Konrad Wróblewski

Urban environmental monitoring increasingly requires methods that translate Earth observation (EO) data into spatial units relevant to ecosystem-service assessment and planning. This study develops a landscape-unit framework for smart monitoring of urban cooling in two Polish Functional Urban Areas (FUAs): Zielona Góra and Gorzów Wielkopolski. Landsat-derived land surface temperature and surface urban heat island (SUHI) intensity were integrated with vegetation, water, imperviousness and NDVI indicators, together with the characteristics of directly adjacent landscape units. Spearman correlations, ordinary least squares and Spatial Error Models (SEMs) were used to diagnose the spatial structure of thermal conditions. The SEM variants removed residual spatial autocorrelation and showed the strongest full-sample fit, with RMSE values of approximately 0.85 SUHI units and MAE values of approximately 0.66; these statistics are used as model-fit diagnostics rather than as measures of out-of-sample predictive performance. The results were translated into a continuous weighted HeatCool Index and seven Jenks natural-break classes that represent the relative capacity of landscape units and their neighbourhoods to support cooling or reinforce warming. The classification is interpreted as an ecosystem-service screening product rather than as an independent validation of its component variables. Independently modelled annual and seasonal NO₂ indicators were then aggregated to the same landscape units as a complementary pressure layer. After EDM correction, the NO₂ model achieved R² = 0.77, RMSE = 3.98 µg m⁻³ and MAE = 2.36 µg m⁻³ on the held-out validation subset. Across the HCI gradient, Classes 1–2 showed the lowest and least variable annual NO₂ conditions, whereas Classes 6–7 exhibited the strongest winter deterioration and Class 7 contained the highest annual NO₂ concentrations. Because NO₂ was not used to construct the HCI, these gradients represent an independent comparison of co-occurring thermal and air-pollution pressures rather than evidence of a causal relationship. The proposed framework extends pixel-based thermal assessment by integrating landscape-unit classification, neighbourhood context and independently modelled air-pollution pressure within a common planning-oriented spatial framework.

Warsaw experiences significant urban heat island (UHI) effects driven by low-albedo surfaces and urban geometry, which pose ongoing challenges for public health and climate adaptation. This study evaluates daytime SUHI intensity at satellite acquisition time across the entire city to provide a comparison of two acquisition dates during heatwaves. Utilizing Landsat 7 and Landsat 9 satellite imagery from July 2015 and July 2022, the research compares six distinct SUHII calculation methods, including spectral indices, statistical normalizations, and area-based temperature differences, as minimum SUHII values differed significantly between the two observations (shifting from approximately −13.8 °C to −7.7 °C). This indicates that suburban areas can become thermally similar to the city due to rapid land conversion and decreased evaporative cooling of vegetation during severe heat. Average intensities calculated via the SUHII 4 method reached 5.80 °C in 2015 and 2.07 °C in 2022. Under the criteria considered in this case study—interpretability, explicit physical units, treatment of water bodies, data requirements, and spatial consistency—SUHII 4 was the most suitable of the six tested formulations for the Warsaw analysis. Conversely, dimensionless spectral indices and purely statistical approaches are not recommended due to interpretative limitations.

Remote satellite observations have played a crucial role in monitoring vegetation since the 1970s, starting with the development of the Normalized Difference Vegetation Index (NDVI) by Rouse (1974) and Tucker (1979). Despite advances in satellite technology, validation with in situ measurements, which are often locally sparse, remains essential. Areas such as grasslands and wetlands, which are vital for CO ₂ balance and water quality, require special attention. Within the GrasSAT project, in situ measurements were conducted in Poland and Norway, and included LAI, soil moisture, and biomass. This article focuses exclusively on in situ studies carried out in Poland, and presents results related to models operating under Polish environmental conditions, highlighting the importance of local factors in the context of comparing satellite and ground data. Different sampling methods, such as linear transect and quadrat methods, were considered. The research aimed to understand and improve the consistency between satellite and ground data, which is crucial for accurate vegetation models.

This article discusses the topic of the use of application maps in precision agriculture (PA), particularly in the context of grassland management, which accounts for over 21% of utilised agricultural area (UAA) in Poland. New technological developments in the area of smart agriculture (Precision Agriculture, Agriculture 4.0), in terms of sensor technology and information processing, are creating a wide range of data acquisition opportunities to document biological production processes with both high temporal and spatial resolution. That information can be used to rationalise production processes and reduce trade-offs between different environmental services. The technologies that support this kind of research are analyses using satellite imagery, and map-based applications like the system developed in the GRASSAT project are discussed in detail in this article. The developed application provides farmers with information on events using free data from the Copernicus Programme (Sentinel-1, Sentinel-2, ERA5-Land reanalyses). Remote sensing indices, such as the Normalised Difference Vegetation Index (NDVI), Leaf Area Index (LAI), and fresh biomass production volumes, are calculated to show the condition of the green vegetation in the grassland plots. Meteorological risks, such as field freezing, are also presented. The GRASSAT application is available in both desktop and mobile versions.

