Introduction

Since prehistory, in the period of continuous valorisation of the Adriatic Sea and its coast, navigation was predominantly preformed using ships powered by sails and oars. Wind, as a major ship driving force prior to invention of steam engines, and later, internal combustion engines, is not always favourable for sailing. Those differences sometimes manifest as upwind, but the most undesirable situations occur during stormy winds. Such wind can not only fracture ships masts and rip sails apart, but the waves it generates can roll over and sink the ship, or push it to the shallow waters and the coast, which can result in human casualties and material losses. It is certainly the cause of numerous shipwrecks later discovered as hydro-archaeological findings (Brusić, 1970; Gluščević, 1994). Moreover, antique and medieval vessels were smaller in size, compared to merchant ships of today, and because of their immense dependence on wind, their maneuvering abilities were limited. Therefore it is logical to presume that such form of navigation needed to be organized in relation to weather conditions and coast configuration. In other words, a careful selection of appropriate navigation time schedules and routes in order to avoid potentially dangerous scenarios, particularly in the open sea, had to be performed.

Fortunately, valuable historical writings about wind and sailing across the Adriatic Sea, especially across its eastern part, do exist, but according to times of their origin, it is necessary to keep in mind that data they contain is gathered by observation, and not by systematic measurements. In addition, until medieval times, and especially until mid Modern Ages, this area was not charted properly – it was perceived from “the eye of beholder” of mariners who were sailing across it. To understand this further, information contained in those writings needed to be compared with measured data¹. The prime hypothesis here is that weather conditions in last two millennia have not been significantly changed in terms of wind flows. That is, their spatial and temporal occurrences and intensity are for most part similar, if not equal, to today’s values. To prove such ideas, wind data for 18 wind monitoring stations (main meteorological and climatological stations) was collected. The main selection criteria was that their layout covers as much of the research area as possible, and that they are distributed as evenly as possible. For the purpose of this paper, the Croatian part of the Adriatic was divided into three parts: northern, central and southern (Figure 1). The collected wind data was provided by Croatian Meteorological and Hydrological Service (DHMZ) and corresponds to thirty-year climate cycle from 1981 to 2011. The data was categorized and classified in the following order: for each station, the wind flow was monitored across the range of 16 directions, and wind flow parameters considered were its frequencies (measured across 12 grades of Beaufort scale and frequency of silence), average velocities and maximum velocities as a component which suggests a possibility of stormy winds and high waves. Those parameters were also measured to get annual average values, and also during the four climatological seasons² to determine seasonal differences in wind directions and intensities. Since the weather stations, when compared to the size of study area, can be cartographically observed as point symbols, each of them was assigned given its buffer impact zone approximately via method of Voronoi Diagrams (in geography better known as Thiessen Polygons) – parts of the Adriatic which are considered to have a direct impact on the measuring results for each station (Figure 1) (URL 1). It was also considered that each station was built on the appropriate location and that there are no obstacles that interfere with wind flow towards stations. Information gathered by analysing those data was then compared to historical writings about sailing to determine the scale of deviation from real i.e. estimated situation and if there were any advances in the description of wind flow and navigation over the years. As a cartographic basis for thematic maps presented here, the data of State Geodetic Directorate (DGU) (Digitalni atlas Republike Hrvatske – DARH) was used together with SRTM 90 m digital elevation model (DMR). The maps are projected in HTRS96/TM reference system. All figures are author’s work, created upon literature data and data received from DHMZ (thematic maps – Fig. 1- 4, Fig. 7) or by sketching interpreted text information contained in several different paragraphs in literature, made for reader’s easier understanding of paragraphs in this paper (schemes showing current course and wind relations for square-rigged ships – Fig. 5 and 6).

