Monitoring of Algae Communities on the Littoral of the Barents Sea Using UAV Imagery

Abstract

The paper presents the results of a study on littoral algae communities along the Murmansk coast from 2021–2024. The emphasis is on fucus algae and green algae communities as the most abundant ones. For the first time, an annual monitoring of littoral algae distribution in the bays of the Barents Sea was performed using a set of methods, allowing a better understanding of the dynamics of their biomass. Unlike most classical studies, which only focus on biomass and population structure, this work shows the results of using UAV-based remote sensing in combination with traditional coastal sampling techniques. The features and limitations of this approach in Arctic latitudes are discussed. According to the monitoring results, an increase in fucus algae biomass is observed in the study area, which may be associated with an increase in summer temperatures and water salinity. Fucus serratus and Pelvetia canaliculata populations remain stable. Ulvophycean algae show seasonal peaks of development with abnormally high biomass in areas of anthropogenic impact, which may indicate local eutrophication. The map of algae spatial distribution is presented. The results are important for understanding the structure and functioning of the Arctic ecosystem and for assessing the environmental impact in the region.

1. Introduction

Under the conditions of a changing climate, as well as increasing anthropogenic pressure, tools are needed to assess the state and dynamics of coast ecosystems [1,2]. These tools should provide opportunities to solve a wide range of problems, from studying ecosystems and the patterns of their structure and functioning to tracking changes under the influence of natural and human-induced causes. One such tool is the environmental monitoring system [3]. Since the list of goals and objectives solved with the help of monitoring is very long, in scientific, environmental and related fields there is a great variety in the definition of the term monitoring, the approaches used and the objects of monitoring.

In this paper, we will consider environmental monitoring as a system of measures aimed at assessing the state of components of biological diversity, including long-term observations of ongoing changes in the state of biological diversity, assessment and forecast of the state of biological diversity.

Macroalgal communities serve as ecological indicators for monitoring and impact assessment, but the number and objectives of macroalgal monitoring programs vary greatly across geographic regions. In Europe, European Union directives have led to the development of indices as a tool for monitoring the ecological quality of coastal systems, and mapping of algal communities in general and Fucales in particular has begun [4,5]. In Russia, an international standard “GOST” has been mandatory for industrial monitoring since 2023. It indicates the need to take into account vegetation, its species diversity and biomass in connection with landscapes and when using geographic information systems [6]. Satellite data are used for remote sensing of many effects of eutrophication [7], allowing the acute response of aquatic ecosystems to be assessed. A combination of water samples and the analysis of biotic indicator species such as macrophytes can provide a more complete picture of the impact of eutrophication on the ecosystem [8]. Ulva intestinalis and Fucus vesiculosus are indicators of eutrophication when measured by their biomass, but the nitrogen and phosphorus content of their thalli, although increased by feeding, does not correlate with the nutrient content of the water [9]. A study of the relationship between the projective cover and biomass of F. vesiculosus in algae-harvesting areas, including using UAV aerial photography, showed that the higher the density of algae remaining, the faster they recover [10].

The use of unmanned aerial vehicles is well suited for monitoring and detecting ecologically important and protected brown algae and a number of other common species [11,12,13,14,15,16,17,18]. UAV images can be used to create high-resolution maps that can be used to identify canopy-forming macroalgae, benthic invertebrate patches (such as mussels) and affected areas with little or no vegetation cover. The availability of high-resolution satellite imagery (e.g., Maxar, 15–30 cm pixels) facilitates wider adoption of this tool for ecosystem studies, but technical features of satellite imaging (e.g., cloud cover, frequency of satellite passes over a particular area, etc.) impose certain limitations. The key to creating marine vegetation maps from unmanned aerial vehicles is the collection and species identification of seagrass samples with their precise coordinates and subsequent calibration of digital UAV photography data [12,14,15,19].

