Historical evolution of the Abalario lagoon complexes (Doñana Natural Park, SW Spain)

85 Historical evolution of the Abalario lagoon complexes (Doñana Natural Park, SW Spain) Arturo Sousa Martín and Pablo García Murillo Departamento de Biología Vegetal y Ecología. Universidad de Sevilla. Apdo. 874. 41080 Sevilla. ABSTRACT This work aims to demostrate which factors have determined surface drainage and distribution of the present-day hygrophytic vegetation of Abalario, in Doñana Natural Park (Huelva). The area is difficult to interpret because its territorial framework is disfigured as a result of an intensive use of its natural resources having been superimposed in the normal territorial dynamics. The area occupied by lagoons has been reduced by more than 30 % in less than 40 years. This percentage is almost 90 % in the case of peat-bogs. Two large lagoons of which historical records exist have completely disappeared namely, Laguna de La Mediana (“the Median Lagoon”) and Laguna de Invierno (“the Winter Lagoon”). A subsidiary aim is to establish the usefulness of the methodology used, i.e. field work carried out in parallel with historical documentation using cartography and records of past land use. This combination of approaches provides a key to interpret actual surface hydrology and vegetation. Key words: Landscape, ecological history, Doñana, lagoons, peat-bogs, SW Spain. RESUMEN Este trabajo pretende mostrar los factores que han determinado el drenaje superficie y la distribución de la vegetación higrófita actual en Abalario, en el Parque Natural de Doñana (Huelva). El área es difícil de interpretar debido a que su marco territorial está desfigurado como resultado de un uso intensivo de sus recursos naturales que han sido superpuestos a la dinámica territorial normal. El área ocupada por las lagunas se ha reducido en más de un 30% ( y casi el 90% en el caso de las turberas) en menos de 40 años. Dos grandes lagunas mencionadas en los registros históricos han desaparecido: Laguna de La Mediana y Laguna de Invierno. Al mismo tiempo, el trabajo intenta establecer la utilidad de la metodología utilizada: trabajo de campo realizado simultáneamente con un estudio documental (cartografía histórica, registros sobre los usos del suelo). Esto proporciona la clave para interpretar la hidrología superficial y la vegetación. Palabras clave: Paisaje, ecología histórica, Doñana, lagunas, turberas. Limnetica 16: 85-98 (1999) © Asociación Española de Limnología, Madrid. Spain. ISSN: 2013-8409. INTRODUCTION The recognition and study of changes in the landscape, entailing the study of the history of ecosystems, is an enormously interesting area of knowledge, not only in itself, but also for the information gained on the relationships between the different elements of natural systems. González Bernáldez (1992) set the scale of historical changes taking place in wetland complexes. He indicated wetlands are arguably the most impacted ecosystems, and estimated that approximately 75 % of European wetland ecosystems have already been destroyed. It is important to perfect a methodology for this type of study, involving historical tracking of changes experienced by natural systems, that could also be extrapolated to other ecosystems of similar nature. The interest in the topic has increased because wetland systems, are very sensitive to large-scale changes, and may act as an indicator ecosystem 86 Sousa & García Murillo for human perturbations. In our case, we will see that it indeed necessary to consider global climatic changes in a complete explanation of regional changes and processes. The study zone (Fig. 1) is situated on the coastal arc of Huelva, between deltas of Rivers Guadalquivir and Tinto. More precisely, it is the area between the Costa de Castilla and Rocina stream, bounded east and west by the C-445 road and the “Montes Propios” of Moguer, respectively (Fig. 2). These boundaries are shared with those of the western sector of Doñana Natural Park, although other land not belonging to the Park is also included. The approximate area under study is 25,000 hectares. Substrate is dominated by Quaternary aeolian deposits, produced by a succession of dune fronts. In certain north-west areas, there is outcropping of Pliocene sands (Leyva & Pastor, 1976). Climate is typically Mediterranean, although with a strong Atlantic influence, shown by milder temperatures and greater precipitations than