**5. Shape Variability**

The mean values (and standard deviations) of the Zingg gravel shape index and flatness index are shown in the Table 2. The table also contains contributions of remaining gravels, but their contributions were very low (below 5%) and are disregarded in the subsequent discussion, as they are not important for insights into the beach lithodynamics.

**Table 2.** The shape measurement data of beach gravels in selected petrographic groups (*n* = 17,882; confidence interval 90%).


The beach was generally dominated by discoidal gravels, which contributed greatly (78.6%) to the sandstones, as well as the limestones (75.1%), crystalline rocks (72.1%), and other gravel (70.8%). A high content of discoidal pebbles is related to processing intensity during storm events or strong winds, when less resistant gravels are flushed out from the shallow foreshore and partially reworked in the surf zone. The lowest standard deviation (10.1%) was found in the mean contribution of crystalline gravels, and the highest in other gravel (29.7%). Ellipsoid gravels contributed most (22.4%) to the other gravel group, but with the highest standard deviation (26.2%). This is one of the reasons why this group is disregarded in the further discussion on lithodynamics. Ellipsoid gravels contributed 16.3, 10.8, and 9.3% to limestone, sandstone, and crystalline gravels, respectively. The lowest standard deviation was found for the crystalline gravels. Spheroid gravels contributed most (13.7%) to the crystalline rock gravels and least (5.9%) to the limestone gravels. Spindle-shaped gravel grains accounted for 4.6 and 2.3% of the crystalline and sandstone gravels, respectively. Figures 5, 6 and 7a,b illustrate the variability of gravel grain shape in the petrographic groups.

As the discoid gravels contributed most to all of the petrographic groups, they should form the basis for evaluating the sedimentary environment dynamics in the coastal zone between Pogorzelica and Dziwnów. The remaining data should be treated as auxiliary.

The preponderance of subsets containing discoid and ellipsoid gravels, in relation to the values within the standard deviation from the mean for the entire coastal section, is indicative of the domination of accumulation over erosion and redeposition in the individual shore sections. On the other hand, the preponderance of subsets containing spheroid and spindle-shaped grains, compared to the values within the standard deviation from the mean along the section analyzed, points to domination of erosion and redeposition over accumulation (instantaneous deposition).

The contributions of the discoid crystalline gravels above half the standard deviation occur primarily in the western part of the Rega Sandbar (364.5–366.5 km), in the area of a stabilized cliff (379.5–381.5 km), and locally in the area of the Dziwna Sandbar shore armoring (386–391 km). A similar distribution is characteristic of the crystalline ellipsoid gravels. It is only in the Rega Sandbar that higher values occur within 363.0–364.5 km, whereas a very similar distribution is found in the Dziwna Sandbar (Figure 5a,b).

**Figure 5.** (**a**) Variability of morphometric indices of mechanically high resistant crystalline gravels along the investigated coast section, (**b**) comparison with coastal dynamics and coastal infrastructure data. The broken lines indicate lack of gravels along the beach (*n* = 17,882; confidence interval 90%).

**Figure 6.** (**a**) Variability of morphometric indices of mechanically low resistant limestone gravels along the investigated coast section, (**b**) comparison with coastal dynamics and coastal infrastructure data. The broken lines indicate lack of gravels along the beach (*n* = 17,882; confidence interval 90%).

**Figure 7.** (**a**) Variability of morphometric indices of mechanically low resistant sandstone gravels along the investigated coast section, (**b**) comparison with coastal dynamics and coastal infrastructure data. The broken lines indicate lack of gravels along the beach (*n* = 17,882; confidence interval 90%).

Contributions of discoid crystalline gravels below half the standard deviation occur mainly in the eroded cliff sections, except for those parts of the cliffs characterized by labile development, and even by dune accretion in front of the cliff. Contributions of ellipsoid gravels less than half of the standard deviation are similar to those of the discoid gravels, except for the areas of the Rega Sandbar and the Dziwna Sandbar (Figure 5a,b).

Contributions of discoid limestones and sandstones accentuate the preponderance of deposition on sandy spit areas, with certain exceptions, as well as domination of deposition in the cliff section (376.0–382.0 km). The distributions of limestone and sandstone gravels are similar to those of the crystalline gravels, but more visible is their increase in the Rega and Dziwna Sandbars, as well as in the western part of the cliff, less susceptible to erosion processes (Figures 6 and 7).

Contributions of spheroid and spindle-shaped gravels above half the standard deviation in all the petrographic groups point to domination of the cliff coast erosion within 367–378 km, and to a labile equilibrium in the western part of the cliff (378–385 km), with a tendency toward stabilization. Contributions below half the standard deviation were found in the Rega Sandbar and, partly, in the Dziwna Sandbar, indicating domination of accumulation over erosion. However, local variation in development trends is visible (Figures 5–7).

Increased values of the gravel flatness index (*Ws*) provide information on the degree of gravel roundness. The values obtained indicate increased mechanical reworking of gravel grains. The *Ws* is related to domination of discoidal and ellipsoidal gravels, and additionally with increased contents of mechanically less resistant rocks. If the *Ws* value is low, it means that the gravels are dominated by rounded (spheroidal or ellipsoidal) pebbles; if it is high or extremely high, it means domination by flat nonrounded grains. The flatness index distributions along the shore are shown for different petrographic groups (Figures 5a, 6 and 7a). The crystalline rock gravels are less rounded than the less resistant limestones and sandstones. In the sandbar areas (363–368 km) and in the eastern part of the Dziwna Sandbar (387–391 km), the limestone and sandstone gravels show higher flatness index values (above the mean and half the standard deviation). In the remaining areas, the *Ws* were very low (except for a few samples). Other gravels, due to their low abundance, are disregarded, although their *Ws* values were similar to those of sandstones and limestones, with a somewhat larger standard deviation from the mean.

The *Ws* values of limestone and sandstone gravels above half the standard deviation occur on the sandbars and locally in the cliff areas, but only in those at the labile development phase or with domination of sediment accumulation. The lowest values are characteristic of erosion-dominated cliff areas. The flatness index values of crystalline gravels show no particular variability along the coastal section under study. It is only the most eroded cliff section (368–378 km) that shows lower values of the index.
