First of all, it should be clarified what a GIS is.[1] Jeffrey Star and John Estes (1990) defined a GIS as a system designed to work with data referenced by spatial or geographic coordinates, a sort of higher-order map. GIS can be understood as a map-making method that overcomes classical cartography's barriers. In the same way that texts have moved from paper to computers, maps have also made this qualitative leap (Olaya Ferrero, 2020).
Geographical Information Systems were originally used by land planners, earth scientists and physical geographers. Gradually, they are being utilized by historians and art historians. They are beneficial in so many ways for these disciplines. They can be applied to creating highly accurate historical maps depicting past boundaries, architectural heritage, trade routes, battles, political networks, etc. However, GIS is not only beneficial for visually structuring research results but as an analysis tool in its own right. As Gregory, Mostern, and Kemp pointed out, GIS "add the capacity for sophisticated spatial thinking to existing historical scholarship through both its ability to portray locations and their attributes in innovative ways and to foster original questions about the relationship of locations to one another" (2001). For example, if we represent on a map all the palaces of the Habsburg family in the 16th century, the map would first be a scientific product in itself, containing information about these residences. But the map would also be a tool that allows researchers to establish possible architectural relationships between the palaces or to determine the time the Habsburgs spent in one place or another, including the estimated travel time between each location. Therefore, the map is no longer an end product but a research tool in GIS.
But what is exactly the functioning of a GIS? This software deals with two types of data. On the one hand, attribute data provides information about a concrete item; this can be a city, a palace, or a specific feature such as a sculpture or a fountain. This attribute data can be just the name of the item but also any information that might be provided about it: time frame, characteristics, style, artists, promotors, etc. On the other hand, GIS deals with spatial data that is represented by coordinates. Each attribute data of an item is connected to spatial data. This provides information about where the item is located on the earth's surface. Therefore, it must first be clear that GIS explicitly brings together a specific item's spatial and attribute data (Gregory, Mostern, and Kemp, 2001).
All the spatial and attribute data collected in a GIS map is organized on an attribute table. This table is usually arranged in an Excel spreadsheet. The sheet can be imported into a GIS software, and the other way around, information can be drawn in GIS and later extracted into a spreadsheet. Once on the map, all the spatial and attributable data is collected on layers. A map layer is a rough equivalent of a database table. Each layer contains information about a particular theme or topic. For example, a layer can collect all the Mexican restaurants in a city. Closer to the present investigation, a layer can contain information about all the grottoes displayed in a palace, while a second layer can gather facts about all the mythological fountains.
But how exactly is spatial information depicted on the GIS map? Coordinates, or sets of coordinates, are the language in which all spatial data is drawn. Specifically, three basic spatial features structure all the information: the point, the line and the polygon. A point is stored in the GIS software as a pair of coordinates (X, Y). They represent a concrete geographical location. In our research, a point usually represents specific fountains, grottoes, wells, springs, etc. A line is created by joining a string of individual points. These points, often called vertices, are placed where the line changes direction. A line must consist of at least two points: the start and end points, usually referred to as nodes. While researching water in specific residences, lines represent natural and manipulated water flow through canalizations or pipelines. But lines can also represent more abstract concepts. Lines can create a network. In our research, lines represent the connections between the multiple palaces where a specific artist worked. Finally, Polygons are created by completely enclosing an area with lines. Waterworks that occupy a considerable surface area, such as tanks, ponds, or lakes, are represented by polygons. In addition, the areas where it is estimated that there were remains of hydraulic infrastructures that cannot be georeferenced through very specific coordinates are also drawn as polygons.
Having understood, broadly speaking, the basic functioning of a GIS, it will be explained how it has been implemented for the study of hydraulics and waterworks in 16th -century residences. Two radically different types of maps have been designed during the investigation. They correspond to the two different scopes of the research: transnational and specific. On the one hand, a generic map of 94 water-related residences all over Europe has been drawn. This pan-European map, working as an analytical tool in itself, has established relationships and networks between the residencies. On the other hand, four individual maps have been created, depicting the particular hydraulic system of each of the four Iberian palaces that are in-depth study cases.
GIS for the comparative analysis of European data about residential hydraulics.