inland (Allue, 1990). Biogeographically, the study zone belongs to the Huelva Littoral Sector of the Gaditano-Onubo- Algarviense chorological province, included in the Mediterranean region (Rivas-Martinez, 1987). Bioclimatically, it is located in the upper thermomediterranean bioclimatic stage (Rivas-Martinez, 1987). It includes land belonging to the townships of Almonte, Bonares, Lucena del Puerto, Moguer, and Palos de la Frontera. After the Spanish Civil War in the 1930s, traditional land uses, such as hunting, grazing, pasture land, and charcoal burning, all of which were compatible with conserving this highly-valued landscape, gave way to strong state intervention (Ojeda, 1987). The new dominant activity, was part of a national, autarchic, post-war economy, and concentrated on the urgent need to obtain raw materials (i.e. primarily wood pulp and essential oils) from rapidly-growing forestry crops. The territory was given over largely to plantations of eucalyptus, except in southernmost zones, where pines were planted instead. The result of such a colonialist design was a radical alteration to the land. The land would no longer be considered as a useless desert (Ojeda, 1992), but a model region, a paradigm of forestry activity, with an exemplary production of raw materials. This ignored its subsequently recognized high environmental value, i.e. numerous lagoons, high spatial heterogeneity, high diversity of organisms and habitats, a wealth of different biological communities and rare species. The gradual decline of forestry, due to the fall in production, inevitable in this type of crop, and the drop in value of raw materials, lead to abandonment of and depopulation (1953-68, according to Ojeda, 1987). The study zone today constitutes a phenosystem difficult to interpret disfigure and indecipherable, after decades of missmanagement. Evidently, present-day communities and lagoon formations cannot be interpreted without closely studying the historical changes, which were deep and far-reaching, causing the original natural systems to conform to a set of physical and biotic elements which are fragmented and disconnected. Such information is a basis or refe- Figure 1. Study area. Área de estudio. Evolution of the Abalario lagoon complexes 87 rence tool for planning programmes (e.g. PORN, PRUG, restoration, public use) aimed at restoring this high-valued ecosystem to close to its original state. The transformations that have taken place in the zone are a good example of the general transformation of the Atlantic littoral landscape linked to wetlands. However, the ecological nature of the elements involved in this system and the relationships between them show that this type of ecosystem is, or acts as if it is, especially sensitive to global climatic changes. Hence the interest in an approach that takes in these different viewpoints. The paper deals with the changes produced in the study zone in the last two centuries. It considers changes to hygrophytic vegetation, to streams and brooks, and the transformations ex- Figure 2. Present-day map of the study area, showing main localities, lagoons, streams, and other toponyms. Mapa de la zona. Mapa actual de la zona de estudio mostrando las principales localidades, lagunas, arroyos y otros topónimos. 88 Sousa & García Murillo perienced by lagoon formations, separately studying both existing types (i.e. seasonal lagoons and peat-bog). Finally, we look at the disappearance of two large lagoons whose existence is documented historically in archives and map libraries. MATERIALAND METHOD The method used consisted in the cartographic reconstruction (scales 1:25 000 and 1:50 000) of changes in both vegetation and surface draining sources of data were field observation laboratory data and archives. Surface measurements were obtained with a planimeter (PLANIX 5000). Resulting error is small given the low relief of the area (with slopes not exceeding 3 %). Resulting maps were analyzed in the laboratory using aerial photographs (i.e. flights made in 1946 -scale 1:40 000-, 1956 -scale 1:33 000-, 1987 -scale 1:20 000-) and satellite images [Lansat-TM (1986 and 1990) and Spot (1989)]. Historical information, gathered from archives and map libraries, has allowed to track the rapid changes in land use and resource exploitation which have occurred in the area. An acceptably reliable reconstruction of changes experienced in the study zone during