Research on correlations between weather conditions and sailing across the Adriatic was carried outby historians, archaeologists and climatologists – mostly in form of articles (Brusić 1970; Gluščević 1994; Bilić, 2012), but also in form of doctoral dissertations (Šešelj, 2009). Undoubtedly, the biggest contribution was made by Mithad Kozličić and his associates (J. Faričić, S. Uglešić), whose bibliography consists of numerous valuable works related to history, cartography and history of navigation (1990; 1995; 1997; 2006). One thing that Kozličić’s works somewhat lack is a more precise wind analysis for area of the Adriatic, which is, although in basic terms, described in climatological and meteorological literature (Šegota, Filipčić, 1996; Penzar et al., 2001; Vučetić, Vučetić, 2002; Gelo, 2010). The wind in those works about climatology is predominantly presented via description method. M. and V. Vučetić focus their work on the description of winds and sea conditions important for modern nautical sailing, while in the work of B. Gelo, there are valuable chapters about analysis of ground air circulation. There are also articles, mainly written by foreign authors, focused on computer modelling of wind and wind-generated waves across the Adriatic, often based on case studies (Heimann, 2001; Bertotti, Cavaleri, 2009; Katalinić et al., 2015). The study area for the majority of articles that describe wind behaviour more precisely is mostly local. One of the articles that directly connect wind and its impact on sailing is Smjer vjetra – jedan od problema u antičkoj plovidbi Jadranom (Bilić, 2012), based upon article Problemi plovidbe Jadranom u predhistoriji i antici, written by Z. Brusić (1970), which does not solve problems of navigation issues completely, although it offers some solutions about how was sailing organized when sailing outwards (the Adriatic). In order to solve that issue more precisely, works that describe influences of introducing navigational technologies such as rudder, keel and lateen sail, with the aim to optimize navigation, were consulted (Champbell, 1995, Castro et al., 2008).

Prevailing winds across the Croatian part of the Adriatic Sea

According to its location, area and shape, the Adriatic Sea is considered an enclosed sea, without direct and strong connection to any of the oceans, particularly when considering that the Mediterranean Sea (whose part is the Adriatic Sea), is connected to the Atlantic and Indian Ocean via two narrow passages (Gibraltar Strait and Suez Canal) (Riđanović, 2002; Magaš, 2013). Across the area of the Adriatic, the influence of global air circulation is not so strong³, and the winds are, for the most part, generated by secondary and tertiary air circulations, caused primarily by thermodynamic air characteristics – air masses and pressure systems (cyclones and anticyclones) that cause bora and jugo winds, and local temperature oscillations of sea and coastal land that cause sea-breeze and land-breeze winds (Filipčić, Šegota, 1996; Vučetić, Vučetić, 2002). Across the Croatian part of the Adriatic, winds flow from every direction of roman (8-point) wind rose⁴, but prevailing winds are jugo, bora and mistral. Jugo⁵ and bora flow most intensively and are able to generate highest waves, while mistral is a wind that flows from NW direction with moderate intensity and is not able to generate threatening sea conditions⁶. Sea and land breezes are also very occasional winds, and neither are able to generate high waves (Gelo, 2010). Other winds that can cause threatening sea conditions (stormy winds⁷) are the eastern wind, so called Levanat, and the southwestern wind, so called Garbin (Libbico). Garbin usually does not develop stormy velocities, except in cases when the eye of the cyclone is located northward and the ridge of Azores high pressure is located southward or southeastward of the Adriatic (Peljar I., Jadransko more – istočna obala, 1999; Brzović, 2001).

Northern Adriatic, mostly Velebit Channel and Kvarner archipelago, are the areas of immense frequencies of bora wind (almost 400 ‰ around Senj and Karlobag), mostly during the cold part of the year, while along the western coast of Istria those values are significantly lower. Bora also reaches its highest average velocities across this part of the Adriatic, range between 5 m/s⁸ across the area of Rijeka and Kvarner archipelago edge and almost 12 m/s in the coastal area at the foot of Velebit. Highest maximum velocities of bora were recorded around Senj (over 32 m/s). The direction of bora flow is in close relation to the surrounding coastal relief longitudinal axis (Figure 2). For example, bora across the western Istrian coast flows from NE direction, while across the area of Rijeka Bay, and further southwards towards Lošinj, N and NNE directions are predominant. In the coastal area at the foot of Velebit, in accordance with curvature of coast line and morphology of neighbouring mountain range, bora flows from ENE⁹ direction across the area of Senj, while southwards, across the area of Karlobag, the flow direction of bora is NNE and NE. Jugo is a less frequent wind across the North Adriatic, particularly in the area at the foot of Velebit, where numerous islands like Pag and Rab block the fetch from southern open sea. Part of the North Adria which is exposed most to jugo events is the southern open sea stretch of Kvarner archipelago, with its S flow direction, and the southern and western part of Istrian coast with SE flow direction (although there are rare occasions of strong breaches from S direction, with frequencies of only 9 ‰ and maximum velocities of 18.5 m/s). This part of Istria is also affected by mistral and the southwestern flows – both winds develop their frequencies and velocities during summer, while maximum velocities of the southwestern wind are present during winter (47 ‰ and 22.7 m/s).

