion. In ou analyses we shall conside hese e ec s o choosing he a eas. In A cGIS we could use he ool Su ace/hillshade wi h he same as e DTM as be o e. This ool could show he po en ial si es ha could be in da kness. As esul s we could conside ha he opog aphy is no e y s eep, and ha he e a e ew bad places in da kness. In yellow, he e a e he loca ions wi h be e insola ion and in da k colou , he wo se ones (Fig. 3). These h ee c i e ia o he o ien a ion, winds and insola ion, seem o be con adic o y a i s and in ac hey a e, because he bes exposu e o he sun is in he sou he ly di ec ion, bu in his case i could be bad o he winds in he a ea. This con adic ion is only seen in ou pa icula e i o y due o he special condi ions in espec o he land di ec ions e e ed o in Columella (Fig. 4). He knew his p oblem and maybe because o ha , he quo ed his uncle’s echnique, abou co e ing he ines wi h palm lea es in o de o educe he e ec s o he bad wind (Col. 5.5.15.). 3. Soil Types The second c i e ion o be conside ed is he soil ype, p obably he mos impo an one when selec ing a place o g owing c ops. In ou s udy a ea (Fig. 5), nowadays he e is la ge p oduc ion o wine and a la ge ine c op as well. The soils a e good enough, and in he pas his ac was p obably he same. All he ag onomis s w o e abou soils, bu wi hou g ea in e es o speci y which he be e ones a e (Ca o Ag . 1.1. Va o R.R. 1.6.). The case o Columella is meaning ul, no only because he w o e speci ically abou ou esea ch a ea, bu also because he w o e abou his uncle, na u al o Gades, p o iding a lo o in o ma ion abou he ma e in Be ic p o ince. Ma cum Columellam pa uum meum, doc issimum e diligen issimum ag icolam, saepe nume o usu passe memo ia epe o, u sabulosis locis c e am inge e e , c e osis ac nimium densis sabulu, a que... ligone succiden , im op imae s e eo a ionis exhibebi . Succidi au em lupinum sabulosis locis opo e , cum secundum lo em, lub icosis, eum e ium ege i . “Ma cus Columella, a e y lea ned and pains aking a me , equen ly employed: ha is, o heap clay on g a elly g ound, and g a el on g ound ha was clayey and oo s i , and in his way o g ow no only luxu ian c ops o g ain bu also e y ine ineya ds” (Col. 2.15.4.). His uncle used hese wo ypes o soils, bo h logical choices, abou he exchange o his g ound ype, in o de o imp o e he quali y o he land. Speci ically he said ha he e a e wo good ypes o soils in he zone o ineya ds, he i s one is u sabulosis locis c e am inge e e , sandy land whe e hey add loam. We conside ed his, be e han he ansla ion be o e, because sabulosis is sandy a he han heap clay. The o he c e osis ac nimium densis sabulu, which means adding sand o loam soil. These wo op ions, sandy and loam soil a e in gene al bad o c ops, i hey a e alone, bu good i hey a e mixed, as was o en he case in Roman ag icul u e (Sáez, 1995). Those wo bes soils in he ex look simila o wo o he common and be e soils in he p o ince, (Regosoles) “alba izas” and (Lu isoles) “a enas” (Sáez, 1987). Bo h ha e he same cha ac e is ics, i you mix hem oge he , he esul ing soil would be be e o ines, as he p e ious ex men ioned. In a ecen a chaeological su ey, he e a e some ancien holes ha co espond o ines in sand (López and Ruiz, 2007). The o he c i e ion is Columella’s e e ence o ma shland ines: Figu e 4: Bad wind o ien a ion and good insola ion ma ch in mos o he a ea. Figu e 3: Land insola ion s udy. TRAPERO, 2016 56 Vi ual A chaeology Re iew, 7(14): 53-60, 2016 Hac condi u a Columella pa uus meus, inlus is ag icola, u i soli us es in his undis, in quibus palus es ineas habeba … “My pa e nal uncle Columella, he dis inguished ag icul u is , used o employ his me hod o p ese a ion on he a ms whe e he had ineya ds on ma shy g ound…” (Col. 12.21.4.). The adjec i e palus es in ancien sou ces a e associa ed wi h we lands. Vines do no o en wan