the last two centuries helped identify the initial landscape disfigurement. The basis has been established for a model of the origin and functioning of surface drainage and vegetational dynamics across the landscape. The regional toponymy, its variation and possible origins, have been analyzed historically from maps supplying data from the beginning of the 17th century to date. RESULTS Changes in the hygrophytic vegetation Floristically, the study zone is characterized (Garcia Murillo et al., 1995; Garcia Murillo & Sousa, 1997) by the large number of taxa with Atlantic distribution (i.e. Sphagnum inundatum, Theliptheris palustris, Osmunda regalis, Erica ciliaris, Pinguicola lusitanica, Potamogeton polygonifolius, Molinia caerulea) and the frequent occurrence of endemisms (i.e. Arenaria algarviensis, Daucus arcanus, Linaria tursica, Vulpia fontquerana). Woody plants are grouped into ten units (Garcia Murillo & Sousa, 1997), each including various phytosociological associations. Following the criteria of Rivas Martínez (1987), these units can be divided into three series of plants. The first are the edaphophilous series of junipers, the second are the riparious oligotrophic woodlands, and the third, the climatophilous series of thermomediterranean coastal cork-oaks. Garcia Murillo & Sousa (1997) studied last forty years vegetational changes describing… a) The decrease, or near loss, of vegetation associated to small streams and gullies. This vegetation is very rich in lianas and creepers (Lonicero hispanicae-Rubetum ulmifolii, Gully Unit). Figure 3. Changes in the hygrophytic vegetation. The decrease and near-disappearance of peat-bogs parallels an increase in hygrophytic shrub of Monte Negro. The area of lagoons has decreased 38.26 % (1301 ha) since 1956, with peat-bogs undergoing the greatest reduction. Cambios de la vegetación higrófita. Se puede ver que la disminución y la casi desaparición de las turberas es paralela al incremento del matorral higrofílico de Monte Negro. El área de las lagunas ha disminuido 38.26% (1301 ha) desde 1956, siendo las turberas las de mayor reducción. Evolution of the Abalario lagoon complexes 89 b) The reduction in area occupied by plant communities associated to peaty environments (Erico ciliaris-Ulicetum (minoris) lusitanicus, Peat-bog Unit). These areas were extensive in the fifties. Today, they are restricted to the Línea de Rivatehilos (Rivatehilos Line). c) The increase in area occupied by hygrophytic shrubs (Erico scopariae-Ulicetum australis, Abalario Unit). This is a consequence of the degradation of peat formations. Hygrophytic plants have been able to take because they are less demanding with regard to flooding and availability of organic matter. Figure 3 shows the changes in the hygrophytic vegetation of the zone during the period 1956-1987. Changes in the streams and brooks The study of the streams and brooks reveals considerable changes (Fig. 4). In 1987 (and today). The length of the few streams opening into the Atlantic has almost halved since 1956. Today these streams hardly reach the road joining Mazagón with Matalascañas (Fig. 4). Torre de la Higuera stream, for instance, has totally disappeared. The Mazagón streams, on the other hand, has been reduced in main length. The grater loss is for Loro stream, particularly when total length is considered to include tributaries (Fig. 4). Overall, reductions are of more than 60% in original length. Today, the area around Médano del Asperillo has drainage, either endorheic nor exorheic. However, there are traces of ancient runoff lines in the area between the southern edge of the peatbogs and the Asperillo dunes. Changes in gullies and brooks opening into the Rocina stream; northernmost in the study area, are less marked. There is, though, a decrease in the number and ramifications of gullies or seasonal streams. Planimetric measurements show, however, that this reduction was of only 30% during the period 1956-1987. Topping and felling of, especially, Eucalyptus globulus has dismantled this complex system of streams. The idea behind this ma- Figure 4. Decrease in length of each stream and brook in the study area (1956-1987). The overall decrease exceeded 60 % for streams, and 30 % for brooks and gullies. Cambios en arroyos y algaidas. Disminución en la longitud de cada arroyo y algaida en la zona de estudio (1956-1987). La reducción promedio excede el 60% para los arroyos y el 30% para las cañadas y algaidas. 