The morphology of coastal relief across the area of Central Adria differs greatly when compared to its northern part. Besides having uninterrupted relief extension axis (the whole relief streches alongside northwest-southeast axis), hinterland relief (Bukovica and Ravni Kotari) is much more flattened. In other words, there is no tight neighbouring of mountain ranges and coastal area. Measured wind parameters are also in close relation with morphology – intensity of bora is much lower across this area, with frequencies usually lower than 200 ‰ and average velocities of about 4 m/s. Moreover, NE direction of bora is merely equal across whole Central Adriatic, with an exception of farthest northeastern part of Zadar archipelago (Silba island group) (Faričić, 2012) where, due to more pronounced influence of Velebit mountain range, the direction of bora flow is more ENE. The direction of jugo across this whole part of Adriatic is SE, but across the parts closer to the open sea there are rare occasions of S direction wind (about 80 ‰), but with higher velocities than they are in cases of SE direction. Central Adriatic is the area of most intense and most equalized flow of mistral (NW flow direction) (Figure 7) across the whole Adriatic, with its highest frequencies along Zadar and Pašman channels and highest velocities on its most northwestern part, which is more exposed to open sea conditions.

South Adriatic is a part of the Republic of Croatia with the biggest contribution of territorial sea within its total area, mostly because of vast open sea areas around Vis and Lastovo islands with the surrounding smaller islands and small islands like Jabuka, Svetac and Palagruža. The area of South Adriatic, and especially its southernmost part, is part of the Adriatic with highest wind exposures (Lastovo station measured 10.8 ‰ of calmness, Komiža station 9 ‰ and Palagruža station only 7.7 ‰), primarily due to jugo conditions in winter and spring and mistral conditions in summer. Prevailing jugo direction across this part of Adriatic is SE (Figure 3), and is conditioned by the surrounding relief, whereas across southernmost open sea area the prevailing direction is S¹⁰, and across narrow Split area the direction is ESE, channeled by Brač channel axis. The average jugo velocities across the open sea are within range of 6 to 8 m/s while maximum velocities reach even 32.7 m/s. Bora is also a very frequent wind there, specially across parts of the sea that are closer to the shore. For example, coastal stations measured frequency of bora from 240 ‰ (Ploče) to 308 ‰ (Split), while in the open sea it rarely exceeds 150 ‰. The prevailing bora direction in coastal part is NNE, while in the open sea, due to lower influence of island relief, bora flow is channelled into N (Komiža, Palagruža, Goveđari) and NE directions (also Komiža and Goveđari). Alongside coastal area, besides relatively high frequencies, with its average velocities from 4 to 6 m/s, while maximum velocities reach about 27 m/s, the velocities of bora are lower than across the coastal area at the foot of Velebit. Across South Adriatic open sea the frequencies of bora are much lower, but having in mind that its average and maximum velocities are similar to those alongside the coast, it is obvious that this part of Adriatic is also exposed to stormy wind breaches from northeast, especially during winter and spring.

Organization of sailing ship navigation across the Croatian part of the Adriatic due to borga, jugo and mistral

