Search Results (18)

Nighttime near-surface air temperature is a critical input for ecological, hydrological, and meteorological models and the Moderate Resolution Imaging Spectroradiometer (MODIS)-derived instantaneous nighttime near-surface air temperature (T_a) and daily minimum temperatures (T_min) can provide spatially continuous monitoring. The MOD07 Level-2 and MYD07 Level-2 atmospheric profile product provides air temperature at various altitude levels, facilitating a more direct estimation of T_a and T_min. However, previous validations mainly focused on daytime, with a lack of validation for nighttime. Therefore, this study comprehensively evaluated the MOD07 Level-2 and MYD07 Level-2 derived T_a by 2168 hourly meteorological measurements over 5000 m altitude spanning in China. Furthermore, a detailed evaluation of their capability to estimate T_min was also compared with MOD11 Level-2 and MYD11 Level-2 land surface temperature. Our results show that the highest available pressure method (HAP) estimated that, in instantaneous nighttime T_a, there was severe underestimation especially in high-altitude areas for both MOD07 (r = 0.95, Bias = −0.27 °C, and RMSE = 4.53 °C) and MYD07 data (r = 0.96, Bias = −0.17 °C, and RMSE = 3.73 °C). The adiabatic lapse rate (ALR) correction effectively reduced these errors, achieving optimal accuracy with MYD07 data (r = 0.97, Bias = −0.05 °C, and RMSE = 3.29 °C). However, the underestimation by the HAP method was still insufficient compared to T_min estimation by land surface temperature (LST). The LST method offers improved accuracy (r = 0.98, Bias = −0.16 °C, RMSE = 2.89 °C). In general, MYD-based estimations consistently outperformed MOD-based estimations. However, seasonal and elevational variability was observed in all methods, with errors increasing notably in mountainous areas (RMSE rapidly increases to 5 °C and above when the altitude exceeds 2000 m). These findings can provide practical guidance for selecting appropriate inversion methods according to terrain and season and support the development of more accurate air temperature products for a range of climate- and environmental-related applications. Full article

Near-surface air temperature (T_a) estimation by geostationary meteorological satellites is mainly carried out under clear-sky conditions. In this study, we propose an all-weather T_a estimation method utilizing FY-4A Advanced Geostationary Radiation Imager (AGRI) and the Global Forecast System (GFS), along with additional auxiliary data. The method includes two neural-network-based T_a estimation models for clear and cloudy skies, respectively. For clear skies, AGRI LST was utilized to estimate the T_a (T_a,clear), whereas cloud top temperature and cloud top height were employed to estimate the T_a for cloudy skies (T_a,cloudy). The estimated T_a was validated using the 2020 data from 1211 stations in China, and the RMSE values of the T_a,clear and T_a,cloudy were 1.80 °C and 1.72 °C, while the correlation coefficients were 0.99 and 0.986, respectively. The performance of the all-weather T_a estimation model showed clear temporal and spatial variation characteristics, with higher accuracy in summer (RMSE = 1.53 °C) and lower accuracy in winter (RMSE = 1.88 °C). The accuracy in southeastern China was substantially better than in western and northern China. In addition, the dependence of the accuracy of the T_a estimation model for LST, CTT, CTH, elevation, and air temperature were analyzed. The global sensitivity analysis shows that AGRI and GFS data are the most important factors for accurate T_a estimation. The AGRI-estimated T_a showed higher accuracy compared to the ERA5-Land data. The proposed models demonstrated potential for T_a estimation under all-weather conditions and are adaptable to other geostationary satellites. Full article

The difference (DIF) between land surface temperature (Ts) and near surface air temperature (Ta) is the key indicator of the energy budget of the land surface, which has a more complex process than the individual Ts or Ta. However, the spatiotemporal variations and influencing factors of DIF remain incomplete. The contribution of vegetation and soil moisture (SM) as key driving factors to DIF is not yet clear. Here, we analyzed the spatiotemporal variation patterns of DIF in China from 2011 to 2023 using in situ Ts and Ta data. A convergent cross-mapping method was employed to explore the causal relationship between SM, NDVI and DIF, and subsequently calculated the contribution of NDVI and SM variations to DIF under different climatic backgrounds. The results indicate that during the study period, DIF values were all above 0 °C and showed a significant increasing trend with a national mean slope of 0.02 °C/a. In general, vegetation and SM have a driving effect on DIF, with vegetation contributing more to DIF (0.11) than SM (0.08) under different surface properties. The background values of SM and temperature have a significant effect on the spatial and temporal distribution of DIF, as well as the correlation of vegetation and soil moisture to DIF. The study outcomes contribute to a better understanding of the coupling relationship between the land surface and atmosphere, which are also crucial for addressing climate change and ecological environmental management. Full article