As mentioned, creating a pan-European map aimed at understanding the historical evolution of the residential water supply systems all over 16th -century Europe. Data on the 94 palaces were recorded in an Excel database (Fig. 1) and later translated into a GIS map (Fig. 2). The resulting map can be consulted on the project webpage.[2] The coordinates, referenced as spatial data, are accompanied by attribute data. This attribute data included information about the years of construction of the residences, the promotors who commissioned them, and the engineers and artists who contributed to the creation of the waterworks. A couple of paragraphs with brief information about these waterworks are also included. The attribute data appears on the map when clicking over the residence that wants to be consulted.
In this pan-European GIS map, an analytical study has been made of the geographical movement of artists related to water and the networks of residencies built by the same promoter. Gren arrows connecting the different palaces (Fig. 2) represent the artists and engineers who worked in different buildings. Red arrows link the residences commissioned by the same promoter. The representation of these "green and red" networks has allowed the drawing of some interesting conclusions, which without GIS would have been more difficult to see:
The map shows considerable regional mobility between hydraulic engineers and water-related artists. Jacques Du Broeucq moved among the primary residences of Belgium at the service of Emperor Charles V. The team formed by Juan Bautista de Toledo and the Dutch engineer Pietre Jansen worked for King Philip II in most of the Reales Sitios of Spain. Bartolomeo Ammannati operated between Rome and Florence, designing fountains and hydraulic devices for the villas Giulia, Medici, Castello, Boboli, and Pratolino. As a last example, Vignola collaborated with Pirro Ligorio in the Bomarzo forest, the Villa Farnese in Caprarola, the Villa Lante and the Villa Giulia.
However, through the GIS, it can also be concluded that there is practically no international circulation among hydraulic engineers, except for four notable exceptions: Pacello da Mercoliano and Fra Giocondo moved from Naples to the French court at the beginning of the 16th century to work in the palaces of Amboise, Blois and Gaillon. The Francini brothers were trained in Pratolino and left Florence at the end of the century to work in Saint-Germain-en-Laye, Fontainebleau and Luxembourg residences. Solomon de Caus worked in the courts of London, Brussels, Heidelberg and France at the beginning of the 17th century. And Giambologna did not move, but his sculptural work ended up in the fountains of the leading European courts of the time.
This pan-European GIS map also shows how promotors were one of the main determining factors in the development of hydraulics throughout the 16th century. The patronage of various palaces with extensive waterworks made them fundamental agents for developing water technology. Among these notable promotors, we must highlight Francis I, Catherine de' Medici, and Henry IV in France. The Holy Roman Emperor Charles V and his son Philip II in Spain. And the secular and religious Italian nobles such as Cosimo I de 'Medici, Ferdinando I de' Medici, Ippolito II d'Este, or Pope Sixtus V.
GIS for the study of individual residential hydraulic infrastructures.
A clear methodology has been developed for researching the water supply systems of the four Iberian residences. This methodology is based on four fundamental steps: First, to understand the structure and history of the palace. Second, to collect all the water-related archival documentation such as plans, contracts, reports, nomination letters, etc. Third, to do an architectonical survey cataloguing and georeferencing in a GIS the remains of the hydraulic infrastructure, recreating an estimated path of the missing parts. Fourth and last, to write a conclusive analysis comparing all the historical and geographical data. Therefore, it is clear that GIS has been essential for developing this investigation.
The GIS maps of the Sintra, Seville, Vila Viçosa and El Escorial palaces always represented two things. The external water supply system, which includes the natural sources where water is obtained and the transportation methods to the palace. And the internal water infrastructure, meaning the networks of fountains, tanks, cisterns, pipelines, etc., inside the residences.
The Monastery of El Escorial can be taken as an example. This royal residence was supplied through a network of more than four kilometres of pipelines based on exploiting the streams Helechal, Tobar and Romeral. For this purpose, large deposits (also known as arches) were built at strategic locations along the course of each stream to capture, purify and redirect water to the monastery through different underground channels. These arches were granite structures, generally rectangular, with hipped roofs and a height of about three metres (De Vicente García, 1991: 39). Inside, the tanks for decanting the water are still preserved (Fig. 3).
