**6. Discussion**

The analyses showed that the beach gravels collected from Pogorzelica to Dziwnów differed widely in their petrographic composition. The cliff sections of the beach, in labile equilibrium, showed the highest proportion of crystalline rock gravels. Their contribution declined in sections affected by erosion and in the sandbar sections. Limestone gravels dominated in active cliff areas (Niechorze–Pustkowo and Łuk ˛ecin–Dziwnówek) and in a part of the Dziwnów Sandbar, indicating increased erosion. Occasionally, their contribution was twice that found in the stable or nonactive cliff sections (e.g., cliff shore in Pobierowo). Attention must be drawn, however, to the eroded foreshore that nourishes the beach, too. Increased reworking of gravel material leaves only the most resistant gravels and their distribution along the shore may be indicative of the direction of the longshore bedload transport.

The direction of the longshore bedload transport in the shore section under study is still unclear. In the 1950s, Zenkowicz [75] pointed to west-to-east bedload transport. Subsequently, B ˛aczyk [43] noticed a divergence of the sandy bedload transport off the Pobierowo cliff, eastwards towards the

Koszalin Bay and westwards towards the Pomeranian Bay. In his studies of the shallow foreshore, Cie´slak [76] demonstrated the direction and intensity of the sandy bedload transport to be both eastward (measured at Niechorze, 368 km) and westward (measured at the River Dziwna mouth, 391.4 km). In the Niechorze–Dziwnów section, Furma ´nczyk [71] showed the locally varying directions of the bedload transport along the shallow foreshore, and associated the variation with cellular currents occasionally producing embayments, as well as larger or smaller cusps. Between Pogorzelica and Dziwnów, channel-like locations termed "underwater circulation gates" were identified [77]. They distort the bedload transport and serve as outlets allowing export of substantial amounts of material during storms.

In general, the coast sections investigated are affected by long term erosion, which can be seasonally less intensive, or limited by coastal infrastructure (so-called "unstable equilibrium"). Gravel circulation is strongly related to erosion and accumulation (direct deposition), considered as a primary removal of sedimentary material, e.g., from morainic cliff sections, and secondary accumulation directly within the cliff base or shallow nearshore, with no significant mechanical reworking patterns. Redeposition occurs due to secondary movement of this material, mainly resulting from strong wave storms. The same process may be related to erosion of shallow seashore, which remobilizes sedimentary material towards the coastline and partially rounds the gravels.

The contribution of individual petrographic groups does not allow for strictly pinpointing the dominant direction of the bedload transport. The increased proportions of crystalline gravels in the cliff area between Rewal and Łuk ˛ecin would be indicative of washing-out of this part of the erosional platform and the deposition of "fresh" gravel material in sandbar–dune areas. This is evidenced by an increased contribution of gravels representing rocks less resistant to destruction, both in the Rega and Dziwna Sandbars.

The longshore distribution of less resistant and more resistant gravels may indicate the absence of any distinct direction of bedload transport along the shore, e.g., from the southwest to the northeast or back. Shoreline armoring (breakwaters, revetments, bulkheads, groins, etc.) significantly limits the unimpeded bedload transport along the shore. Most likely, the most intensive bedload movement proceeds seaward from the shore and back, and it is then that the material is most exposed to destruction. Zones with dominance of crystalline magmatic and metamorphic pebbles may indicate increased redeposition. On the other hand, the increased amount of soft limestones or sandstones may suggest longshore bedload transport (high rate of accumulation). The greater contents of crystalline pebbles found along the cliff bench sections with developed foredunes may indicate a long period of mechanical reworking. Thus, the soft pebbles were mechanically destroyed or chemically dissolved. Additionally, the greater contents of crystalline pebbles may be related to redeposition from cliff sections or shallow foreland.

The increased proportions of discoid and ellipsoid gravel grains and higher flattening indices are indicative of the dominance of material accumulation, and were recorded mainly in the sandbar–dune areas, or in the western part of the cliff shore (378.0–386.0 km). These areas include shore sections with lower proportions of the discoid and ellipsoid gravels, indicative of intensified erosion of the beach. Most likely, this is associated with earlier gravel deposition during periods of strong storms. Shore sections with evidence of accumulation (elevated contributions of discoid and ellipsoid gravels) are separated by sections with lower proportions of these gravels. This is most likely associated with the presence of 100–500 m long cusps and embayments in the shore area discussed. Those cusps and embayments were shown by Furma ´nczyk to be structures affecting the shore dynamics [71]. Therefore, areas of increased accumulation may feature sections indicating erosion characterized by erosion–redeposition processes (instantaneous redeposition).

Increased proportions of spheroid and spindle-shaped gravels were found mainly in the eastern part of the cliff shore (368.0–378.0 km), indicating domination of erosion processes and further mechanical reworking of the sedimentary material.

Generally, the lithostatistical data provided indicate increased erosion and redeposition processes occurring in the middle part of coast section studied, which are significantly reduced in the area of Dziwna and Rega Sandbars. The variability of *Ws* index in the group of limestones indicates low rates of erosion in areas of seawalls and groins, where bedload transport is also significantly reduced, especially between 367–369 km and 387–391 km. The large number of flat and relatively mechanically low resistant gravels, such as limestones and marbles, indicate tendencies of dominating accumulation. These gravels do not experience sufficient time and energy to be rounded. Discoidal or ellipsoidal grains may indicate lower energy and shorter bedload transport.

Elevated contents of spheroidal gravels in the crystalline and limestone groups, visible especially between 370 and 375 km, may be related to increased erosion and material reworking, and disturbed by the presence of seawalls and groins. A similar situation occurs between 374 and 375 km, or in the eastern part Dziwnów embankments, where low number of spheroidal gravels and slightly increased discoidal and ellipsoidal shape indicate low accumulation. The variability of sandstones is comparable, with highest contents of spheroidal pebbles in low erosion sections, and discoidal or ellipsoidal pebbles in areas of increased accumulation.

It must be borne in mind that the coastal zone dynamics pattern presented here, as concluded from the petrographic composition and gravel grain shapes, reflects the post-early-autumn storm conditions. Generally, characteristics indicative of material deposition and redeposition contribute to the developmental trends in the part of the Polish coast studied. Additionally, the results can inform comprehensive coastal protection programs.

**Author Contributions:** Conceptualization, C.S. and T.K.; methodology, C.S.; software, C.S. and R.B.; formal analysis, C.S., R.B. and Ł.M.; investigation, C.S., R.B. and T.K.; resources, C.S. and T.K.; writing—original draft preparation, C.S.; writing—review and editing, C.S., Ł.M. and R.B.; visualization, C.S.; funding acquisition, C.S., R.B. and T.K. All authors have read and agreed to the published version of the manuscript.

**Funding:** The publication was cofinanced by Faculty of Civil Engineering and Architecture, West Pomeranian University of Technology in Szczecin, Poland.

**Conflicts of Interest:** The authors declare no conflict of interest.
