Litter on the coastline is one of the most obvious signs of marine litter pollution (see e.g. UNEP 2009). Beached marine litter such as broken glass, medical waste, fishing line, and discarded syringes can harm beach users as well as the risks associated with the leaching of poisonous chemicals (Sheavly and Register 2007, Thompson et al. 2009). The most abundant components of beach litter are plastics (Frost and Cullen 1997, Gabrielides et al. 1991, Tudor and Williams 2001).

More indirect effects come in the form of wider social impacts, influencing coastal tourism and the local economy as a result. According to the Convention for the Protection of the Marine Environment of the North-East Atlantic (OSPAR 2009), major land-based sources in the North Atlantic region include tourism recreation, illegal dumping, waste disposal sites and input from rivers. Major sea-based sources are fisheries, commercial shipping, pleasure crafts and offshore installations (UNEP 2009).

Surveys of litter stranded on the coastline are a primary tool for monitoring the load of litter in the marine environment and have been used world-wide to quantify and describe marine litter pollution (see e.g. Ribic et al. 2010). They can be used to measure the effectiveness of management or mitigation measures, identify the sources and activities leading to litter pollution and determine threats to marine biota and ecosystems (Martinez-Ribes et al. 2007, Cheshire et al. 2009). A number of studies have quantified beach litter and defined its components (see e.g. Ariza et al. 2008).

Monitoring programs that address clear questions are an effective way to gather information that can fruitfully be integrated with research and adaptive management (Lovett et al. 2007), as in the case of beach litter. The coming into effect of the European Marine Strategy Framework Directive (MSFD; Directive 2008/56/EC) will propel member states to create mechanisms for the management of marine litter, in order to achieve the goals established in the regulation. This is the first time that beach litter has been specifically addressed in an integrated way for the protection of the marine environment in a European directive. As pointed out by Galgani et al. (2010), the overriding objective of the MSFD on beach litter would be "a measurable and significant decrease in the total amount of litter by 2020".

In order to study beach litter on the Galician coast (NW Spain), three beaches were sampled from 2001 to 2010. The Galician coast has a length of over 1200 km, of which 820 km correspond to cliffs and 278 km to Galicia’s 723 beaches. The three beaches studied (A Lanzada, Baldaio and O Rostro) are wide, gently sloping sandy beaches in dune areas exposed to the open sea, located close to major shipping lanes and fishing harbours but not close to urban areas, river mouths or sewage outfalls.

In this paper, beach litter data were analyzed in order to 1) characterize the main elements, 2) search for significant seasonal patterns and 3) determine the main sources of litter on the Galician coast. This work may also be helpful for studying local/regional factors of beach litter management on North Atlantic beaches. To our knowledge, this is the first time that data on beach litter have been studied in the Galician region.

A major issue in producing and using data series is to assure their continuity, quality and accuracy, in order to provide good and homogeneous data sets. In practice, this factor is far from trivial. Despite the efforts in the region by municipal authorities, NGO’s and other bodies to clean up beaches, the majority of these actions are made without a detailed data register or a standardised protocol.

The monitoring methodology developed during the pilot project on marine litter of the OSPAR convention (www.ospar.org) was used throughout the sampling period in our study to ensure the homogeneity of the data base. The beaches were selected according to the agreed guidelines and protocols on the OSPAR pilot project (OSPAR 2007). The beaches should be:

- composed of sand or gravel and exposed to the sea
- visually and/or frequently littered (large marine items)
- accessible for ease of marine litter removal
- over 1 km in length
- not located near other input sources of waste, such as rivers
- already included in national marine beach litter activities

In addition, in order to comply with the requirements of the MSFD, which was transposed in 2010 to the Spanish national law (Law 41/2010 on the protection of the marine environment), the sampling on the beaches continued during the years 2007 to 2010 using the monitoring methodology developed within the OSPAR pilot project. One more beach (O Rostro; see Fig. 1) was also sampled in the region during this period.

Our study covers nine years (2001-2010) of beach litter data with a frequency of four samplings per year (winter, spring, summer and autumn) at A Lanzada and Baldaio beach. In the case of O Rostro, the sampling period was only from 2007 to 2010, also with four samplings per year. A total of 79 surveys were conducted from 2001 to 2010: 32 at A Lanzada, 33 at Baldaio and 14 at O Rostro. To our knowledge, our data base is one of the oldest and largest characterizing the Galician coast.