we soils and o ha eason he was p obably e e ing o a soil cha ac e is ic, a ype o g ound ha could be looded in some mon hs o he yea , such as he e isols “bujeo” (De Ca a, 2010). We could conside ha because Columella spoke in his book abou a p ac ice o ines, i consis ed o making a hole o he plan in o de o main ain he soil mois u e, and ha could be done in no h A ica bu no in a we o ainy a ea, because he excess o wa e could damage he ine (Col. 5.5.4.). The e a e o he soil ypes in he wo king a ea. The lu isols a e common soils nea a i e , o med by allu ial deposi s ha could also be used o plan ines, as long as he ines we e no co e ed by wa e o some mon hs o he yea . Some soils a e no sui able o ines such as li osols, ypically in he moun ains o planosols which a e common in ma shland soils; bo h used o li es ock a ming and o he uses. The in o ma ion on hese kinds o soils and hei cha ac e is ics can be ound in he classi ica ion sys em p oposed by he FAO Soils Po al (2016). The las ype o soils a e cambisols and o he simila ypes, ha a e good enough o e e y ype o c op and he e o e we canno disce n whe he Romans used hem o ines o no . The analysis o his in o ma ion by A cGIS could be done using he se e Spa ial Da a In as uc u e o Andalusia (IDEAndalucia, 2016), whe e i is possible o iew he soil cha ac e is ics o he land. Then he only s ep is o geo e e ence he da a in o a polygon laye and eclassi y he new shape in numbe s, anging om 10 i he soil is good down o 0 i i is bad. 4. Land Slopes Ano he c i e ion o conside is he land slope. The e ain slope is p esen in Va o (Va . R.R. 1.6.5.): P op e haec ia as igia o mae disc imina quaedam iun sa ionum, quod sege es melio es exis iman u esse campes es, ineae collinae, sil ae mon anae “Owing o hese h ee ypes o con igu a ion di e en c ops a e plan ed, g ain being conside ed bes adap ed o he plains, ines o he hills, and o es s o he moun ains”. This summa ises he main idea ha Columella had al eady w i en (Col. 3.1.8.). A medium slope is app op ia e o plan ing ineya ds, and e en oday i is he same. We can assign he plain soil o o he c ops such as whea . A s eep slope would be mo e use ul o li es ock a ming o o he ac i i ies, as he soil is ai ly shallow and s ony. The s udy o his c i e ion could be done by he use o he su ace/slope ool (Fig. 7). I is challenging o know he alue o an a e age slope, because by esea ching he sou ces we do no ha e he measu ing alues o la ge and small slopes. The slope ool shows whe e he e ain makes an ab up change, e en i i is in a dep ession. In he s udy, we could see ha he e i o y is gene ally la , wi h ele a ions in he Cádiz moun ain ange and all he e i o y has plen y o small changes which a e shown in g een. The e a e occasional high slopes in he a ea (Fig. 6). We ha e e iewed he na u al cha ac e is ics o he e ain, despi e he ac he e could be o he s. Ne e heless, hese a e he mos impo an o Columella. In he same way, i is ob ious ha Columella is ep esen a i e and he bes sou ce o lea n whe e he po en ial places o Roman ineya ds we e. 5. P oximi y F om now on he c i e ia will be mo e his o ical, based on human ac o s a he han na u al c i e ia. The ac o s we conside in his sec ion a e summa ised by Ca o (Ca o Ag i. 1.3.): Figu e 5: Soil ypes Figu e 6: Slope o he s udy a ea. ROMAN VITICULTURE ANALYSIS BASED ON LATIN AGRONOMISTS AND THE APPLICATION OF A GEOGRAPHIC INFORMATION SYSTEM IN LOWER GUADALQUIVIR Vi ual A chaeology Re iew, 7(14): 53-60, 2016 57 Si po e is, sub adice mon is sie , in me idiemspec e , loco salub i; ope a ío um copia sie , bonumque aqua ium, oppidum ualidum p ope sie au ma e au amnisqua ñaues ambulan , au uia bona ceíeb isque… “Take ca e o choose a good clima e, no subjec o des uc i e s o ms, and a soil ha is na u ally s ong. I possible, you a m should be a he oo o a moun ain, acing