Prehistory and Antique Sailing

The coastal part of Croatian Adriatic has been populated since prehistory (Brusić, 1970; Stražičić, 1989), and the earliest archaeological findings of ships from this area were dated up to third millennia B.C. Generally speaking, those findings are remains of dugout canoes and rafts, but a ceramic drawing of square-rigged ship, found within Grapčeva cave, Hvar Island, points to the earliest usage of vessels rigged with sails across the Adriatic between 2500 and 2400 B.C. (Brusić, 1970).¹¹ There are no preserved writings about navigation routes from that area, but it is possible to reconstruct them (at least partially) by studying locations of prehistoric shipwrecks. Based upon archaeological findings from Neolithic and later ages, and guided by the hypothesis that square-rigged ships were able to sail only downwind, Brusić alleged some of fundamental navigation routes across Adriatic – longitudinal and transversal.¹² He concluded that, during prehistory, longitudinal directions stretched mostly alongside the eastern Adriatic coast (Figure 7) (Brusić, 1970). That statement is supported by the fact that the eastern coast is full of numerous longitudinally pointed channels between islands and between islands and the coast. Winds that could safely power vessels with lower maneuvering abilities are land-breeze and mild jugo in the direction of sailing in, and mistral and sea-breeze in direction of sailing out of the Adriatic Sea. The earliest writings about navigation across this area originate from Ancient Greece, and describe sailing mostly as a long, scout-type navigation, and for example, Theopompos wrote that sailing across the whole Adriatic lasted for 6 days.¹³ The Antiquity is also the age of first proto-pilots like Pseudo Scylax’s Periplus and Pseudo Scymnus’s Periegesis, which contain information that sailing across the Adriatic was organized during daytime and nighttime, and that numerous islands along the eastern Adriatic made sailing much easier to perform (Brusić, 1970; Kozličić, 1990; Gluščević, 1994). According to M. Kozličić (1990), whose interpretations followed the works of Croatian historian Ivan Lučić (1604 – 1679), during the Antiquity there were three types of navigation routes, depending on the ship size; interadriatic route that stretched across centre of the Adriatic and was used for the largest vessels, routes between outer and central islands used for mid-sized vessels, and routes between islands and coast used for the smallest of vessels (Figure 7). This route order was common in situations when the weather conditions enabled sailing across the open seas (i.e. in summer), while during the remaining part of the year, vessels of all sizes sailed in close proximity to the coast whenever possible. The description of eastern Adriatic island range as an area safe for navigation and synchronizing routes with ship size also point to the fact that Ancient Greeks were aware of local stormy winds and wavy sea conditions and organized navigation accordingly. In his book Historijska geografija istočnog Jadrana u starom vijeku (1990) Kozličić also included a map of antique shipwrecks, which, when compared to bora average velocities (i.e. average annual values), shows that many of those shipwrecks took place in the areas prone to higher bora average velocities (areas with higher probability of excessive bora velocities). Such areas are the coast at the foot of Velebit, Kvarner archipelago, Šibenik littoral area and cape Ploča (Punta Planka) – a place with frequent alternations between bora and jugo in relatively short periods (Šešelj, 2009) (Figure 4). Although neither wind nor other natural factors (i.e. stranding on a reef) could necessary be the main or the only cause of shipwrecks, there is a possibility that some the mapped shipwrecks are related to sudden bora breaches. Bora is a wind that occurs suddenly and generates extremely unfavourable sailing conditions, especially for vessels with lower maneuvering abilities.¹⁴

There were doubts about how sophisticated the maneuvering abilities of antique vessels in context of harvesting wind as a power resource were. Assumptions stretch from those that they could sail exclusively downwind (Bilić, 2012, according to Brusić, 1970), to the idea that the range of available sail directions for certain wind flow extended for about 100° port and 100° starboard, which sums up to 9 points of compass¹⁵ (Champbell, 1995; Bilić, 2012). In other words, the ancient square-rigged ships could not sail upwind, and on courses which relatively differ from the opposing wind for less than 7 points of compass. For example, in a scenario of jugo flowing from SE (135°), an antique ship could not sail on course larger than 56.25° (~ENE) and smaller than 213.75° (~SSW) – it could not, for example, sail towards east (90°) or south (180°) (Figure 5) (Champbell, 1995; Bilić, 2012)