The entire network of aqueducts of El Escorial has been tracked and georeferenced by means of in-situ work. The architectural remains of the system have been searched for in the mountains surrounding the palace, and the coordinates of each item have been registered into the GIS. The result is an interactive and navigable map representing all the structures (Fig. 4). This map can also be consulted online on the project website.[3] Symbology codes, common to all the palaces, have been followed. The yellow squares represent the arches built for water retention, storage and redirection in the aqueducts. The continuous lines correspond to the preserved aqueduct sections, while the dashed lines estimate the route of the aqueduct sections that have been lost or travel underground. The smaller red dots symbolize specific coordinates that serve as reference locations. The items can be identified on the map itself. Clicking over the symbol, a window is open with its name.
Suppose we enlarge the GIS image in the space where the monastery is located. In that case, the entire internal infrastructure of the building is depicted (Fig. 5). A complete article dedicated to its analysis will be published (Gumiel Campos, 2023). Still, a brief synthesis of its morphology can be done so far. The internal hydraulic infrastructure consists of a network of eight pipelines that depart from a deposit known as the Arc of Repartimientos, situated in the last part of the aqueduct (Fig. 4: Red cross in the map north to the building). These pipelines ran to the main areas of the monastery. The first four canalizations supplied the southern courtyards where the chambers dedicated to the monks' residence were displayed. They supplied the fountains, taps and tanks of the kitchens, latrines or apothecaries. The fifth and sixth pipes supplied the church, the sacristy of the seminary, the kitchen of the palace and the twelve rectangular fountains in the garden. The seventh and eighth pipes were dedicated to supplying all the fountains and taps of the seminary in the northern area of the building. In addition, eleven rainwater cisterns were constructed to complement the water supply network (De Andrés, 1965).
As with the external supply system, each of the water items has been represented using a specific symbology (Fig. 6). The eight pipes have been drawn in an identifiable colour differentiating one from the other. Fountains have been represented as blue circles, latrines as green triangles, kitchens as orange circles and fish stores as yellow pentagons. On the other hand, the large rainwater cisterns are represented as green polygons and the tanks clear blue. Drainage pipes of the sewer system are also represented as brown lines with arrows. Each of these symbols is grouped in a specific layer of the GIS, representing specific water features, allowing their whole selection at once. The same georeferencing process and symbolic language have been followed in the other three palaces (Fig. 7). In fact, it is hoped that this geographical language will be used in future research to analyze any other water infrastructure of residences that may be case studies.
The individual maps have not only served as a showcase for the final result of the research but have also been an essential part of it. Thanks to the analysis performed over the spatial data represented in the GIS, together with the study of the historical sources, it has been possible to draw a series of conclusions regarding the constructive evolution of the hydraulic infrastructures. For instance, the results of El Escorial will be published briefly in PALATIUM e-Publications volume 6 (Gumiel Campos, 2023). Another useful analytical feature of ArcGIS Pro is the ability to create three-dimensional maps. In cases where the orography is quite extreme, as in the National Palace of Sintra, 3D helps to understand the functioning of the aqueducts through the principle of gravity (Fig. 8). This undoubtedly, also facilitates the analysis of spatial data, generating more accurate conclusions about the functioning of the hydraulic infrastructure.
Nonetheless, the study of the water systems of each palace has presented certain challenges. For example, very little is preserved of the Caños de Carmona, the aqueduct that supplied the Alcázar of Seville (Fernández Chaves, 2011). And the few remains that do exist are underground and inaccessible, making it impossible to do accurate georeferencing on the GIS. On the other hand, the hydraulic remains of Vila Viçosa are better preserved, but historical documentation is practically non-existent, complicating accurate dating. In Sintra, the difficulty also arose when establishing a definite chronology, as the aqueduct underwent numerous modifications from the 13th to the 20th century.
Apart from the many advantages of GIS, one drawback must be solved in the future: the instability of the digital support on which the projects are based (see web pages). Although the layout of maps on web pages makes science more accessible and brings research closer to the general public, this type of support is unstable. It runs the risk of being eliminated after a few years. Moreover, it tends to be incompatible with the academic format of publications, which in book or article form, do not allow the introduction of maps in GIS format.