The Murmansk coast as a study area is characterized by extensive sandy and boulder beaches that dry out during low tide and algae thickets, primarily Fucus and ulvophycean algae, are exposed on them. Description of vegetation distribution and assessment of algae reserves in such vast areas are possible only using low-altitude aerial photography [20,21,22,23]. At the same time, aerial photography does not allow direct assessment of algae biomass. Studies combining hyperspectral aerial imagery and field collections of algae for direct weighing and species identification in the laboratory have been found to be the most productive and accurate [14,15,24,25]. Drone-based approaches are being developed worldwide and provide very high-accuracy estimates of algae distribution, for example, to estimate the distribution of the fucus alga Ascophyllum nodosum [15]. The spatial distribution of intertidal algae has traditionally been monitored using a variety of methods, ranging from underwater observations, photography and/or videography to aerial and space-based remote sensing using aircraft or satellite technologies, with the best results obtained using a combination of methods [26]. At the moment, the problem of species identification of objects from aerial photographs remains unsolved, but when assessing the distribution of species groups, for example, Fucus in general, UAV shooting gives good results. But serious problems in shooting the coastline at high latitudes are the low position of the sun, the prevalence of cloudy weather and high humidity, which significantly degrade the quality of images and require calibration and finding an adequate range of shooting spectra [27].

In the Russian Arctic, the need for monitoring is most acute for the ecosystem of the Kola Bay, since several large ports and large and medium-sized cities with industrial and ship repair enterprises are located on its shores and the traffic of ships, including tankers, is steadily growing in the water area. The second most important monitoring area should be the Teriberskaya Bay, since it is a large bay of the Kola Peninsula, on the shore of which the village of Teriberka has been located for many centuries, where the fishing industry has always flourished, and in recent years the tourist flow has been growing. In 2021, the coast received the status of a natural park, which has further increased the recreational load on the ecosystem of the Teriberskaya Bay. Yarnyshnaya Bay was chosen as an area with a typical ecosystem that has not been affected by human influence. This water area has been subjected to a high degree of study, but economic activity and tourism in this area are practically not represented. Thus, we selected three bays, the observation of which will allow us to assess the main trends of changes for the entire Murmansk coast of the Barents Sea.

The purpose of this study is to describe the current state of littoral algal communities in the Barents Sea, focusing on key areas along the Murmansk coast, including Kola Bay, Teriberskaya Bay and Yarnyshnaya Bay. To achieve this, we used modern integrated survey methods, including ground surveys and aerial photography using UAVs. We also consider the methodological aspects of this monitoring in detail.

2. Materials and Methods

2.1. The Study Area and Its Features

The object of the study is the algae communities on the Murmansk coast of the Barents Sea. This section of the coast is located on the Kola Peninsula, in its northern part, and covers areas from the Russian–Norwegian border to Cape Svyatoy Nos. Its total length along the coastline is ~ 1500 km. It is characterized by the presence of bays, inlets and coves with various hydrodynamic conditions and a wide variety of landscapes near their shores.

Taking into account the characteristics of the region, three areas were selected for monitoring studies: Kola Bay, Teriberskaya Bay and Yarnyshnaya Bay (see Figure 1). Within them, in addition, for greater detail of observations, the most typical and significant polygons were identified (

Table 1. Chronology of coastal surveys.

Area	Polygon	Research Period
		2021	2022	2023	2024
Teriberskaya bay	Lodeynaya Bay	+	+	+	+
Kola bay	Cape Belokamenny	+	+	+	−
	Retinskaya Bay	−	+	+	−
	Khlebnaya Bay	−	+	+	−
	Cape Elovy—Kola Bay bridge	−	−	+	+
Yarnyshnaya bay	Yarnyshnaya Bay	−	−	+	−

The plus indicates whether a survey was conducted, and the minus sign means that there was no survey for that period.

).

The selected areas differ fundamentally in the nature of anthropogenic impact. Kola Bay is heavily urbanized: at least 10% of the coastline is occupied by port infrastructure, three large cities and rural settlements and large industrial facilities and the waters of the bay are polluted. The anthropogenic impact has been constantly increasing throughout the development period and further development of the transport hub is planned. Littoral vegetation monitoring sites were established on both banks of the middle part and in the southern part of the bay (see Figure 1). On the shore of Teriberskaya Bay there is a rural settlement and a national park. Yarnyshnaya Bay is remote from populated areas. It can be considered as potentially important for algae harvesting. In the upper reaches of the bay there is a vast littoral dry land, which was chosen as an observation site.

