Using accurate land cover data is essential to monitor land use and cover changes and assess the effectiveness of various environmental policies. This study evaluates the accuracy of contemporary global land cover products with 10 m spatial resolution, including Google’s Dynamic World (GDW), European Space Agency’s World Cover (ESA WC) and Esri Land Cover (ELC) in mapping forested areas in Poland, aiming to test an assumption if the combination of these products may improve forest mapping accuracy compared to any individual product. Three global datasets and their combinations were assessed with the 2022 EU Land Use/Cover Area Frame Survey (LUCAS). A land cover map of Poland (S2GLC PL) for 2021 served as an auxiliary reference data set. Forest cover classification accuracy was evaluated using precision, recall, and F1-score metrics, and spatial agreement of binary forest maps in the thematic global products was measured with the Intersection over Union (IoU) at two various scale levels (country and province). Our results showed that forest mapping accuracy of three global products varies for Poland, with F1-score equal to 72.2% for ELC, 76.9% for ESA WC, and 68.8% for GDW. IoU against S2GLC PL was equal to 82.6%, 82.3% and 75.2%, for ELC, ESA WC and GDW, respectively, and slightly exceeded 70.5% for three global products. A specific combination of binary forest maps from global products, where the output forest area consisted of forests mapped at the same time by all three products and forests mapped at the same time only by GDW and ESA WC yielded better accuracy indicators than any single product and other tested combinations (F1-score equal to 80.4%, and IoU against S2GLC PL equal to 87.1%).

Deep learning models have contributed recently to significant advances in pattern recognition. In the fields of remote sensing and geoinformation, they have largely replaced labour-intensive manual workflows, allowing researchers to quickly collect accurate data over large study areas. However, when working with historical datasets, for instance old aerial imagery or maps, temporally-equivalent digital data in vector formats for labelling and training are typically not available, and the limited availability of training data often constrains model performance.. In this research, we developed and tested a method to quickly generate training data to detect buildings on historical black-and-white aerial imagery acquired in 2003 over south-eastern Poland. A large simulated training dataset was developed by [1] adjusting blur, noise and contrast in modern aerial imagery to match historical black-and-white imagery, and [2] extracting building footprints using contemporary topographic data in the vector format. Then we compared the performance of various instance segmentation models trained on the large simulated dataset against a control dataset created by simply converting modern imagery to black and white, and a small training sample retrieved manually from the historical aerial imagery. Models trained on the simulated dataset showed better performance (F1 ranging from 0.93 to 0.96) than models trained using the control dataset (F1 ranging from 0.89 to 0.96) and the small training sample (F1 ranging from 0.88 to 0.92). Finally, we demonstrated the applicability of the model in analysing settlement development in a case study area in south-eastern Poland.

This paper investigates changes in building numbers and density in Poland over the past 50 years. By combining building data from historical maps (1970s-1980s) with contemporary digital topographic datasets, we analyze changes in building numbers, distribution, and density at a national scale with high spatial resolution. We calculated a change index (Normalized Buildings Number Change Index, NBCI) within a 2 x 2 km grid covering the entire country. An additional map shows changes in building density at the commune level. The maps highlight areas with significant increases in building numbers around major cities and areas with stagnation or decreases, mostly in economically deprived regions. Some patterns reflect historical legacies related to boundary changes since the eighteenth century. Overall, the maps and data serve as valuable tools for discussions on settlement structure changes, migration patterns, and land use trends in Poland.

Accurate information on forest tree species composition is vital for various scientific applications, as well as for forest inventory and management purposes. Country-wide, detailed species maps are a valuable resource for environmental management, conservation, research, and planning. Here, we performed the classification of 16 dominant tree species and genera in Poland using time series of Sentinel-2 imagery. To generate comprehensive spectral–temporal information, we created Sentinel-2 seasonal aggregations known as spectral–temporal metrics (STMs) within the Google Earth Engine (GEE). STMs were computed for short periods of 15–30 d during spring, summer, and autumn, covering multi-annual observations from 2018 to 2021. The Polish Forest Data Bank served as reference data, and, to obtain robust samples with pure stands only, the data were validated through automated and visual inspection based on very-high-resolution orthoimagery, resulting in 4500 polygons serving as training and test data. The forest mask was derived from available land cover datasets in GEE, namely the ESA WorldCover and Dynamic World dataset. Additionally, we incorporated various topographic and climatic variables from GEE to enhance classification accuracy. The random forest algorithm was employed for the classification process, and an area-adjusted accuracy assessment was conducted through cross-validation and test datasets. The results demonstrate that the country-wide forest stand species mapping achieved an accuracy exceeding 80 %; however, this varies greatly depending on species, region, and observation frequency. We provide freely accessible resources, including the forest tree species map and training and test data: https://doi.org/10.5281/zenodo.10180469 (Grabska-Szwagrzyk, 2023a).