Near-surface air temperature (T_a) is a key variable in global climate studies. Global climate models such as ERA5 and CMIP6 predict various parameters at coarse spatial resolution (>9 km). As a result, local phenomena such as the urban heat islands are not reflected in the model’s outputs. In this study, we address this limitation by downscaling the resolution of ERA5 (9 km) and CMIP6 (27 km) T_a to 1 km, employing two different machine learning algorithms (XGBoost and Deep Learning). Our models leverage a diverse set of features, including data from satellites (land surface temperature and normalized difference vegetation index), from ERA5 and CMIP6 climate models (e.g., solar and thermal radiation, wind), and from digital elevation models to develop accurate machine learning prediction models. These models were rigorously validated against observations from 98 meteorological stations in the East Mediterranean (Israel) using a standard cross-validation technique as well as a leave-one-group-out on the station ID evaluation methodology to avoid overfitting and dependence on geographic location. We demonstrate the sensitivity of the downscaled T_a to local land cover and topography, which is missing in the climate models. Our results demonstrate impressive accuracy with the Deep Learning-based models, obtaining Root Mean Squared Error (RMSE) values of 0.98 °C (ERA5) and 1.86 °C (CMIP6) for daily T_a and 2.20 °C (ERA5) for hourly T_a. Additionally, we explore the impact of the various input features and offer an extended application for future climate predictions. Finally, we propose an enhanced evaluation framework, which addresses the problem of model overfitting. This work provides practical tools and insights for building and evaluating T_a downscaling models. The code and data are publicly shared online. Full article

The Urban Heat Island (UHI) effect is of critical concern for cities’ adaptation to climate change. The UHI effect shows substantial intra-urban variation at the city microscale, causing disparities in thermal comfort and energy consumption. Therefore, air temperature assessment should be prioritized for effective heat mitigation and climate adaptation. However, meteorological stations’ spatial distribution is far from meeting the scale that the UHI and its driving parameters operate. This limitation hampers demonstrating the intra-city variability of UHI and its origin of sources; for example, most studies employ Land Surface Temperature (LST), usually without demonstrating the relationship between UHI and LST. The current body of knowledge on urban climate implies a much better understanding and more detailed information on the spatial pattern of UHI and the driving factors to provide decision-makers with tools to develop effective UHI mitigation and adaptation strategies. In an attempt to address the adequacy of the use of LST and UPs in describing the intra-city variability of UHI, this study investigates the relationship between LST daytime and nighttime, and air temperature (Ta) daytime and nighttime, and driving urban parameters (UPs) of UHI together. Although it is well recognized that the intensity of the UHI is characterized by Ta, particularly at night, so-called nocturnal UHI, the use of remotely sensed LST is common, owing to the lack of spatially detailed Ta data in cities. Our findings showed that nocturnal UHI is weakly correlated with nighttime LST with a Pearson correlation (r) of 0.335 at p > 0.05 and that it is not correlated with daytime LST for the case study, highlighting the need for Ta observations for representing the intra-urban variation of nocturnal UHI. Among UPs, Sky View Factor (SVF), Building Volume Density (BVD), and Road Network Density (RND) explained 69% of the variability of Ta nighttime that characterizes nocturnal UHI. Therefore, UPs that performed well in estimating nocturnal UHI may be used in the absence of densely distributed Ta measurements. In a further investigation of the urban cooling phenomenon based on UHI diurnal changes, a particular region with high nighttime temperatures spoiled the Ta daytime and nighttime coherence. This region is characterized by high Mean Building Height (MBH), BFD, and BVD that re-emits heat, low SVF that prevents urban cooling, and high RND that releases extra heat at night. These particular UPs can be of prior interest for urban cooling. The present study, exploring the relationships of LST and Ta in a diurnal context, offers a further understanding of the preference of LST, Ta, or UPs to characterize UHI. Ta, in relation to major causative factors (UPs), provides insights into addressing the localities most vulnerable to the UHI effect and possible strategies targeting heat mitigation for sustainability and climate change resilience. Full article