The test sites were selected primarily based on landscape features. Of particular interest were the vast littoral drainage areas at the tops of the bays with complex microrelief, which combine grounds of different granulometric composition (sandy and silty areas of the gently sloping littoral, boulder ridges, individual boulders, etc.), fresh watercourses and elevation changes and other factors that form a complex picture of vegetation with different types of plant (algal) communities. For the safety of flights over the test site, an important factor was the lack of coastal infrastructure in the immediate vicinity. The logistical accessibility of the areas also imposed some restrictions on the performance of work. The Teriberskaya and Yarnyshnaya bays are remote from large cities and have restrictions on transport accessibility. In unfavorable weather conditions (in winter and to a lesser extent in spring and autumn), they can be practically inaccessible. Kola Bay, due to the active development of port and other coastal infrastructure, is logistically accessible along most of its length (except for the northern part). However, the presence of a large number of economic entities imposes a number of administrative restrictions on obtaining flight permits.

In total, 6 polygons on the Murmansk coast of the Barents Sea were surveyed from 2021 to 2024 (Figure 1, Table 1). For most sites, the survey frequency was once a year. Given the specifics of the approval procedures, as well as weather conditions and logistics, it was not possible to ensure annual monitoring with aerial observations for each site. Fucus biomass sampling was performed at all study sites each year (Appendix A). Ulvophycean algae samples were collected from March to November each year at all study sites. It is also important to note that these results are part of a larger monitoring effort performed by the authors. Information on monitoring results for some additional areas is presented in Section 3.2 and Section 3.4.

2.2. The Data

2.2.1. The Studied Species and Community

Species identification of algae was performed for Fucus algae based on morphological characteristics directly in the littoral zone and for other species in the laboratory using light microscopy. We used LOMO Mikmed-6 and LOMO MSP-2 microscopes (St. Petersburg, Russia) with a magnification range from 1 to 1000 times and a digital system of two video cameras and a laptop with LOMO Microsystem software ver. 3.0 (LOMO Microsystem, ltd., St. Petersburg, Russia) for image analysis. Species identification was performed using the corresponding algae guides for the North Atlantic and Arctic seas [28,29]. This was based on modern ideas about the systematics of algae based on the AlgaeBase database [30].

The projective cover (%) was defined as the proportion of the bottom covered by algal thalli during the low-water period. The biomass was estimated by the plot count method [31,32]. Samples were collected in triplicate using a 0.5 m × 0.5 m frame for Fucus algae and 0.1 m × 0.1 m for ulvophycean algae. Note that the samples were collected in areas with 100% projective cover, so the data are fully comparable. Sample weight was determined with an accuracy of 0.1 g. Biomass (B, kg/m²) was calculated to estimate the population density. Biomass data are presented as mean values ± 1 standard error, and a significance level of 0.05 and a trend level of 0.1 were used for all statistical tests. All data analyses were performed in PAST.

The homogeneity of the distribution of species in the littoral zone was assessed by the uniformity of the projective bottom cover, the nature of the ground and the size of individual thalli (visually on site and control measurements of averages in the laboratory) [31]. The uniformity of communities was also assessed by appearance and the distribution of associated species.

For all data sets, the normality of the distribution of data was tested using the Shapiro–Wilk test, and the homogeneity of data variance was tested using the Bartlett test. Both tests were confirmed by diagnostic plots. Post hoc analyses of statistically significant effects were performed using appropriate pairwise comparisons (Bonferroni and Tukey).

2.2.2. Aerial Photography from a UAV

Aerial photography was performed using a UAV—Mavic Air 2s (SZ DJI Technology Co., ltd., Shenzhen, China). Photography of coastal areas was performed in automated mode. The flight mission preparation was performed in the Drone Harmony application (Dübendorf, Switzerland). Depending on the terrain features and research tasks, the shooting was performed with the following parameters: camera tilt at an angle of 90°, perpendicular to the surface; UAV movement speed from 3–7 m/s; overlap between adjacent images is at least 60%; shooting height 15–200 m; camera focusing in automatic mode, shutter release—automatically, every 2 s. The take-off point and flight transect directions were chosen in such a way that the UAV moved from the shore to the water’s edge parallel to it.

The shooting altitudes were primarily chosen based on the terrain features and flight safety considerations. At some sites, the shooting was performed from different altitudes. An assessment of images detailing influence on their suitability for analyzing visual characteristics was made.

2.2.3. GIS Processing and Mapping

The resulting set of photographs was processed using Agisoft Metashape ver. 2.2.1 (Agisoft LLC, St. Petersburg, Russia). An orthomosaic was constructed according to the manual [33]. Default parameters were used. To assess the influence of topography on the distribution of algae in the littoral zone, a digital terrain model (DTM) was prepared. The Metashape toolkit was also used to generate the DTM.