The Wildland-Urban Interface (WUI) is the area where natural vegetation is close to housing and area of concern due to various negative consequences for humans and the environment including fire ignitions, landscape fragmentation and human-wildlife interactions. The WUI is a global phenomenon, and widespread in many countries but long-term WUI dynamics and the main factors causing WUI growth are unknown. Our goal was to assess WUI changes in the Polish Carpathians since the mid-19th century, based on high-resolution spatial data for 1860s, 1970s and 2013. We found that WUI covered already 30% of the study area in the 1860s but grew to cover nearly half by 2013, especially at lower elevations. Detailed analysis of WUI determinants confirmed the areas closer to regional administrative centres or located on steep slopes were more WUI-prone. Tourist trail density also fostered WUI occurrence. We conclude that in Central Europe, with a long history of human settlements and agricultural activities, WUI has been a persistent landscape feature for centuries, but increased in area in recent decades due to widespread abandonment of agricultural land combined with development of new residential areas.

The aim of this study was to create a dataset of building locations in Poland from the 1970s–1980s. The source information was the historical 1:10 000 Polish topographic map. Building footprints were detected and extracted from approximately 8,500 scanned map sheets using the Mask R-CNN model implemented in Esri ArcGIS Pro software, and converted to point building locations. The dataset of building locations covers the entire country and contains approximately 11 million points representing buildings. The accuracy of the dataset was assessed manually on randomly selected map sheets. The overall accuracy is 95% (F1 = 0.98). The dataset may be used in conjunction with various contemporary land use, land cover and cadastral datasets in a broad range of applications related to long-term changes in rural and urban areas, including urban sprawl and its environmental and social consequences. It can also serve as a highly reliable reference dataset for regional or global settlement products derived, e.g., from early Landsat data.

Abstract The EU Biodiversity Strategy 2030 aims to better protect natural ecosystems with high biodiversity and climate change mitigation potential. To achieve this goal, it is crucial to identify forests worth protecting, such as those characterized by long continuity and old age. Here, we propose a robust approach that combines historical maps from the mid‐19th century with remote sensing data to identify areas that have been forested for over 150 years, aiming to reduce the need for detailed and time‐consuming field investigations in potential conservation areas. We tested this approach in a 20,000 km2 region in the Polish Carpathians, an area likely containing many unprotected forests of high conservation value, where historical maps from the mid‐19th century are representative of much of Central Europe. Our results showed that an area of approximately 4200 km2 has been continuously forested since at least the mid‐19th century. Currently, 50% of these forests are outside protected areas, representing an important conservation opportunity. More generally, our approach can support the identification of valuable forests worth protecting before they are lost and provide useful insights for policymakers and stakeholders in the ongoing debate on the challenges of improving forest conservation in the Carpathians, and Europe more widely.

CORONA spy satellites offer high spatial resolution imagery acquired in the 1960s and early 1970s and declassified in 1995, and they have been used in various scientific fields, such as archaeology, geomorphology, geology, and land change research. The images are panchromatic but contain many details of objects on the land surface due to their high spatial resolution. This systematic review aims to study the use of CORONA imagery in land use and land cover change (LULC) research. Based on a set of queries conducted on the SCOPUS database, we identified and examined 54 research papers using such data in their study of LULC. Our analysis considered case-study area distributions, LULC classes and LULC changes, as well as the methods and types of geospatial data used alongside CORONA data. While the use of CORONA images has increased over time, their potential has not been fully explored due to difficulties in processing CORONA images. In most cases, study areas are small and below 5000 km2 because of the reported drawbacks related to data acquisition frequency, data quality and analysis. While CORONA imagery allows analyzing built-up areas, infrastructure and individual buildings due to its high spatial resolution and initial mission design, in LULC studies, researchers use the data mostly to study forests. In most case studies, CORONA imagery was used to extend the study period into the 1960s, with only some examples of using CORONA alongside older historical data. Our analysis proves that in order to detect LULC changes, CORONA can be compared with various contemporary geospatial data, particularly high and very high-resolution satellite imagery, as well as aerial imagery.