Sou h, in a heal hy si ua ion, whe e labou and ca le can be had, well wa e ed, nea a good sized own, and ei he on he sea o a na igable i e , o else on a good and much equen ed oad”. When he Romans chose a place o build a illa i was impo an o loca e i nea o a ci y, oad o a lu ial poin , such as i e s and seas. Also he ex e e s o he need o a wa e supply and he p oximi y o wo ke s. These wo las condi ions a e less impo an in ou s udy, because he wa e supply o uni iga ed a mland is no necessa y o ines and he manpowe c i e ion is ela ed wi h he p oximi y o a ci y. I is possible o know he p oximi y using A cGIS. The i s hing is o make some changes in he DTM. By eclassi ying he as e we could show classi ica ion abou opog aphy in o ma ion, and by using he ool cos /cos dis ance, i is possible o calcula e he dis ance o a p e iously chosen poin (Lagos ena, 2011; Mignone, 2013). 5.1. Loca ion The i s p oximi y c i e ion is he case o being loca ed nea o a ci y o o he se lemen . I is ele an in his case o know whe e he ci ies o se lemen s a e. We can loca e some la ge ci ies in he e i o y; he mos impo an one is Has a Regia colony. T he e a e o he poin s nea by ha could be ci ies co esponding o indings o complex a chaeological si es, such as “Ce o de las Vacas” o “Gibalbin” (González, 2014). O he s such as Tu is Caepionis o Lux Dubia a e no ci ies, bu a e ele an o ou s udy, because hey a e co es, o example a owe and a eligious sanc ua y (Ga cía e al., 2008). Anyway he knowledge o hese kinds o loca ions is necessa y as a as being nea a illage o ci y is conce ned. In his case Columella old us: Censeo igi u in p opinquo ag um me ca i, quo e equen e dominus enia e equen ius en u um se… “I am o he opinion, he e o e, ha land should be pu chased nea by, so ha he owne may isi i o en and announce ha his isi s will be mo e equen han he eally in ends hem o be…” (Col. 1.2.1) I is some hing signi ican o be nea a ci y, no only o he ma ke oppo uni ies p o ided he e, bu also o good managemen by an owne . This las poin can be a li le con o e sial because o he aim o Columella’s books. He aims o each con empo a ies how o be a good a me (Col. P ae .) Fo ha eason, we could conside whe he he owne needed o be nea a ci y o i ha was he idea ha Columella ied o each, bu was no ul illed by he majo i y (Ma ín, 1971). Anyway, he p oximi y o a ci y is an impo an c i e ion o place a plan a ion, because o he icini y o he ma ke . This would no be especially necessa y o ines bu o he sale o esh g apes and o he esh p oduce o be sold. Fig. 7 shows he cos dis ance o se lemen s. In pa icula , he a eas pain ed in g een a e he close ones o se lemen s. This me hod does no only conside he s aigh -line dis ance bu also he di icul y o mo emen in he opog aphy o he e ain. Figu e 7: Cos dis ance o se lemen s. 5.2. Sea and i e In he same way, he p oximi y o ei he a oad o a wa e way is an impo an c i e ion because anspo a ion was one o he ac i i ies ha would mos a ec he p ice o he inal p oduc , which would be cheape o e sho e dis ances: Quod si o o o una subsc ibi ,ag um habebimus salub i caelo, ube i glaeba, pa e campes i, pa e alia collibus el ad o ien em el ad me idiem molli e de exis; e enisque aliis a que aliis sil es ibus e aspe is, nec p ocul a ma i el na igabili lumine, quo depo a i uc us e pe quod me ces in ehi possin . “Bu i o une a ends ou p aye , we shall ha e a a m in a heal hy clima e, wi h e ile soil, pa ly le el, pa ly hilly wi h a gen le eas e n o sou he n slope; wi h some pa s o he land cul i a ed, and o he pa s wooded and ough; no a om he sea o a wa e way, by which i s p oduc s may be ca ied o and supplies b ough in” (Col. 1.2.3.) Na igable communica ions seem o be signi ican , pa icula ly o expo ing hese p oduc s o o he p o inces. Flu ial and ma i ime communica ions could be a ec ed by he Lacus Ligus inus, he ac ual coas and he Guadale e Ri e (Fig 8). The implemen ed me hodology is he same as epo ed abo e and will be conside ed o he ollowing ones (Sillie es, 2001). To know he po en ial o na igable places, in he analysis we used he econs uc ed es ua y o he Guadalqui i and he Guadale e laye . The i s was done by con ou lines o Doñana and he i e by geo e e encing he hyd og aphy (Ga ala, 1959). TRAPERO, 2016 58 Vi ual A chaeology Re iew, 7(14): 53-60, 2016 Figu e 8: Cos dis ance o sea and i e . 