For the obtained orthophotos, the total area of the littoral of the study area (S_lit) was estimated. For this purpose, the digital contour of the coastline boundary (upper tide limit) obtained from the navigation map was loaded into the GIS. For the Kola Bay, navigation maps of a 1:25,000 scale were used as a topographic basis, and for the Teriberskaya and Yarnyshnaya bays, the shoreline contours were taken from a map of a 1:10,000 scale. The application of maps with different scales is determined only by data availability. Since only the position of the coastline boundary was used, its level of detail should not have a significant impact on the result. The coastline boundary from the navigation map was loaded as a data layer on top of the orthophoto. If changes were detected visually in the orthophotos (for example, new berthing facilities), then the corresponding changes were made to the coastline boundary. The position of the water edge was determined using the orthophotos. The water surface in the photos is covered by waves or has a specific “shine” due to light reflection. These features made it easy to separate the littoral areas free of water from those covered with water. The water boundary was determined visually and contoured. The “draw line” tool in Metashape was used. Using these contours (shore and water edge), the littoral area free from water at the time of shooting was formed.

To estimate the projective cover, the area occupied by algae was determined. For all polygons, the area occupied by brown algae (S_F) was determined and, for some of them, the area occupied by green algae (S_Chl). This task was performed by processing orthophotos in the ESRI ArcGIS ver.10.8 application (Environmental Systems Research Institute, Inc., Redlands, California, U.S.). In the early stages of the work, the boundary of algae growth was contoured manually and the area of the resulting polygons was measured using standard GIS tools. This approach is very labor-intensive and is not suitable for large areas, for example, in the Yarnyshnaya Bay, in the littoral area of ~0.5 km², the number of separately growing clusters of algae is over 1 million. Moreover, the accuracy of assessment using this approach is dependent on the operator’s scientific experience. In this regard, to assess the projective cover of most areas, the values were estimated using automated tools—spatial analysis (ArcGIS ver.10.8: Spatial analyst tools—Surface—Contour) in the ArcGIS GIS application—and the results were checked by the operator for falsely identified objects (see Figure 2). The role of the operator in this process is not to measure the area values but to identify incorrectly defined objects (Figure 2a).

As a result, we obtained the values of the total littoral zone within polygon S_lit and the area occupied by Fucus communities at low tide S_F—the area where Fucus and Ascophyllum thalli are located.

2.2.4. Construction of the Kola Bay Vegetation Map

The materials obtained by the authors during previous years’ expeditions were used as a basis for constructing the plant community’s biomass map [34,35]. Our previous studies were based on the results of classical sampling methods in combination with GIS technologies. When constructing a vegetation map, we previously proceeded from the fact that the biomasses should be homogeneous when macrophytes grow in the same hydrological conditions. The depth of macrophyte growth was considered as the main factor available for analysis. Information on the depths was taken from navigation charts. When updating the vegetation maps based on the observation results for the period from 2021 to 2024 for 6 surveyed polygons, orthophotos also became available, which made it possible to clarify the previously presented results. Orthophotos were used to assess the homogeneity of the distribution of macrophytes within the littoral zone.

3. Results

3.1. Biomass of Fucus Algae

The projective cover of Fucus algae has increased significantly at all research sites. The projective cover in the Kola Bay is lower and does not exceed 60% on the rocky bottom. It is likely that the summer decrease in coastal runoff and high salinity of the water, coupled with heat waves, stimulated the growth of Fucus.

The biomass of A. nodosum has been constantly increasing (Figure 3). In 2024, it averaged 14.78 ± 2.42 kg/m², with the maximum in the Teriberskaya and Khlebnaya bays of the Kola Bay. At the same time, the biomass remains low at Cape Belokamenny of the Kola Bay.

Populations of species with low population density—Fucus serratus, Pelvetia canaliculata and Fucus spiralis—were found in 2024 in the same areas as in the previous two years, with thalli of different ages and reproducing. It can be assumed that they are in a stable state. A large number of thalli of Fucus vesiculosus of the first year of life (up to three dichotomous branches) were noted on the middle and upper horizons of the littoral. The proportion of Ascophyllum nodosum in communities and its linear dimensions increased.

3.2. Biomass of Ulvophycean Algae

In general, the polygons for 2021–2024 showed a pattern of spring development of ulvophycean algae, which is normal for the Murmansk coast [28]. The exceptions were the Gryaznaya (Figure A4) and Khlebnaya (Figure A5) bays, where an increased density of algae was detected throughout the summer of 2024.