90 Sousa & García Murillo nagement is that the system will reorganize in a subparallel network, similar to its previous state. With regard to the situation of streams before the spanish Civil War, historical maps are available of areas to the coast (Gussefeld, 1781; Saavedra, 1810; Coello, 1869; Montojo y Salcedo, 1875; Valverde, 1880; Gonzalo Y Tarin, 1887; Carrasco, 1892; Noriega & Cobo del Guzmán, 1897; Anon., 19th century; 19th century naval maps; Ibáñez de Ibero, 1902). Topographic and geological maps drawn in the first half of the 20th century are also available (San Miguel, 1913; Gavala, 1936; Mapa Topográfico Nacional, 1947; Mapa Topográfico Nacional, 1949; Mapa Topográfico Nacional, 1951). The longest streams west of Médano del Loro (Loro Dune), Loro, Harinoso, Harinosillo, Rompeculos, Mazagón, Hueso, and Morla are undoubtedly the Mazagón and the Loro (Fig. 4). The latter could be the original drainage channel of the Winter Lagoon, as discussed below. In addition, there were other smaller ravines today extinct (i.e. the Atarazanas, Tojar del Lagarto, Carboneros streams, and the Higuera channel). It would be excessively speculative to associate disappearance and regression of these streams solely to forestry activities over the last 50 years. In fact, it seems the decline of fluvial habitats began long before the XIX century. Regression became more marked during the nineteenth century, and experienced strongest impacts with forestry work starting in the 1940s. Historical documents published by Mora Figueroa (1981), referring to the construction of watchtowers on the Huelva coast in the 18th century, indicated that the greatest availability of surface waters was found in the Asperillo stretch, and that there was no water in the higher stretch and very little in the coastal stretch west of River Oro (sic) (Sousa & Garcia Murillo, 1998). Changes in the peat-bogs The presence of communities of hairy heathers (Erico ciliaris-Ulicetum (minoris) lusitanicus) points to a peaty substrate originated by a process of pseudogleization requiring sluggish waters with abundant vegetation. Today, the area occupied by these peaty lagoons (the peat-bog unit of Garcia Murillo & Sousa, 1997) extends over the zone known as “Rivatehilos” or “El Hilo” (“The String”). This unit or “string” (the local name for a series of aligned lagoons) divides the territory into two zones known as Mediana Baja (Low Median) and Mediana Alta (High Median). The two units have a different geomorphology, with high and low aeolic deposits respectively (Borja & Diaz del Olmo, 1987). On the other hand, Mediana Baja is characterized by its Baja shrub and Mediana Alta by its Abalario shrub (Garcia Murillo & Sousa, 1997). These data suggest a different geological origin to both areas. The “reforestation” with eucalyptus, E. globulus in the high zones and E. camaldulensis in the more water-logged areas, after the Civil War lead to great evapotranspiration losses from mainly forests, and to a deepening of the water table. This was more marked with distance southwards from Rivatehilos. The extent of forestry practices impact on hydrology and ecosystems (Márquez, 1985; Ojeda, 1987; Espina & Estévez, 1992, 1993), especially acute after the Civil War, is still not fully known. Vegetational changes indicate considerable reductions in water availability. The impact of monoculture forestry lead, for example, to a decrease in the area flooded by lagoons and to the replacement of peat-bog vegetation [Erico ciliaris-Ulicetum (minoris) lusitanicus] by hygrophytic shrubs of “monte negro” (i.e. local name for communities of woody heath: Erico scopariae-Ulicetum australis).In addition to data obtained from analyse of aerial photographs, management-related alterations are unquestionably revealed by written documents (Gonzalo & Tarín, 1887; Castro 1900a, 1900b, 1912; de la Lama, 1951; Kith Tassara, 1952; Espina & Estevez 1992, 1993), oral testimony, from both natives of the zone (Espina & Estévez, 1992), and from former managers (De La Lama, pers. comm.; Rodríguez Navarro, pers. comm.) Peat bogs of the study zone show the clearest impacts caused by forestry. Some 40 years ago, the peat-bog vegetation extended over a much wider area, whose southern boundary coincided Evolution of the Abalario lagoon complexes 91 Figure 5. Gradual disappearance of the southernmost peat-bogs, before the 19th century. The