5.3. Roads Rega ding o e land communica ions, Columella ad ises no o be nea an impo an oadway due o he p oblem o c iminals and passe s-by. He said no o be oo close o he oad, bu close enough o p oduc anspo a ion (Fig 9): Haec au em p ae e eun em ia o um popula ionibus e adsiduis de e en ium hospi iis in es a em amilia em. P op e quae censeo eius modi i a e incommoda illamque nec in ia nec a ia p ocul edi io e si u conde e, sic u ons eius ad o ien em aequinoc ialem di ec a si . “Mo eo e , he highway, impai s an es a e h ough he dep eda ions by passing a elle s and he cons an en e ainmen o hose who u n in o lodging. Fo hese easons my ad ice is o a oid he disad an ages o his so and o place he illa nei he on a highway no a om a highway, heighe , and o build in such a way ha i aces he poin whe e he sun ises a he ime o he equinox.” (Col. 1.5.7.) In he spa ial a ea, wo impo an oadways a e known; he i s one is he Via Augus a and he second is men ioned by Ra enna Cosmog aphy (Ra . 317.5) a oad ha wen om Hispalis o Baessipo (Sillie es, 1977). 5.4. Wa e supply The las ac o o conside is he p oximi y o wa e , no speci ically o consump ion o li es ock a ming, bu also o a ming some c ops, plan s like cane, osie and ushes which a e ela ed o he c op o ineya ds. They a e necessa y o his kind o cul i a ion and i hey do no ha e hem, i could inc ease he p ice o he inal p oduc . Quoniam cons i uendis colendisque ineis, quae ideban u u ili e p aecipi posse, disse uimus, pedaminum iugo umque e iminum p ospiciendo um adenda a io es . Haec enim quasi quaedam do es ineis an e p aepa an u . Quibus si de ici u ag icola, causam aciendi ine a non habe , cum omnia, quae sun necessa ia, ex a undum quae enda sin ; nec emp ionis an um. “In as much as we ha e discussed hose ma e s which seemed ha could be augh o ake ad an age o o es ablishing and cul i a ing ineya ds, a me hod mus be se down o he p o ision o p ops, ames, and wi hes. Fo hese a e p epa ed be o ehand, as dow ies, so o speak, o he ineya ds. And i he a me is des i u e o hese, he has no eason o making ineya ds, since e e y hing ha is needed will ha e o be sough ou side he a m; and, jus as A icus says, no only does he cos o pu chase pu a bu den upon he accoun s o he o e see , bu also he p ocu ing o hem is a e y g ea annoyance.” (Col. 4.30.1.) I is ocussed on ineya ds because o he ma e ial equi ed o cul i a e hem. As al eady men ioned, Romans had a ious ypes o ines ha all needed wooden ma e ials o suppo hem. Mos o he wood- p oducing plan s we e needed o suppo ing s icks and we e ound in looded a eas, such as i e s, lakes, s eams and ma shland. A coas al a ea could ha e supplied hose kinds o needed ma e ials. The e o e, he con enien a ea o si ing a ineya d would be nea o a coas line. Fo his map we use he hyd og aphical in o ma ion o Andalusia o c ea e a laye wi h s eams, lakes, i e s and he p esump ion o he ancien es ua y o Guadalqui i . Figu e 10: P oximi y o s eams, i e s and sea. Figu e 9: Cos dis ance o oads. ROMAN VITICULTURE ANALYSIS BASED ON LATIN AGRONOMISTS AND THE APPLICATION OF A GEOGRAPHIC INFORMATION SYSTEM IN LOWER GUADALQUIVIR Vi ual A chaeology Re iew, 7(14): 53-60, 2016 59 6. Discussion The condi ions we a e discussing he e, a e no he only ones ha could be conside ed, bu hey a e he mos impo an ones o his s udy. Uni ying he condi ions