In June and July 2024, a mass development of ulvophycean algae was observed at Cape Retinsky, and the projective cover reached 90% on the grounds not occupied by Fucus at the beginning of the month and 100% at the end. Monostroma grevillei, Ulva intestinalis and Ulotrix flacca prevailed. Kornmannia leptoderma was found on Fucus. At Cape Belokamenny, the projective cover of the littoral with green algae was 100%, and the dominants were the same. At Abram-Mys, the development of green algae was less, and the projective cover was only 40%. On the lower horizon of the littoral, the ground is silty-sandy with individual boulders of different diameters. In June, very large Ulvaria obscura and Porphyra umbilicalis grew on them. In Teriberskaya Bay at the mouth of the river, thalli of M. grevillei covered 50% of the lower and middle horizons of the littoral. In Zavalishina Bay, development of Protomonostroma undulatum was observed (projective cover 5%), in Korabelnaya Bay—Acrosiphonia arcta + Ulva intestinalis (projective cover 15%). In Pechenga Bay, P. littoralis and M. grevillei were widely developed on the littoral of the left bank. In the Ura Bay estuary, U. intestinalis + M. grevillei + Urospora penicilliformis were numerous (projective cover 40%). In Vaenga Bay, the green ones did not change their density after spring—separate accumulations of U. intestinalis, but unattached thalli of U. obscura, were numerous, probably brought in from outside. In the Khlebnaya and Gryaznaya bays, green algae covered the entire littoral, including the upper horizon, with a thick layer of several centimeters. The thalli of M. grevillei were fragile, without a specific shape, and formed a mass, holding water between the thalli at low tide.

In August, the biomass of green algae decreased significantly; in places of mass development in previous months, there were many dead thalli of plate algae (faded, fragile thalli with torn edges, attached to the ground). F. vesiculosus did not shed receptacles, but there were many Fucus sprouts in the littoral.

Observation of green algae development on the Kola Bay littoral in autumn and early winter revealed differences in the dynamics between the study areas, but there was a general tendency for vegetative thalli to disappear in autumn with cooling (Figure 4). In the development of the population, expressed as a percentage of projective cover, two maxima, spring and autumn, are clearly traced in most areas and one maximum, summer, in Khlebnaya Bay. The autumn generation of ulvophycean algae was most clearly manifested in the littoral near Cape Elovy and Cape Krestovy, in the estuary of the Kola Bay. The greatest development of ulvophycean algae was observed in Khlebnaya Bay and in the area of Cape Belokamenny—almost complete coverage of the bottom throughout the summer and increased coverage in autumn with high biomass (Figure 4 and Figure 5). In the spring, small accumulations were noted in the same areas. Two generations are typical for the Murmansk ulvophycean algae [28]. But the autumn one is larger than the spring one this year, probably due to high air and water temperatures. The reasons for the high density of ulvophycean algae in two areas are still being clarified. But there is reason to believe that the blooming zones are local and do not pose a threat to the bay ecosystem.

3.3. Detailing of Aerial Photos for Littoral Research

Due to the significant volume of remaining methodological issues, this work does not pretend to be a full-fledged methodology for conducting aerial photography for littoral studies. Below are described the results and analysis of the experience of using aerial photography to study the spatial distribution of algae.

An important aspect when using aerial photography to identify and measure objects is the level of detail in the final results. The possibilities of using images can vary from forming a general picture of algae distribution within the littoral to identifying individual sections of the thallus (see Figure 6). The level of detail in images during remote sensing (including aerial photography) is usually expressed by the ground sample distance (GSD), presented in cm/pix. This value links the characteristics of the camera and the flight altitude [36].

The GSD value can be changed (the lower the value, the more detail). The value is affected by: decreasing the altitude, increasing the camera resolution, increasing the focal length [37]. In graphical form, the relationship between the flight altitude and GSD can be characterized by the chart shown in Figure 7. The figure shows the calculated values for the DJI Mavic Air 2S camera. Values for other altitude ranges, or for other UAVs, can be obtained using a GSD calculator, for example, [38].

GSD strongly influences visual recognition of objects. We used the same UAV in our study, so camera characteristics remained the same. GSD depends solely on the altitude of flight. Figure 7 shows that GSD grows slowly in the height range from 1–10 m, then continues to grow in the range 10–50 m and accelerates significantly after 50 m. This is likely due to the discreteness of values considered. In practice, images with a GSD of ~4 cm/pixel are suitable for assessing the overall distribution of relief features and bottom vegetation. Images with a GSD ~2 cm/pix are good for projective algae coverage and comparing it with interannual dynamics to identify individual algae associations; and images at a GSD ~1 cm/px allow for describing species composition.

