Badland systems in Turkey: A holistic approach to understand the formation, controlling factors and geomorphologic characteristics

Abstract

Badlands are extremely rugged, outstanding landscapes that can be seen in all ice-free climate regions over erosion-susceptible unconsolidated materials, and they have drawn attention with their spectacular and iconic. The distribution of badlands is primarily influenced by the occurrence of loose and unconsolidated rocks (such as marl, sandstone, mudstone, etc.). These lithological units, in conjunction with various environmental elements such as climate, tectonics, vegetation, and topography, collectively shape the appearance and dynamics of badland landscapes. Since badlands have the largest sediment flux and erosion rates, they are considered to be erosional hot spots. However, the morphological characteristics, origins, regional characteristics, and development processes of these badlands are not well known. Unlike nearly all badland studies conducted at the experimental site and watershed scale, the broader-scale evaluation has been neglected in the analysis of badland distribution, characteristics, and dynamics. The first part of the thesis provides an integrative new insight into badland landscapes by investigating the distribution, characteristics, and controlling factors of Turkish badlands on a broad, regional scale. Turkish badlands were inventoried using aerial imagery and studied their distribution using K-means clustering, an unsupervised machine learning algorithm, based on a set of major conditional geo-environmental factors that control the regional distribution and characteristics of badlands, including tectonics, lithology, topography, climate, and vegetation. Here, a total of 4494 km2 of badland areas were identified which are non-uniformly distributed across Turkey, substantially clustered in the Central Anatolian Plateau (CAP). According to regional analyses, a total of five badland regions have been determined comprising three major types classified as Semi-arid, Mediterranean, and Montane (humid), together with two transitional types in-between the Semi-arid and Montane badland regions. The results indicate that temperature seasonality, mean annual precipitation, and precipitation seasonality are predominantly assigned to the badlands clusters. The clastic rocks are revealed as the most crucial and inevitable factor for the development of Turkish badlands, which are represented in a wide geologic time-scale (Cretaceous to Quaternary) and diverse lithological units (i.e., lacustrine, volcaniclastics, and terrestrial). Neogene and Paleogene terrestrial clastics (77 %) constitute the majority of the lithologic settings of these badland landscapes. The active and complex tectonic history of Turkey has portrayed the fundamental frame of the identified badland regions, by providing a susceptible environment (i.e., development of sedimentary basins) and promoting badland development through successive base-level changes. Furthermore, tectonically-modulated (i.e., formation of orogenic belts, and uplifting of CAP) climate dynamics outline the distribution pattern and differentiation of the regional characteristics of badlands in Turkey. Overall, the regional-scale approach to badland mapping and regional synthesis may decipher not only the tectonic and climatic conditions of the identified badlands regions but may also contribute to the implementation of future effective strategies for the detection and mapping of erosion-susceptible and high sediment flux areas in very broad spatial contexts of similar unexplored territories. An ever-present intricate association between weathering and erosional processes is largely controlled by climate disparities. Weathering as a predisposing process for natural hazards, landform evolution, and sediment mobility hosts key uncertainties in our understanding of how climate controls differential weathering types and rates. Here, the second part of thesis shows an approach to test the hypothesis controlling of weathering trends with the influence of seasonal characteristics of precipitation and temperature, which has been not well understood in badland landscapes, yet. Previous studies have adopted a rainfall simulation approach either in the field or laboratory, which is essential but is also limiting in that only steady climate conditions can be monitored in understanding weathering dynamics. In this context, by taking advantage of the laboratory environment to accomplish the aim of this part, the climate settings were simulated by exposing samples collected from four different climates and sedimentary environments previously defined as characteristic badland landscapes in Turkey to the seasonal precipitation and temperature conditions in twelve sequential seasons mirroring 3-years in order to understand the weathering rebounds. The pH, electrical conductivity (EC), ion, and surficial changes have been used as chemical and physical proxies, respectively, in referring to certain types and trends of weathering. Based on the incontrovertible influence of sediment physicochemical properties (especially high SAR values) on weathering processes, the findings reveal that sinusoidal trends attaining their peak level in spring in Na+, which overcomes other ions, contribute to an accelerated dispersion degree with concurrently decreasing pH in marly sediments in the arid region. Additionally, the recurrent increase pattern of Ca2+, particularly in winter seasons, can enhance the extent of sediment aggregations in Mediterranean sandy mudstones. In conclusion, consistent with previous studies, wetting–drying cycles are crucial in physical weathering and regolith behavior, which resulted in cyclic deep crust formations in the spring and summer seasons due to the higher swelling capacity of samples. Overall, this study demonstrates how seasonal changes in climate regulate the degree of chemical and physical weathering processes in badland landscapes. The combination of bedrock weathering, climate seasonality, and the controversial contributions of hillslope and river erosion processes, as part of the earth's dynamic systems, is what primarily develop badland environments. Although modern definitions make it clear that gully channels dominated by overland flow, mass movements dominated by gravitational processes, and piping driven by subsurface processes are all necessary for badland initiation, the questions remain to be tackled regarding their topographic position in the landscape. The third and last part of thesis focused on this problem by comparing Turkish badlands and globally known badland sites in order to understand the geomorphometric and topographic imprints of badlands. As a result, the topographic character of badlands at the sub-catchment scale may provide an adequate example of a transitional domain from a diffusive erosional process to a fluvial erosion process.