on a map could help o loca e and isualize he po en ial places suscep ible o be su icien ly sui able o ineya ds. The e a e eigh condi ions (c i e ia) ha we ha e conside ed, wi h he associa ed geog aphical map. Using he ool weigh ed o e lay can combine all in o a single map ha shows he bes and wo s places o ineya ds (Fig. 11). These esul s could help o loca e a illa, no only on he main la ge map, bu also in a pa icula place. I his model we e used in a smalle a ea, i could con ibu e o p o iding eliable esul s. This model ies o be p edic i e, looking o po en ial places o be p ospe ous. The eigh c i e ia o he s udy a e weigh ed wi h he same alue, 12.5%, adding up o a o al o 100%. We ga e hem he same impo ance because none o hem a e comple ely de e mined. We use he same weigh ing o each c i e ion, because he mos e icien way is o balance hei alues in o de o disc imina e none o hem and because he in o ma ion o ancien sou ces do no gi e us he impo ance ha ag onomis s conside o e e y case. Tha could be he subjec o a u u e s udy. The esul s shown indica e some a eas ha we e bad o ineya ds. I is wo h no icing ha he s udy could be ca ied ou wi h o he pa ame e s, such as wa e consump ion necessa y o he illae (bu no speci ically o he ineya d). I we compa e he esul s wi h he cu en ag icul u al land use, he b own colo on he map is he cu en c op o ines and he g een a eas deno e he bes places acco ding o he esea ch (Fig. 12). We can obse e ha he e a e some a eas whe e he condi ions a e in he same loca ions as c ops oday whe eas o he s a e no . In he la e case we mus conside why. Fi s ly, mos o he b own a eas o cu en ineya ds a e in a ligh g een o yellow, which a e no bad o cul i a ion wi h he pa ame e s o he Roman ag onomis s. The second is ha he logic o he loca ions a e no he same, because oday he impo ance o i icul u e is in Je ez and he majo i y o he cul i a ion is nea his ci y. The o he poin s a e nea Sanlúca , Pue o de San a Ma ía and T ebujena, hese ci ies ha ing a signi ican ineya d cul u e oo. 7. Conclusion The use o his his o ical in o ma ion o c ea ing a map using a GIS, allows us o es ima e he po en ial places o ineya ds. Howe e , i could be unde aken no only o ines, bu also o addi ional economic ac i i ies such as li es ock a ming, oli e oil p oduc ion and maybe mining and ishing. Wi h o he pa ame e s, i is possible o econs uc ca og aphy using condi ions adap ed o he s udy a ea designed o esea ch and in es iga ion. Fo example, in o he places, wind p oblems should be di e en . The main idea o his wo k was no o each a o mal conclusion ha closes his issue, bu o ind whe e he a eas would be loca ed. The main pu pose was o show he applica ion o GIS o es ima e land loca ion in o de o ca y ou u u e p ospec ion. The nea es egion o Has a Regia, ac ually no h o Je ez, in he weigh ed map was a good esul , p obably because in ancien imes wine p oduc ion would be close o he wa e , whe e he p oduc s migh ha e been expo ed, due o cheape wa e way anspo a ion and he p oximi y o a big ci y such as he Has a Regia colony, o Gades (Ma ín-A oyo, 2013). The ideal s udy a eas (highligh ed in g een in Fig. 9) a e loca ed in he Je ez a ea called “Ma co del Je ez”, which is cu en ly an impo an wine-g owing egion. As i is demons a ed he ein, i seems o be ha cul i a ing ineya ds was also impo an in he Roman age. Figu e 11: Final weigh ed map. G een: be e places o ineya ds. Figu e 12: Resul s in he cu en land use. TRAPERO, 2016 60 Vi ual A chaeology Re iew, 7(14): 53-60, 2016 Re e ences De Ca a Ga cía, J. A. (2010). Ca ac e ís icas ag oclimá icas de la id. Agencia Es a al de Me eo ología, Se icio de Aplicaciones Ag ícolas e Hid ológicas. Re ie ed 3 Janua y 2016, om h p://www.di ulgame eo.es/uploads/Ca ac e %C3%ADs icas- id.pd Ca e as Mon o , C. (2001). P oducción de Hal e n 70 y D essel 7-11 