3.4. Green Algae Distribution Analysis Based on Aerial Photo

The detailing level of orthophotoplans allows identification of macrophyte associations by their characteristic appearance. Fucus and Laminariaceae can be identified in low-altitude photographs. Two ulvophycean groups are typical for phytocenoses of the Murmansk littoral: Ulva intestinalis, which tends to be found on coarse-grained sands, is easily identified from clusters dominated by Monostroma grevillei by its lighter and brighter shade of green (Figure 8). GSD values of 10 cm/pix and higher are suitable for assessing the distribution boundaries of macrophytes in the littoral (flight altitudes of 100–200 m); species differentiation of algae is possible with GSD values of 0.5 cm/pix and higher (flight altitudes of 1–10 m).

The ability to identify the type of algal community, i.e., the plant association, by photos depends primarily on the lighting conditions.

Lighting conditions in polar latitudes significantly limit aerial photography in different seasons. During the polar night, in the absence of natural light, shooting is impossible, while during the polar day, shooting is possible at night. Wind direction and speed did not pose significant limitations during the work. Despite the wind speed restrictions of up to 10 m/s specified in the equipment documentation (DJI Mavic Air 2S), the work was not interrupted by gusts of up to 15 m/s, subject to additional safety measures.

Table 2. Distribution of Ulvophycean algae clusters at two polygons.

Polygon	Year	Slit, m2	SChl, m2
Cape Elovy	2023	250,222	78,255.91
Cape Elovy	2024	227,425	7078.11
Cape Belokamenny	2022	117,509	22,154.39
Cape Belokamenny	2023	103,311	15,765.30

S_lit: Area of surveyed littoral zone, m². S_Chl: Area occupied by green algae, m^2.

shows the results of green algae biomass assessment at two polygons in the Kola Bay. Ulva intestinalis was dominant, covering the bottom with a dense layer of intertwined thalli of different ages and sizes. The biomass during the period of maximum development (June–July) reached significant values, even 2926 ± 297 g/m² was recorded (Cape Elovy in the southern part of the bay, 2024). The example (Table 2) shows that the total biomass of U. intestinalis in the littoral of the Kola Bay can be quite significant and it is important to take it into account in environmental measurements.

3.5. Distribution of Vegetation Biomass in the Kola Bay

The biomass of fucoids in the Kola Bay is distributed along the coastline quite unevenly (Figure 9). The biomass in different areas fluctuates from 0.1 to 6 kg/m². The map shows the average biomass of algae across the entire width of the intertidal zone, although F. vesiculosus and A. nodosum have biomass of up to 25 kg/m² at the average horizon of the littoral. The main mass in the littoral zone consists of the species Fucus vesiculosus, Fucus distichus, Fucus serratus and Ascophyllum nodosum. They account for 80–100% of the total mass. Distribution depends primarily on the landscape. The most favorable conditions for fucoids are formed in closed bays with boulder beaches. Such beaches are concentrated in the middle part of the bay near the eastern shore and in the northern part near the western shore.

4. Discussion

To assess stocks and changes in communities, it is necessary to combine classical hydrobiological samples and aerial photography. Species identification of ulvophycean algae is only possible in laboratory conditions [28], as well as for many other species, but the assessment of the projective cover and the area occupied by a certain community is much more correct when using orthophotos.

The advantages of aerial photography are illustrated by the example of assessing the distribution of biomass in the littoral zone. Samples are taken using classic methods to obtain a number of data points on the coast with known values, but the number of samples is limited due to labor intensity and time constraints caused by tidal cycles. To estimate the values for the entire coastline of a specific area (bay or inlet), it is necessary to convert the data from the points into area data and extrapolate values from a number of sample points to the whole area under study. An approach to this problem assumes that biomass is homogeneous within areas with similar hydrological conditions, which is based on navigation charts and field studies, but this approach has limitations and can lead to inaccurate estimates of biomass reserves. Aerial photography data can clearly show heterogeneity and mosaic distributions of algae within the littoral zone.