Nowadays, Poland is one of the European countries most affected by agricultural land abandonment (ALA). Though considered to be a negative phenomenon, ALA opens up several options for planning future land uses critical for biodiversity conservation or future carbon sequestration. So far, many studies of ALA have been done in the mountainous areas in Poland, but less is known about the magnitude of ALA in other regions. In this paper we use the declassified CORONA satellite imagery (1969) to backdate the information on land cover and land abandonment from topographic maps from 1970s for the region located in central Poland and currently affected by widespread ALA. The information from archival materials is compared with current High-Resolution Layers and airborne laser scanning products, indicating that a forest cover increase of 23% was observed. The output of vegetation height analysis confirmed significant land use transformation from non-forest and ALA into forest area. Additionally, analysis of forest pattern change revealed that although forest core areas have increased since 1970, its share in total forest cover decreased due to newly established small forest patches. Our research shows the importance of archival remote sensing materials and indicates their role in understanding ALA-related forest cover change in Poland over the last 50 years.

E-learning is widely used in academic education, and currently, the COVID-19 pandemic is increasing the demand for e-learning resources. This report describes the results achieved and the experiences gained in the Erasmus+ CBHE (Capacity Building in Higher Education) project “Innovation on Remote Sensing Education and Learning (IRSEL)”. European and Asian universities created an innovative open source e-learning platform in the field of remote sensing. Twenty modules tailored to remote sensing study programs at the four Asian partner universities were developed. Principles of remote sensing as well as specific thematic applications are part of the modules, and a knowledge pool of e-learning teaching and learning materials was created. The focus was given to case studies covering a broad range of applications. Piloting with students gave evidence about the usefulness and quality of the developed modules. In particular, teachers and students who tested the modules appreciated the balance of theory and practice. Currently, the modules are being integrated into the curricula of the participating Asian universities. The content will be available to a broader public.

Accurate estimations of the extent of agricultural land abandonment (ALA) are critical to the sustainable management of agricultural resources and forestry, the understanding of ALA determinants, and the development of future agricultural policies. Although ALA is widespread in Europe, mapping it over large areas using remote sensing data is difficult as a result of the complexity of this phenomenon. This study aims to develop methods for a detailed wall-to-wall regional-scale mapping of ALA using vegetation height and secondary forest succession indicators. The rates and distribution of ALA were analyzed at the parcel and communal level in the Polish Carpathians using a high-resolution vegetation height model (VHM) derived from Light Detection and Ranging (LiDAR) point clouds and topographic data. Depending on the parcel-level secondary forest succession threshold (10, 20, and 50%), the regional ALA rates were 18.8, 9.0, and 2.1%, respectively. Regardless of the threshold, abandoned grasslands covered about three times more area than abandoned croplands. The highest ALA rates were observed in communes located in the western part of the study area, as well as east and south of Rzeszów. We found that areas receiving European Union Common Agricultural Policy payments very rarely showed signs of secondary forest succession and land abandonment. The developed method proved to be effective for detailed ALA mapping at various spatial scales.

Geoinformatica Polonica » Vol. 18 (2019) » Integrating contemporary spatial forest cover data in the polish Carpathians: does abundance of data increase knowledge or uncertainty? A A A

The Polish Carpathians, like many mountain areas in Europe, are currently facing dynamic land use changes that will shape their future landscapes. As there are many different possible scenarios of potential change, we compared three different land use scenarios up until the year 2060 and assessed their impact on the potential habitat connectivity of two large carnivoreswolf (Canis lupus) and lynx (Lynx lynx). We first analysed the main directions of change within and outside the pan-European wildlife corridor located in the western part of the Polish Carpathians. Then we calculated and compared least-cost paths among randomly selected points for each land use scenario separately. Our results showed that the main direction of changeforest cover increasemay positively influence habitat connectivity for both wolf and lynx. However, due to the future spread of settlements, this positive impact might be locally limited. Therefore, to realise the potential conservation opportunities resulting from on-going land use changes, adequate orientation of spatial planning towards habitat connectivity is crucial.