The aim of this study was to analyze and compare evapotranspiration estimates obtained from different data sources over grassland regions in central-western Poland during the vegetation seasons in the years 2021 and 2022. The dataset provided includes evapotranspiration (ET) estimates derived from three sources: (1) evapotranspiration measurements from the ECOSTRESS satellite; (2) evapotranspiration estimates calculated using the energy balance method based on ERA5-Land meteorological data with land surface temperature (LST) from MODIS; and (3) evapotranspiration estimates with meteorological data derived from ground measurements replacing ERA5-Land data and using MODIS LST for the surface temperature. For the second and third sources, where the energy balance method (Penman–Monteith) was applied, the data used for the ET calculation were obtained from the nearest ground-based meteorological station to the test fields, with the most distant fields being up to 40 km away in a straight line. In addition, for comparison, the MOD16 global evapotranspiration product was added. In a study conducted in the central-western region of Poland, specifically in Wielkopolska (NUTS2–PL41), 18 grassland plots ranging in size from 0.36 to 21.34 ha were studied, providing valuable insights into the complex relationships between environmental parameters and evapotranspiration processes. The evapotranspiration derived from different sources was tested by applying correlation with soil moisture and the height of the grass obtained from ground measurements. It was found that the evapotranspiration data derived from ECOSTRESS had the best correlation with soil moisture (r = 0.46, p < 0.05) and the height of the grass (r = 0.45, p < 0.05), both of which were statistically significant. The values of the ground measurements (soil moisture and vegetation height were considered as verification for the evapotranspiration precision). In addition, the information about precipitation and air temperature during the time of measurements was considered as the verification for the evapotranspiration conditions. Comparisons between ECOSTRESS data and other sources suggest that ECOSTRESS measurements may offer the most precise estimates of evapotranspiration in the studied region.

In this study, the emphasis is on assessing how satellite-derived vegetation indices respond to drought stress characterized by meteorological observations. This study aimed to understand the dynamics of grassland vegetation and assess the impact of drought in the Wielkopolskie (PL41) and Podlaskie (PL84) regions of Poland. Spatial and temporal characteristics of grassland dynamics regarding drought occurrences from 2020 to 2023 were examined. Pearson correlation coefficients with standard errors were used to analyze vegetation indices, including NDVI, NDII, NDWI, and NDDI, in response to drought, characterized by the meteorological parameter the Hydrothermal Coefficient of Selyaninov (HTC), along with ground-based soil moisture measurements (SM). Among the vegetation indices studied, NDDI showed the strongest correlations with HTC at r = −0.75, R² = 0.56, RMSE = 1.58, and SM at r = −0.82, R² = 0.67, and RMSE = 16.33. The results indicated drought severity in 2023 within grassland fields in Wielkopolskie. Spatial–temporal analysis of NDDI revealed that approximately 50% of fields were at risk of drought during the initial decades of the growing season in 2023. Drought conditions intensified, notably in western Poland, while grasslands in northeastern Poland showed resilience to drought. These findings provide valuable insights for individual farmers through web and mobile applications, assisting in the development of strategies to mitigate the adverse effects of drought on grasslands and thereby reduce associated losses.

This study aimed to compare remotely sensed data with in situ data using the AquaCrop simulation model for accurately monitoring growth conditions and predict grassland biomass in the north-eastern and central-western regions of Poland from 2020 to 2022. The model was calibrated using input data, including daily climate parameters from the ERA5-Land Daily Aggregated dataset, crop characteristics (initial canopy cover, maximum canopy cover, and harvest index), and soil characteristics. Additionally, parameters such as the leaf area index (LAI), soil texture classes, and plant growth stages were obtained through field campaigns. The grassland’s biomass simulation results indicate that the root mean square error (RMSE) values for the north-eastern region ranged from 0.12 to 0.35 t·ha⁻¹, while for the central-western region, they ranged from 0.07 to 0.12 t·ha⁻¹. Overall, the outcomes obtained from Sentinel-2 data perform comparably to the in situ measurements, and in some instances, even yield superior results. This study contributes valuable insights into grass production management on farms, providing essential information and tools for managers to better understand grass growth and development.