en las inmediaciones del Lacus Ligus inus (Las Ma ismas, Bajo Guadalqui i ). In Ac as del Cong eso In e nacional, Ex Bae ica Ampho ae. Conse as, acei e y ino de la Bé ica en el Impe io Romano. Écija: 419-426. Columella (1941). De e us ica, (H.B. ASH T ans.). On ag icul u e de e us ica, Vol. 1, books I-IV. Ca o & Va o (1967). On Ag icul u e. (W.D. Hoope T ans.). Loeb Classical Lib a y. FAO Soils Po al. (2016). Food and Ag icul u e O ganiza ion o he Uni ed Na ions. Fao.o g. Re ie ed 3 May 2016, om h p://www. ao.o g/soils-po al. Ga cía Va gas, E., Fe e Albelda, E. & Ga cía Fe nández, F.J. (2008). La omanización del bajo Guadalqui i : ciudad, e i o io y economía (Siglos II-I a.C.). Mainake, 30, 247-270. Ga ala, J. (1959). Geología de la cos a y Bahía de Cádiz. Cádiz: Dipu ación P o incial de Cádiz. González Fe nández, J. (2014). Insc ipción omana del yacimien o de Gibalbín (Cádiz) con indicación de su condición de municipio. SPAL, 23, 191-196. Doi:10.12795/spal.2014.i23.11. Goodchild, H. (2013). GIS models o Roman ag icul u al p oduc ion. In A. Bowman & A. Wilson (Eds.), The Roman ag icul u al economy: o ganiza ion, in es men , and p oduc ion, 55-86. h p://doi.o g/10.1093/acp o :oso/9780199665723.001.0001 IDEAndalucia (2016). Ideandalucia.es. Re ie ed 3 May 2016, om h p://ideandalucia.es. Lágos ena Ba ios, L. (2001). La p oducción de salsas y conse as de pescado en la Hispania omana (II a.C.-VI d.C). Ba celona. Colección Ins umen a 11. Lagós ena Ba ios, L. (2007). Columela, de e us ica VIII, 16-17: una uen e pa a el conocimien o de la piscicul u a en Bae ica. Ce a iae, 109-115. Lagós ena Ba ios, L., Zule a Alejand o, F.B., Cas o Ga cía, M., Bas os Za andie a, A. & Tala e a Cos a J. (2011). In aes uc u as hid áulicas en Bae ica: p opues a pa a el es udio del azado de sus acueduc os. In V. Mayo al He e a & S. Celes ino Pé ez (Eds.), Tecnologías de in o mación geog á ica y análisis a queológico del e i o io, (pp. 271-279) Mé ida. Lagós ena Ba ios, L. (2014). La pe cepción de la ibe a en la cos a a lán ica de la p o incia Hispania Ul e io Bae ica. El lacusLigus inus. In E. He mon & A. Wa ele (Eds.), Ripa ia, un pa imoine cul u el. La ges ión in èg èe des bo ds de l’eau, (pp. 187-197). López Amado , J.J. & Ruiz Gil, J.A. (2007). A queología de la id y el ino en El Pue o de San a Ma ía. Re is a de His o ia del Pue o, 38. 11-36. Ma ín R. (1971). Reche ches su les ag onomes la ins e leu s concep ions économiques e sociales. Socie éd’édi ion les belles le es. Pa is. Ma ín-A oyo Sánchez, D.J. (2013). P oyección e i o ial y colonización omana. El caso de Has a Regia. Cádiz. Mignone, P. (2013). El camino hacia el San ua io Incaico del llullaillaco: el complejo a queológico más al o del mundo analizado a a és de un SIG. Vi ual A queology Re iew 4. 145-154. h p://doi.o g/10.4995/ a .2013.4356 Sáez Fe nández, P. (1987). Ag icul u a omana de la Bé ica. Se illa, 5-15. Sáez Fe nández, P. (1995). El ino en la Bé ica omana. In J.J. Iglesias Rod íguez, (Ed.). His o ia y cul u a del ino en Andalucía. 125. Se illa. Sánchez López, E & Gonzalbes C a io o, E. (2012). Los usos del agua en la Hispania omana. Vínculos de His o ia, 1, 14-19. Sillie es, P. (1977). P ospec ions le long de la Via Augus a. Habis, 8, 331-344. Sillie es, P. (2001). Voies d’eau e esso economique de l’Hispanie, Vías ma í imas, lu iales y desa ollo económico de Hispania. Zephy us 53-54. 433-442. Su e , J.G. & Palacios y Guillen M. (1867). Plano Gene al de los e enos inícolas de je ez. Pue o de San a Ma ía y Sanluca de Ba ameda. Van Joolen, E. (2003). A chaeological land e alua ion. A econs uc ion o he sui abili y o ancien landscapes o a ious land uses in I aly ocused on he i s millennium BC. G oninguen. Vii anen, E.M. (2010). Locus Bonus, he ela ionship o he oman illa o i s en i onmen in he icini y o Rome. Helsinki.