Studies of antique writings about the Adriatic, such as: Pseudo Scylax’s Periplus, Pseudo Scymnus’s Periegesis and Apollonius of Rhodes’s Argonautica, showed that eastern Adriatic was described in a way as if it were sailed across in southeast direction. It leads to a conclusion that such direction was, in fact, the prevailing antique sailing route. Moreover, during those times, summer was the period of most intensive sailings – March, April, May and October were considered as months dangerous for safe navigation, while during the winter months sailings usually were not performed (Šešelj, 2009; Bilić, 2012). If we consider summer wind conditions across the eastern Adriatic with mistral as a power resource, and compare it to the antique order of coast description (outwards sailing direction), it is evident that they overlap for most part, which means that they harvested mistral as a propulsive wind for southward and southeastward sailings. According to T. Bilić (2012), directions of sailing inwards took place along both longitudinal Adriatic coasts, harvesting favourable winds, and that during the periods of calmness, oars were used as a propulsion devices. Favourable winds that flow northwards could be jugo in spring (and also in summer, but less often) and sea- and land-breezes that could be harvested as side winds flowing from port and starboard direction during coastal navigation. Later, during medieval times, smaller vessels were rigged with lateen sails. Since large scale lateen sail is difficult to adjust quickly; larger vessels had square-rigged main masts and lateen-rigged mizzen-masts (Castro et al., 2008). It is assumed that the addition of lateen-rigged mizzen-mast at ship stern extended the range of harvestable winds for about a half of point of compass on each side (with usage of 16-point wind rose, and for about one point each side for a 32-point wind rose) (Champbell, 1995). If, in any case, navigation needed to be done on a course that is opposite (or nearly opposite) to the wind, then the route was not straight-lined. The course needs to be constantly alternated clockwise and counterclockwise, usually by using one of two techniques – a wavy route called tacking or an opposing loop route called wearing, so that the course is plotted at angles that allows upwind to hit ship’s sides (Figure 6) (URL 5).

Medieval and Modern Age routes used for sailings for religious purposes

One of the most reliable and most precise written sources of sailing across the eastern Adriatic, at least longitudinal sailing, that took place during medieval and modern times are pilgrimage travels that thoroughly describe sailing from Europe to the Holy Land. There are also writings that describe voyages of two very important Christian persons across the Adriatic – Pope Alexander III in 1177 and Francis of Assisi in 1212 and 1219 – 1920 (Faričić, 2014; 2015). Undoubtedly, such endeavours required organization on the highest level which had to be planned and executed in accordance with local weather conditions, including wind. So, it is presumable that such sailing routes in some part reflect past climate conditions, obviously favourable in terms of sailing along coasts. Late Middle Ages are the times of technological advancements in navigation across the Mediterranean, the Adriatic included. During the thirteenth century, rudder has been introduced, and invention and application of simple compass in the twelfth century and dry compass in the fourteenth century present major technological leaps forward. Additionally, the thirteenth century and the fourteenth century are the times when first nautical charts that showing Adriatic area were produced (Faričić, 2014; 2015).

Pilgrimage sailing endeavours, whose target was the easternmost part of Mediterranean, the so called Levant, usually started from Venice, mostly because, by that time, Venice had made intense connections with the Mediterranean and the central Europe, and also had the greatest part of sailing routes towards Levant under its control (Pavić, 2007; Faričić, 2014). Sailings that started from Venice would begin during spring or summer months (from April till August), and its segment that included sailing across Adriatic lasted from 8 to 20 days, depending on weather conditions, chosen route, number of shore visits and time spent within ports. Such type of navigation was usually organized alongside Adriatic coastal longitudinal routes. After sailing off from Venice, navigation would continue along western coast of Istria (visits to ports of Poreč, Rovinj or Pula were often included). After leaving the coast of Istria, one of the routes, the so called, outer branch (Faričić, 2014; 2015), continued over the open sea, usually without stops, along outer island belt towards Strait of Otranto, the Island of Corfu and Ionian Sea (although, there was an occasional stop at the port of Dubrovnik). Another possibility, the so-called, inner branch (Faričić, 2014; 2015) was navigation through open sea border of Kvarner archipelago between Lošinj and Unije islands. After that, the course was altered towards the coast through Sedmovraće¹⁶, in order to reach towns like Zadar, Šibenik and Split, and later, through Splitska vrata, alongside islands Hvar and Korčula islands the sailing was directed towards Dubrovnik. One of the routes included sailing from western Kvarner archipelago towards Jabuka, Sv. Andrija, Vis and Palagruža islands (Kozličić, 1997; Pavić, 2007). It is worth mentioning that not all pilgrimage routes had Venice as a starting point – transversal sailing routes also existed in order to connect western and eastern Adriatic coasts (i.e. ports like Ancona, Pesaro, Pescara and Bari with ports of Zadar, Šibenik, Split and Dubrovnik; Faričić, 2014). Afterwards, navigation would usually continue alongside eastern Adriatic coast (Figure 7).