Flooding is a major environmental hazard in Poland with risks that are likely to increase in the future. Land use and land cover (LULC) have a strong influencing on flood risk. In the Polish Carpathians, the two main projected land use change processes are forest expansion and urbanization. These processes have a contradictory impact on flood risk, which makes the future impact of LULC changes on flooding in the Carpathians hard to estimate. In this paper, we investigate the impact of the projected LULC changes on future flood risk in the Polish Carpathians for the test area of Ropa river basin. We used three models of spatially explicit future LULC scenarios for the year 2060. We conduct hydrological simulations for the current state and for the three projected land use scenarios (trend extrapolation, ‘liberalization’ and ‘self-sufficiency’). In addition, we calculated the amount of flood-related monetary losses, based on the current flood plain area and both actual and projected land use maps under each of the three scenarios. The results show that in the Ropa river, depending on scenario, either peak discharge decreases due to the forest expansion or the peak discharge remains constant—the impact of LULC changes on the hydrology of such mountainous basins is relatively low. However, the peak discharges are very diverse across sub-catchments within the modeling area. Despite the overall decrease of peak discharge, there are areas of flow increase and there is a substantial projected increase in flood-related monetary losses within the already flood-prone areas, related to the projected degree of urbanization.

We produced the first spatially explicit, cross-border, digital map of long-term (160 years) land use in the Carpathian Ecoregion, the Hungarian part of the Pannonian plains and the historical region of Moravia in the Czech Republic. We mapped land use in a regular 2 × 2 km point grid. Our dataset comprises of 91,310 points covering 365,240 km2 in seven countries (Czechia, Slovakia, Austria, Hungary, Poland, Ukraine and Romania). We digitized three time layers: (1) for the Habsburg period, we used maps of the second Habsburg military survey from years 1819–1873 at the scale 1:28,800 and the Szatmari's maps from years 1855–1858 at scale 1:57,600; (2) The World Wars period was covered by national topographic maps from years 1915–1945 and scales here ranged between 1:20,000–1:100,000; and (3) the Socialist period was mapped from national topographic maps for the years 1950–1983 at scales between 1:25,000–1:50,000. We collected metadata about the years of mapping and map sources. We used a hierarchical legend for our maps, so that the land use classification for the entire region consisted of 9 categories at the most general level and of 22 categories depending on the period and a country.

This article examines the impact of the 1850 Austro-Hungarian customs union on production land-use outcomes. Using newly digitized data from the Second Military Survey of the Habsburg Monarchy, we apply a spatial discontinuity design to estimate the impact of trade liberalization on land use. We find that the customs union increased cropland area by 8 percent per year in Hungary between 1850 and 1855, while forestland area decreased by 6 percent. We provide suggestive evidence that this result is not confounded by the emancipation of the serfs, population growth, or technological change in agriculture.

Understanding the causes and consequences of forest-fragmentation changes is critical for preserving various ecosystem services and to maintain biodiversity levels. We used long-term (1860s–2010s) and large-scale data on historical forest cover in the Polish Carpathians to identify the trajectories of forest fragmentation. Past forest cover was reconstructed for the 1860s, 1930s, 1970s and 2010s using historical maps and the contemporary national database of topographic objects. We analyzed forest-cover changes in 127 randomly selected circular test areas. Forest fragmentation was quantified with GuidosToolbox software using measures based on a landscape hypsometric curve (LHC). Despite a general increase in forest cover, forest fragmentation showed divergent trajectories: a decrease between the 1860s and 1930s (in 57% of test areas), and an increase between the 1930s and 1970s and between the 1970s and 2010s (in 58% and 72% of test areas, respectively). Although deforestation typically involves the increasing fragmentation of forest habitats, we found that forest expansion may not necessarily lead to more homogenous forested landscape, due to complex land-ownership and land-use legacy patterns. This is both a challenge and an opportunity for policy makers to tune policies in such a way as to maintain the desired fragmentation of forest habitats.