Winter and Median lagoons have disappeared, probably extinct by the end of the 19th century. The map also shows the direction of disappearance of lagoons (i.e. from SW to NE). Cambios en las lagunas turbosas. Se muestra la progresiva desaparición de las lagunas turbosas más meridionales desde finales del s. XIX. Además destaca la desaparición completa de la laguna de Invierno y de la laguna Mediana, probablemente desde finales del s. XIX. 92 Sousa & García Murillo approximately with the path joining the Poleosas lagoons with the village of Abalario. Indeed, past disttribution relicts are found in isolated “peaty points” (Erico ciliaris-Ulicetum (minoris) lusitanicus), far from Rivatehilos. The area of peatbogs has almost disappeared (Fig. 5), reduced today to 180 ha, scattered across about 30 isolated spots. This process can be assumed to have begun before the appearance of forestry. However, intensification of eucalyptus exploitation for essential oils and wood pulp, accelerated it. The result was a decrease of almost 90 % in the area of peat-bogs between 1956 and 1987. Changes in the seasonal lagoons Lagoon complexes and peat-bogs differ in their drainage and filling cycle. Changes in lagoons are not as drastic as those in peat-bogs (Fig. 6). No important changes are observed in the easternmost lagoons, except the artificializing of the Acebuche, La Pajas, and Huerto lagoons in the 80s to make a reception area for visitors to Doñana National Park. This lagoon complex appears to be linked to, or is a continuation of, lagoons more to the east, which are within the confines of Doñana National Park. It seems western lagoons formed part of a great ancient seasonal lagoon, mentioned in historical documents as the “Winter Lagoon”. Western seasonal and permanent lagoons have maintained their overall area in the period 1956- 1987 (Fig. 3). However, it is probable that some ha- ve disappeared and been converted into “Monte Negro”. On the other hand, new ones have appeared as a result of degradation of the ancient peat-bogs. The Median Lagoon The historical maps and texts suggest that the origin of the name “Mediana” is in the past existence of a great lagoon at the western edge of Abalario, the “Median Lagoon”. The size of this lagoon (Madoz, 1848) would have amply exceeded any of the extant peat-bogs. This great lagoon, shown on various maps (Coello, 1869; Valverde, 1880; Carrasco, 1892; Ibáñez, 1902), gave rise to a string of lagoons, demonstrating an interphase or discontinuity. The same maps allow reflection of the name “Mediana”. This might indicate the size of the lagoon, since it would have been the largest of the peat-bogs, but smaller than the great “winter lagoon”. Hence, the name “median” shows it as been intermediate size between the various marshy formations of the zone. This territorial conformation fully coincides with property deeds of the 40s and 50s regarding sale, purchase, and valuation reports, as well as with the perception of landscape held by those involved in forestry work. Originally thus, it seems there was a great seasonal lagoon (Winter Lagoon), another large, permanent lagoon (Median Lagoon), albeit slightly smaller than Winter Lagoon. These lagoons would have had different vegetation, associated to their different hydrology. Also, there would have been numerous very small lagoons, the Rivatehilos complex of peat-bogs. The group of peat-bogs characterized by the community Erico ciliaris-Ulicetum (lusitanicus) minoris and originated from this “median” lagoon to the whole most of which have now disappeared, lead to the extension of use of the name “Median” area. This toponym is present since at least the 19th century. Similarly, the name “Tojar Grande” (“Big Furze Shrub”) on maps of 1945, 1981, and 1987, undoubtedly refers to a large area occupied by hygrophytic shrub (heather). This site approximately coincides with that of the extinct Median Lagoon (1880). It can therefore be speculated that the vanished great lagoon was replaced by hygrophytic vegetation (heather) and that, from the 1940s, the name “Tojar Grande” appeared in place of “Mediana”. To distinguish topographically, zones are named “High Median” and “Low Median”. The Winter Lagoon Archives and historical maps show the existence of a great ancient seasonal lagoon named the Evolution of the Abalario lagoon complexes 93 Figure 6. Changes in the seasonal lagoons. Cambios en las lagunas temporales.