When longitudinal pilgrimage routes and seasons during which those routes were used are compared to the direction and intensity of wind across the eastern Adriatic, it is noticeable that navigation was organized during months with mild and predictable weather conditions (with lower storm probabilities) and with longer daylight to optimize navigation using terrestrial objects (Faričić, 2014; 2015). Besides that, the prevailing northwestern wind flow occurs during summer across the eastern Adriatic. Those wind flows are known as etesian and sea-breeze, which are very favourable for longitudinal sailings in southeast direction, especially across the Central Adriatic, with their frequency somewhat higher across the open sea. The Adriatic open sea is also a region of high wind frequencies and velocities all year round, so the same situation occurs in summer in cases of mistral as the predominant wind, making it very suitable for propelling large vessels. It is noticeable that sailing routes that took place during late medieval times and early modern period do not deviate much from the prehistoric and antique routes, except for outer longitudinal routes that stretch across centre of the Adriatic due to the usage of large vessels with better maneuvering abilities and more sophisticated navigational equipment.

However, not all pilgrimage sailings were carried out without difficulties arising from the wind conditions, which, according to written sources, mostly occurred due to sudden bora breaches or due to jugo that would strengthen from the opposite direction. For example, Sir Richard Guylforde, during his voyage to the Holy Land 1506 – 1507, wrote that unfavourable wind occurred during their sailing from Vis to Hvar (Kozlićić, 1997), which was obviously a bora event, while Sir Richard Torkington on his voyage during June of 1517 mentioned “opposing wind along coast of Istria” (Kozličić, 1997), which can be considered as a jugo event. John Locke, during his voyage to the Holy Land 1585 described Palagruža as a “dangerous island” (Kozličić, 1997), which is not surprising because of immense annual wind frequencies and velocities across island and its surrounding area.

Earliest modern pilots

Writings that describe navigation along coasts and islands of the eastern Adriatic, which were not influenced by religion, are portolan charts and isolarios¹⁷, created by numerous authors from medieval times and modern period, for example, Benedetto Bordone, Piri Reis, or Giovanni Francesco Cammoccio (Kozličić, 1995; Novak et al., 2005). One of the most important works that describe navigation routes across the Adriatic was the isolario Viaggio da Venetia a Constantinopoli (Voyage from Venice to Constantinople) written by Giuseppe Rosaccio in 1598, and Senj Pilot from 1639 (Kozličić et al., 2012). The first written source about how navigation is affected by wind comes from Dubrovnik author Benedikt Kotruljević, entitled O plovidbi (About Navigation) from 1464 (Kozličić, 2006), but the earliest detailed scientific contribution to understanding wind conditions across the Adriatic was made by a French hydrographer-engineer at the Hydrographic Naval Institute of Paris, Beautemps-Beaupré during his two hydrographic surveys of the Adriatic – the first in 1806, and the second from 1808 to 1809 (Kozličić, 2006). The pilot entitled Portolano del Mare Adriatico (Adriatic Sea Pilot) from 1830, written by an Italian captain Giacomo Marieni, was the first exhaustive and extensive work of that kind written for the eastern Adriatic (Zore-Armanda, 2000; Kozličić et al., 2012; Faričić, Mirošević, 2017).

Giuseppe Rosaccio's isolario

The sailing route across the eastern Adriatic, as a segment of longer navigation across the Mediterranean, according to Rosaccio’s isolario, started at Venice, and was similar to aforementioned pilgrimage sailing routes. Rossaccio also mentioned alternative “unfavourable” routes which are, in general, routes stretching alongside the coastal area at the foot of Velebit and Biokovo. An important transversal route which was directed southward and southwestward alongside mid-Dalmatian islands, and later across Palagruža and Tremiti islands towards western coast of the Adriatic was also mentioned (Pavić, 2000; 2003). It is noticeable that such a route does not deviate much from pilgrimage routes (Figure 7), and not even from antique routes, which leads to the conclusion that climate conditions, including winds, have not changed (significantly) during the whole period, and that maneuvering abilities of vessels powered by sails and oars had not improved a lot until the appearance of steam ships in mid nineteenth century (Pavić, 2000). It can be assumed that, although Rosaccio in his isolario did not recommend the routes at the foot of Velebit and Biokovo due to threats posed by the Uskoks of Senj and Ottoman presence in mid-Dalmatian littoral area (Pavić, 2000), their peripheral role was probably a consequence of weak land transport connections of that area with its hinterland and steep and inhospitable coasts, and also of frequent and intensive bora flows across those areas, unfavourable for navigation (Kozličić et al., 2012).

