The Impact of land use and depopulation on burial mounds in the Kazanlak Valley, Bulgaria: an ordered logit predictive model

This article uses an ordered logistic regression (logit) model to assess the vulnerability of ancient burial mounds to human activity in the Kazanlak Valley, Bulgaria. This model yields probabilities of damage to burial mounds subject to changing conditions, based on the present condition and situation of a large dataset of mounds (n = 773), as estimated through direct visual assessment. Results for the Kazanlak Valley indicate that changing land use (conversion of pasture to arable land) and depopulation or de-urbanisation (increased distance to the nearest city, town, or village) represent two anthropogenic factors that degrade burial mounds. These factors likely represent threats from ploughing related to annual agriculture, and looting fostered by the decreased scrutiny associated with remoteness. After an initial survey to acquire the requisite data, local cultural heritage personnel can use this approach to predict quickly and continuously how mound vulnerability will respond to changing circumstances, and then direct resources to the most vulnerable monuments. Unlike typical predictive modelling for cultural heritage management, use of a logit regression on a large dataset quantifies the probable impact of changing circumstances on monuments without relying on site location models, prior knowledge of specific hazards, or forecasts of future development. This approach can be applied widely, wherever sufficient observational data are available. Our results also provide a reminder that agriculture is not wholly benign, and that depopulation – not just urban sprawl – can threaten cultural heritage.