Figure 10 shows a “view” of the littoral using different data sources. Figure 10A displays a fragment of the navigation map, Figure 10B shows an orthophoto fragment and Figure 10C presents a photograph of the site overview from the nearest elevation. Numbers 1 and 2 show examples of sites with different biomasses of macrophytes on the littoral. It is evident that on the navigation map (Figure 10A) the entire littoral area is uniform, while in the orthophoto (Figure 10B) a mosaic distribution of macrophytes is visible. Observation from an elevation (Figure 10C) also gives a good view of the macrophyte distribution heterogeneity within the littoral, however, a regular photo does not contain spatial data that allows one to identify the boundaries of growth and plot them on a map. Thus, in our opinion, orthophotos make a significant contribution to the study of the spatial distribution of macrophytes. It is worth noting that the use of satellite images does not always give the desired results. Limitations of satellite images include the distance from the surface being photographed and the lack of the ability to select the time of accurate shooting (photos are needed during the period of low water).

The next stage of the analysis should be the construction of species vegetation maps. Species identification is based on a combination of morphological, anatomical and reproductive features, as well as their growth form and habitat preferences. Key features include thallus shape and size and reproductive strategies. The list of visual differential criteria for ulvophycean and Fucus algae that we obtained is presented in Appendix A 

Table A3. Identification of common algae species.

Taxa	Species Characteristics	Features Identified from Aerial Photographs
Genus Ascophyllum, Fucus, Pelvetia	Easily identified by external morphological characteristics—location of air bubbles, shape of receptacles, shape of thalli branches	Fucus species are clearly recognizable in photographs taken from low and ultra-low altitudes. It is possible to distinguish species characteristics in high-resolution images from ultra-low altitudes Except F. spiralis, rare species, which requires microscopy of receptacles
Saccharina latissima, Alaria esculenta, Laminaria digitata	Reliably identified by the shape of the thallus. It is possible to confuse L. digitata with L. hyperborea; they differ only in adulthood in the presence of mucous canals in the thallus. But the second species is not found in the littoral zone on the Murmansk coast	They are easily identified in photographs even from low altitudes, but they grow on the lower horizon of the littoral zone and are not hidden under water only at spring tides
Palmaria palmata	Specific palm shape and bright green or crimson color of the thallus	Well recognizable in photographs from ultra-low altitudes
Ulva intestinalis, Ulva prolifera	Large forms are easily recognized by their tubular, corrugated, weakly branched, brightly colored form, but species are reliably distinguished from each other only by the shape of their cells; microscopy is required. Young thalli can be confused with other types of green algae	The accumulation of thalli of these species is well recognizable by their specific bright green color and their association with sandy soils. But reliable results can be obtained in conjunction with laboratory sample processing
Pylaiella littoralis, Ectocarpus siliculosus, Phleospora brachiata, Pylaiella varia and some other Acinetosporaceae and Ectocarpaceae	Outwardly, they all look similar and are easily distinguished from other groups but are distinguishable from each other only upon microscopic examination. It is necessary to evaluate the shape of chloroplasts, unicellular and multicellular sporangia, branching	Distinguishable in photographs from ultra-low altitudes if they form large clusters. May interfere with Fucus analysis
Ulvaria obscura, Monostroma grevillei, Ulva lactuca	In most cases, they can be identified by the shape of the thallus and density, but often verification is required by the shape and size of the cells, the number of cell layers	In aerial photographs, only clusters can be distinguished and identification is impossible
Genus Ulothrix, Urospora, Rhizoclonium, Cladophora, Rama, Acrosiphonia	Identification using microscopy only	Poorly visible

.

The advantages of parallel use of hydrobiological methods and GIS are as follows: there are no distortions in assessing the projective cover caused by complex microrelief that impedes visibility; it provides the ability to accurately describe the distribution of species by altitude, vegetation zonation and other parameters; determination of the total biomass of algae; it provides the ability to assess the dynamics of algae reserves and their distribution.