Beatutemps-Beaupré’s Hydrographic survey of the Adriatic

The earliest thorough study about the Adriatic winds and their influence on sailing was performed by Beautemps-Beaupré in 1806 and 1808 – 1809 during his hydrographic survey of the Adriatic. Beautemps-Beaupré classified winds into main winds and other winds, where bora and jugo are treated as main winds (Kozličić, 2006). He claimed that bora is a wind which is threatening to navigation due to its immense intensity; it flows at its maximum at the end of winter, appears instantly and blows in sudden breaches (with highest intensity across the north Adriatic and Kvarner archipelago), and it “throws” ships towards Italian coasts. Nevertheless, Beautemps-Beaupré also mentioned positive effects of bora events on sailing – ships could harvest bora as a power resource when sailing from eastern mid-Adriatic (from Dugi Otok Island to Vis Island) towards Cape Gargano (Italy), positioned southwest of Palagruža. Comparing that data with the contemporary wind monitoring and coastal relief it is noticeable that bora velocities across that part of Adriatic are lower in scale and less prone to oscillations and sudden breaches due to relatively flat relief, so it could be more suitable for sailing. The other way of harvesting bora, according to Beautemps-Beaupré, is in situations of sailing outwards from the Adriatic, when the ship is positioned in the open seas of the South Adriatic (Kozličić, 2006). Beautemps-Beaupré stated that jugo does not cause such threats as bora does, but during winters it is followed by thick fog that blocks the sight when observing coast from ship’s decks or masts. He also stated that there are frequent occasions of multiple winds generated across the Adriatic at the same time – for example, a situation of bora event across Kvarner archipelago and simultaneous jugo event ongoing southwards across open seas. Other winds, less threatening than jugo, as described by Beautemps-Beaupré, are N and NW winds, generate smaller waves, and are favourable for sailing outwards the Adriatic. Mistral, as he claimed, occurs during spring and summer, with frequent inter-substitutions of mistral and jugo with ongoing periods of calmness between the two. He also mentioned nighttime winds coming from coastal land mass, which flow at lower and higher velocities, and can (sometimes) cause difficulties to ships sailing into ports (Kozličić, 2006).

Giacomo Marieni's Adriatic Sea Pilot

A valuable study of the influence of wind on sailing was given by Giacomo Marineni in his 1830 work entitled Adriatic Sea Pilot. The pilot was created upon data collected during systematic and thorough hydrographical survey of the Adriatic performed in cooperation of the Kingdom of Austria, Kingdom of Two Sicilias and the United Kingdom 1818 – 1819. Marieni’s pilot, except for having a separate chapter about winds, as in Beautemps-Beaupré’s work, contains particular wind properties across separate eastern Adriatic sub-regions. Marieni divided the eastern Adriatic into Istria coast, Kvarner archipelago, coastal area at the foot of Velebit, and north, central and south Dalmatia. Additionally, warnings about prevailing stormy winds were included in form of notes on segments of the nautical chart entitled Carta di cabbotaggiodel Mare Adriatico (Adriatic Sea Cabotage Chart), published by Milano Military-Geographic Institute 1822 – 1824 (Faričić, Mirošević, 2017).

Marieni considers that the eastern Adriatic coast is much more favourable for navigation than its western coast, for both directions – sailing inwards or outwards of Adriatic, while the western coast is favourable for sailing outwards only in spring (Marieni, 1830). According to Marieni, bora is an umbrella term for winds that flow from NNE, NE and ENE directions; it is the most prevailing wind besides jugo and causes fear among mariners due to its sudden appearance, not allowing them to adapt the sails and course of the ship adequately. In addition, he claims that bora most frequently occurs in winter (in intervals lasting from 9 to 30 days, and up to 3 days in summer) across Kvarner archipelago, whilst often generating unpredictable sea conditions that can throw ships towards western Adriatic coast. Jugo is described as a wind entering the Adriatic from the Strait of Otranto, then following the axis of the Adriatic, causing wavy sea conditions and can be threatening to navigation (due to its intensity and 3 to 9 day intervals). Mistral is described as a wind that occurs in spring, and is substituted by land-breeze during nighttime. Marieni considers mistral as the most favourable wind for sailing outwards the Adriatic, and very favourable for sailing in general because it is not able to generate high waves (Marieni, 1830). It needs to be emphasized here that Marieni’s observation over differences in bora flows is a highly valuable record, but at the same time his claims that spring is the season of most frequent mistral events do not go hand in hand with contemporary measured data. Mistral is a type of air circulation caused, besides local sea-land thermal imbalance, by specific summertime synoptic situation covering the Mediterranean that occurs only in summer.