Counting the number of individual beach litter items provides the easiest and cheapest information for the formulation of management measures at all levels (linking items to sources and use), as recently indicated by a marine litter group of experts (Piha et al. 2011). It is also the most practical method, though other additional methods such as assessment of the weight of the items found (Piha et al. 2011) can be valuable.

On average we found 975±1173, 85±56 and 276±151 BLI per survey on the 100-m stretch of beach at A Lanzada, Baldaio and O Rostro, respectively. In the OSPAR pilot project on monitoring beach litter an average of 542 items per 100 metres were counted on the reference beaches (throughout the OSPAR region). On the 1-km stretch we found on average 181±174, 47±29 and 80±41 BLI per survey at A Lanzada, Baldaio and O Rostro, respectively compared with the average of 67 items found in the OSPAR pilot project.

Therefore, the number of beach litter items found on the Galician coast during the 100-m stretch surveys (mainly small items) was lower than the number found in the OSPAR area, indicating that this area is less polluted with beach litter than the average (with the exception of A Lanzada, with values twice the average). On the other hand, the number of items found on the 1-km stretch (bigger items) was quite similar to the average value and the number found at A Lanzada was nearly three times the average in the OSPAR region.

It is very difficult to compare beach litter concentrations of various coastal areas obtained from various studies with different methodologies. Nevertheless, compared with some recent studies on beach litter, the average densities on the Galician coast seem to be lower than in tourist areas of the Mediterranean (3600 BLI per 100 m; Martinez-Ribes et al. 2007), the Black sea (2400 BLI per 100 m; Topçu et al. 2013) and populated areas of Brazil (500 to 10000 BLI per 100 m; Tourinho and Fililman 2011). Comparisons with other studies in the Atlantic show that our average values are very similar to those reported for England (85 BLI per 100 m; Willoughby et al. 1997) and the US Atlantic coast (72 BLI per 100 m; Ribic 1998) and very close to those of the North Atlantic (66 BLI per 100; Barnes and Milner 2005).

As indicated in the OSPAR pilot project, there is great variation in the number of marine litter items found on reference beaches all across the OSPAR maritime area (OSPAR 2007). The reasons for this variation lie in the source (close to river mouth or major harbours), oceanographic processes and the cleaning frequency on the beaches (beaches in the major tourist destinations are cleaned nearly on a daily basis, whereas others are cleaned only once per season or even per year). This makes comparison more difficult.

Therefore, in order to evaluate the main sources, the measures to implement and their effectiveness for reducing beach litter, a regional analysis is more valuable. The local factors (rivers, ports, beachgoers, currents,) influencing beach litter quantity and type need a specific analyses that can be masked in global analysis.

The great differences between seasons (at least in the Galician region, as demonstrated herein) suggest that the values should be compared on a seasonal basis for the same region. These differences are due to several factors, but probably mainly beach cleaning during specific periods and intensified accumulation of litter on beaches caused by oceanographic processes.

It is also important to note that beach cleaning frequency is very important to interpret the statistical analyses. If surveys are made on beaches that have not been cleaned until recently, and especially after monitoring has started, the reduction could be artificially generated by the increased frequency of cleaning.

Beach cleaning can be less of a problem for surveys if it is controlled and regular, i.e. the same intensity is maintained over time. For example, if surveys are done in each season previous to the beach cleaning, it will be possible to detect trends in beach litter quantity and composition. In our case, at A Lanzada and Baldaio beaches mechanical cleaning takes place regularly (daily and weekly, respectively) during the summer tourist season. Beach cleaning at O Rostro is usually manual and takes place only once per year, at the beginning of the summer season.

A remarkable fact during the sampling period in the Galician region is the well-known Prestige tanker incident (see e.g. Albaiges et al. 2006). The Prestige broke and sank close to the Galician coast (about 42°10’N and 12°05’W, 133 nm off the coast) in November 2002. Galician beaches were cleaned during the winter and spring of 2002-2003, removing the fuel and underlying sand, mainly with the aid of shovels. In May 2003, most of the Galician beaches (98.3%) were cleaned, and cleanup activities were limited to removing tar balls that occasionally reached the beaches (Junoy et al. 2005). This incident could affect to the initial values of our study, as low values could be due to these intensive cleanup activities on the beaches. In fact, the lowest values of beach litter items in A Lanzada and Baldaio beaches in winter-autumn were observed in 2003 (see Figs 2A and B). Similar effects are expected to be observed in zones affected by incidents like this, such as the French coast after the "Erika" tanker disaster.

