Mapping Relict Charcoal Hearths in New England Using Deep Convolutional Neural Networks and LiDAR Data

Advanced deep learning methods combined with regional, open access, airborne Light Detection and Ranging (LiDAR) data have great potential to study the spatial extent of historic land use features preserved under the forest canopy throughout New England, a region in the northeastern United States. Mapping anthropogenic features plays a key role in understanding historic land use dynamics during the 17th to early 20th centuries, however previous studies have primarily used manual or semi-automated digitization methods, which are time consuming for broad-scale mapping. This study applies fully-automated deep convolutional neural networks (i.e., U-Net) with LiDAR derivatives to identify relict charcoal hearths (RCHs), a type of historical land use feature. Results show that slope, hillshade, and Visualization for Archaeological Topography (VAT) rasters work well in six localized test regions (spatial scale: <1.5 km2, best F1 score: 95.5%), but also at broader extents at the town level (spatial scale: 493 km2, best F1 score: 86%). The model performed best in areas with deciduous forest and high slope terrain (e.g., >15 degrees) (F1 score: 86.8%) compared to coniferous forest and low slope terrain (e.g., <15 degrees) (F1 score: 70.1%). Overall, our results contribute to current methodological discussions regarding automated extraction of historical cultural features using deep learning and LiDAR.

A geoachaeological view on the Roman economy of Baetica – The case study Munigua (Seville, SW-Spain).

&lt;p&gt;The economy of Roman cities in Baetica, was largely diversified and depending on the city for example specialized in olive oil, halieutic or metallurgical production. The economy of the Hispano-Roman city Munigua (municipium Flavium Muniguense) was particularly based on mining and in the 1&lt;sup&gt;st&lt;/sup&gt; and 2&lt;sup&gt;nd &lt;/sup&gt;centuries CE Munigua was the largest producer of copper and iron in the Sierra Morena. This contribution focuses on the evaluation of soil potential for practicing agriculture and evidences for prehistoric and historic land use in the vicinity of Munigua. It aims to provide new information to the food supply strategy of Munigua and furthermore a geoarchaeological few on the diversification debate of the urban economy in the conventus Hispalensis. Applying a pedo-geomorphic approach the present study i) shows that the landscape around Munigua has the potential for an agricultural use in larger areas and would have certainly allowed a production of agricultural goods in Roman times. Additionally, the study ii) provides clear evidence for prehistoric and historic land use in region documented by multi-layered colluvial deposits and a preserved Roman hortic soil. Hence, the results pointing to a local cultivation of agricultural products as an active contribution to the food supply of Munigua. Moreover, the study provides geoarchaeological evidences supporting the concept of an economic diversification of Roman cities in Baetica province and Hispania.&lt;/p&gt;

Current and future land use and land cover scenarios in the Arroio Marrecas watershed

ABSTRACT This study evaluated historic land use and land cover changes in the Arroio Marrecas watershed (Caxias do Sul, Rio Grande do Sul state of Brazil) and simulated future land use scenarios until 2034. Spatial and temporal simulations were conducted with the Conversion of Land Use and its Effects - Small Regional Extent (CLUE-S) model. Three land use scenarios were developed to include public policies and agricultural trends in the study region for 20 years (2015-2034). Geospatial analysis of different land uses showed that areas that were originally covered grasslands and forests decreased, which can be attributed to the expansion of intensive agricultural uses, such as fruit farming/forestry and urbanization. The reallocation dynamics of land use with CLUE-S was primarily driven by the soil class (especially Typic and Humic Dystrudepts and Lithic Undorthents), with limited effects of altitude and slope. Analysis of the land use maps of the Arroio Marrecas watershed until 2015 identified major human-induced changes that were driven by expanding agricultural production and urbanization. The allocation of land use derived from the proposed future scenarios with CLUE-S showed that in this regional context, Humic Dystrudepts and Rhodic Kanhapludults were the key drivers of the allocation of agricultural expansion. Conversely, natural resource conservation was indicated to most likely occur in Typic Dystrudepts. The main limitation of this approach is the recognition of driving factors that have a high correlation with each land use as effective predictor variables.