An annual assessment of the state of Fucus algae in several areas along the coast allowed us to identify a general trend towards increased biomass. In the period 2023–24, there was an increase in the spread of values for biomass. Since the leading factors are rising temperatures and increased storm frequency, these opposing forces create complex pattern of population dynamics along the Murmansk coast. Using the analytical reports of the Murmansk Hydro-Meteorological Center [39] it is clear that the increase in biomass is closely linked to positive temperature anomalies in spring (Figure 11). Official reports from the Hydromet Service of Russia indicate that 2021 had moderate warmth, with a 1 °C excess of average annual temperature compared to the climatological norm, while April and June stood out as particularly warm. The year 2022 also had warmth (excess of 1.4 °C), while 2023 was moderate (0.7 °C), and 2024 saw negative temperatures in April and May, with temperature anomalies. A. nodosum is particularly sensitive to changes in spring water temperatures. An increase in its biomass at the northerly edge of its range has been noted in other areas with warming [40], as well as significant growth in the entire thalli in warmer years, which significantly increases biomass and thallus size [41]. The subarctic littoral community structure is quite stable, and climate change is not expected to cause significant changes, as shown by studies in Greenland [42]. These findings emphasize the importance of annual monitoring of Fucus thickets along the Murmansk coast.

Fucus species along the Murmansk coast begin to actively grow in March, and samples are collected annually to assess biomass in July. The monthly average temperature in Figure 1 is therefore related to this time period.

According to the literature, the closely related species Ulva prolifera is the most common species in large “green tides” [43]. In our studies, we identified these two morphologically very similar species only by the features of their cellular structure according to [28]. There is information about the mass reproduction of U. intestinalis, for example, on the Romanian coast of the Black Sea [44]. However, it has been proposed to consider the mass of green algae not only as a dangerous phenomenon but also as a carbon utilizer [45,46]. The mass development of Ulva algae revealed in this study and its consequences for the health of the Kola Bay ecosystem have yet to be assessed, since this will require additional measurements of the nutrition and chemical composition of the algae.

In total, all the data collected made it possible to accurately describe littoral phytocenoses at a level previously unavailable. Monthly surveys during the growing season probably allow tracking the growth of Fucus algae, depending on microconditions, which was previously an impossible task for algae scientists. Similar work—mapping of bottom vegetation using GIS—has been successfully performed, for example, in the Black sea [47], and the practical significance of this data has been demonstrated.

Methodological Recommendations for Organizing Monitoring

Our research experience allows us to recommend conducting monitoring with aerial photography at least once a year, in the summer. Images with a GSD of no more than 4 cm/pixel (shooting altitude of 100–150 m, for DJI Mavic Air 2S) allow obtaining orthophotos for assessing the general distribution of relief forms and bottom vegetation.

Images with a GSD of no more than 2 cm/pixel (shooting altitude 45–75 m for DJI Mavic Air 2S) allow obtaining orthophotos for the algae projective coverage and comparing it in interannual dynamics with the identification of individual algae associations.

Images with GSD no higher than 1 cm/pixel (shooting altitude 10–35 m for DJI Mavic Air 2S) allow obtaining orthophotos for describing species composition and distribution by heights.

5. Conclusions

The studies of littoral algae communities of the Murmansk coast from 2021–2024 made it possible to estimate the areas occupied by various plant associations at the polygons and determine the algae’s projective cover and its interannual dynamics. It is recommended to conduct annual surveys at heights of ~100 m for monitoring and for individual areas with high species diversity at 30 m or less to obtain data on the general nature of the landscape, projective cover and species composition of phytocenoses.

1. Effectiveness of the combined approach: The combination of UAV aerial photography, GIS analysis and traditional hydrobiological methods allowed us to achieve high accuracy in assessing the distribution, biomass and dynamics of macroalgae. This approach is especially important for hard-to-reach Arctic regions, where field work is limited by logistics and climate. Automated processing of orthophotos (e.g., in ArcGIS) reduced the labor intensity of mapping large littoral areas but requires manual verification to minimize errors (primarily in species identification).

2. Phytocenosis dynamics: An increase in the biomass of Fucus algae (especially Ascophyllum nodosum and Fucus vesiculosus) was recorded for the period 2021–2024, which may be associated with an increase in summer temperatures and water salinity. The populations of Fucus serratus and Pelvetia canaliculata remain stable.

Ulva intestinalis and Monostroma grevillei algae demonstrate seasonal peaks of development with abnormally high biomass in areas of anthropogenic impact (the middle part of the Kola Bay), which may indicate local eutrophication.

3. Limitations and prospects: High-latitude weather conditions (cloudiness, low sun) reduce data quality. The solution is to calibrate spectral ranges and shoot during the polar day. To accurately assess biomass, it is necessary to develop AI algorithms that link UAV data with field measurements (e.g., through machine learning based on RGB indices). The resulting vegetation maps can be integrated into environmental management systems of Arctic ports to monitor pollution and assess ecosystem sustainability.