Plastic accounted for 63%, 47% and 38% of BLI at A Lanzada, Baldaio and O Rostro, respectively. The OSPAR survey found an average of 75% of non-degradable plastic and/or polystyrene in all the 100-m stretch surveys. Our coastal zone therefore showed a lower percentage than the OSPAR region. Similar results showing the ubiquitous presence of plastics items in the marine environment are found world-wide and indicate the predominance of plastics among marine litter (see e.g. Derraik 2002). Though the methods were not assessed to ensure that the results were comparable, the proportion varies between 32% and 88% of total beach litter (Derraik 2002).

This is a threat to humanity because plastic does not degrade but is only broken into small particles called microplastics. For example, Browne et al. (2010) found 65% of litter sampled in Plymouth, UK, was microplastic. This is also a problem of concern in the Iberian region; a recent study (Martins and Sobral 2011) reported 72% of microplastics in litter from five beaches in Portugal.

The unknown or unassigned source items represent 46%, 37% and 36% of BLI at A Lanzada, Baldaio and O Rostro, respectively. These values are similar to the average value determined in the OSPAR pilot project, around 40%.

It is also important to mention that in the OSPAR pilot project only beach litter items from the fishing sector showed a statistically significant increase on the sampled beaches. Our results for the Galician region (with more temporal coverage than the OSPAR pilot project) are similar: an increase in the fishing sector items on all the beaches. The fishing and aquaculture sector is very important in the Galician region. In Galicia 4.6% of the active population is employed in the fisheries sector, a much higher percentage than in any other European Union country (Vazquez Seijas 1998).

Beach litter can only provide some of the information necessary to supply a full picture of the total load of marine litter in the marine environment. Any interpretation of data on beached marine litter, as a reflection of the overall marine litter situation in a given area, should be made with this uncertainty in mind, especially referring to the section of our work on sources of beach litter. For example, only floating marine litter from shipping will reach the beaches; other elements of this litter will sink in the water column and accumulate on the seabed.

We should be cautious about a long-term interpretation of our results since we studied a period of only 10 years. Nevertheless, our study could be interesting for a larger geographical level, at least covering the Iberian upwelling system.

Finally, Tudor and Williams (2001) verified that the use of volunteers to conduct beach litter surveys is a reliable method, showing no statistical difference between the results of data gathered by inexperienced and experienced surveyors.

Beach litter monitoring programmes based on regular surveys of stretches of beaches can be particularly valuable for defining background values and determining trends. This is especially important in the case of descriptor 10 (marine litter) of the MSFD. These programmes are extraordinarily difficult to maintain and few survive for periods longer than a decade, like the national marine debris monitoring programme of the National Oceanic and Atmospheric Administration (Ribic et al. 2010). The OSPAR pilot project on monitoring beach litter has therefore been a successful pioneering European initiative that extended in time to the Galician coast because of the requirements of the MSFD.

Our work provides a baseline value for beach litter on the Galician coast referring to the year 2001 and will be very valuable for determining the effectiveness of the measures and policies aimed at reaching a good environmental status in this region by 2020, as defined in the MSFD.

This monitoring programme on the Galician coast is expected to continue in the near future in order to evaluate the temporal evolution in beach litter on this area. It should also be stressed that this information can be used as a basis for marine management decisions and policy development, such as in the definition of operational targets in the MSFD.

Our study confirms the importance of plastics in the marine environment, as indicated by several studies (Derraik 2002, Thompson et al. 2004). The fragmentation of plastic into small particles (not detected by the human eye) should be considered in new monitoring programmes, especially because of the slow degradability of these materials.

It is not surprising that in a region with a high dependence on fishing and aquaculture this sector represents the main source of beach litter. The contribution from the fisheries/aquaculture sector to beach litter in the studied region has increased during our sampling period. However, this does not indicate an inappropriate activity in this sector because the items are mainly lost during routine activities or during severe meteorological conditions. Nevertheless, in order to restrict litter related to the fishing and shipping industry, measures such as better schooling of fishermen are needed.

Finally, it should be pointed out that there are limited data on the overall influence that beach litter has on the environment and society, and further research is needed (UNEP 2005, Mouat et al. 2010).