Marieni stated that along the eastern coast of Istria bora is the most intense wind that flows in parallel with the shore, and that most immense bora breaches are along Limski channel. Jugo is, as he wrote, a less threatening wind flowing from SE direction which can generate high waves in the open sea, most frequently in winter and spring, and Cape Premantura is the most dangerous part of Istrian coast during jugo events. He also wrote that alongside coast of Istria there are events of NW breeze at daytime, which is probably mistral, and not pure sea-breeze, because sea-breeze should have the direction closer to SW. According to Marieni, Kvarner archipelago is part of the Adriatic with most intense bora events which are able to throw ships to open seas, and that in summer bora has lower intensity and is called burin. Jugo is not such a threatening wind across Kvarner archipelago, except for the events of sudden bora-jugo alterations. Marieni highlighted the coastal area at the foot of Velebit as a region of excessive bora events, especially in the vicinity of Senj, and that there are scenarios of furious bora breaches across Senj region, while few miles ahead, across open sea there is some other wind or even total calmness. Dalmatia is divided by Marieni into north, central and south part. According to Marieni, the border between North and Central Dalmatia is Cape Planka (Ploča). North Dalmatia is full of safe havens for ships. Littoral area of Šibenik is region of most intense bora, while the area around Dugi Otok is most prone to south and southeast wind. For ships that sail from Zadar to Šibenik during winter, Marieni recommended routes that go through channels between the islands, and to avoid sailing past outer coast of Dugi Otok. Cape Planka and Pelješac peninsula are Marieni’s borders of Central Dalmatia, a region he says is prone to northern wind breaches, especially along its coastal part-He also claims that jugo causes storms in winter but flows mildly in summer. He also wrote that there are events of east and northwest breeze across this region but that island relief dampens stronger flows from NW and SW directions. According to Marieni, South Dalmatia is a region prone to bora events, particularly around Neretva river mouth, whilst jugo is the most unpleasant wind across open seas, southeast of Dubrovnik, with its intensity somewhat declining in areas closer to the island coasts (Marieni, 1830).

Conclusion

Archaeological findings and posterior written sources of sailing across the Croatian part of the Adriatic that took place since prehistory till Modern Age undoubtedly prove that winds were an extremely important factor in terms of sailing organization. Throughout the history of sailing in this area it has been reflected in two ways– spatially by choosing the appropriate sailing routes and temporarily by organizing sailing during warm part of the year with more favourable weather conditions. During prehistory and antiquity, longitudinal sailing routes stretched mostly in close proximity to the coast and through channels between islands, while later, especially medieval and modern age routes went mostly along island open sea boundaries as bigger ships and new navigational technologies were introduced (dry compass, nautical charts). It is obvious that winds such as mistral, mild jugo and local air circulation were favourable for mariners, with mistral being the most favourable one. Winds that were usually not recommended for sailing were bora and stormy jugo, although, according to Beautemps-Beaupré, even bora, after declining in strength, could be efficiently harvested across open seas for sailing southwestwards.

A comparison between the information about winds contained in Beautemps-Beaupré’s and Marieni’s works and similar older sources showed that their writings are an improvement in terms of precise wind descriptions, while core descriptions of specific winds and how they affected sailings are basically the same – a confirmation of the hypothesis that weather conditions have not been changed, same as their influence on sailing. A comparison between those writings and contemporary wind data (frequencies and velocities), showed that they are a testament of millennia-old knowledge of winds across eastern Adriatic. It also showed that Beautemps-Beaupré’s and Marieni’s works are, despite the lack of precise measurements and exact numerical data, highly valuable material, not only in terms of historiography, but in terms of their possible application even today.

Notes

The free online CSS minimizer tool will compress the style files for your websites in seconds.

Never forget to clean HTML code and double check your content before publishing an article!