Using in situ near-surface observations of solar-induced chlorophyll fluorescence (SIF) and gross primary productivity (GPP) of a subtropical evergreen coniferous forest in southern China, this study analyzed the dynamics of SIF, GPP and their environmental responses, and explored the potential of SIF in characterizing the variation of GPP. The results showed that SIF and GPP have similar diurnal and seasonal variation and both reach the highest value in summer, indicating that the SIF can be applied to indicate the seasonal variation of GPP for the subtropical evergreen co-niferous. With the increase in temporal scale, the correlation between SIF and GPP becomes more linear. The diurnal variations of both SIF and GPP were characterized by photosynthetically active radiation (PAR), the seasonal variations of SIF and GPP were influenced by air temperature (Ta) and PAR. Probably due to the absent of drought stress during the study period, no significant correlation was detected between soil water content (SWC) and either SIF or GPP. With the in-crease in Ta, PAR or SWC, the linear correlation between the SIF and GPP gradually decreased, and when Ta or PAR was relatively higher, the correlation between SIF and GPP become weakly. Further research is still needed to illustrate the relationship between SIF and GPP under drought condition which occurred frequently in this region based on longer observation. Full article

A better understanding of the relationship between land surface temperature (Ts) and near-surface air temperature (Ta) is crucial for improving the simulation accuracy of climate models, developing retrieval schemes for soil and vegetation moisture, and estimating large-scale Ta from satellite-based Ts observations. In this study, we investigated the relationship between multiple satellite-based Ts products, derived from the Atmospheric Infrared Sounder (AIRS) and the Moderate Resolution Imaging Spectroradiometer (MODIS) onboard Aqua satellite, and Ta from 204 meteorological stations over Brazil during 2003–2016. Monthly satellite-based Ts products used in this study include: (1) AIRS Version 6 with 1° spatial resolution, (2) AIRS Version 7 with 1° spatial resolution, (3) MODIS Collection 6 with 0.05° spatial resolution, and (4) MODIS Collection 6 with 1° spatial resolution re-sampled from (3) for a direct comparison with AIRS products. We found that satellite-based Ts is lower than Ta over the forest area, but higher than Ta over the non-forest area. Nevertheless, the correlation coefficients (R) between monthly Ta and four Ts products during 2003–2016 are greater than 0.8 over most stations. The long-term trend analysis shows a general warming trend in temperatures, particularly over the central and eastern parts of Brazil. The satellite products could also observe the increasing Ts over the deforestation region. Furthermore, we examined the temperature anomalies during three drought events in the dry season of 2005, 2010, and 2015. All products show similar spatio-temporal patterns, with positive temperature anomalies expanding in areal coverage and magnitude from the 2005 to 2015 event. The above results show that satellite-based Ts is sensitive in reflecting environmental changes such as deforestation and extreme climatic events, and can be used as an alternative to Ta for climatological studies. Moreover, the observed differences between Ts and Ta may inform how thermal assumptions can be improved in satellite-based retrievals of soil and vegetation moisture or evapotranspiration. Full article

Meteorologically observed air temperature (T_a) is limited due to low density and uneven distribution that leads to uncertain accuracy. Therefore, remote sensing data have been widely used to estimate near-surface T_a on various temporal scales due to their spatially continuous characteristics. However, few studies have focused on instantaneous T_a when satellites overpass. This study aims to produce both daily and instantaneous T_a datasets at 1 km resolution for the Jingjinji area, China during 2018–2019, using machine learning methods based on remote sensing data, dense meteorological observation station data, and auxiliary data (such as elevation and normalized difference vegetation index). Newly released Moderate Resolution Imaging Spectroradiometer (MODIS) Collection 6 surface Downward Shortwave Radiation (DSR) was introduced to improve the accuracy of T_a estimation. Five machine learning algorithms were implemented and compared so that the optimal one could be selected. The random forest (RF) algorithm outperformed the others (such as decision tree, feedforward neural network, generalized linear model) and RF obtained the highest accuracy in model validation with a daily root mean square error (RMSE) of 1.29 °C, mean absolute error (MAE) of 0.94 °C, daytime instantaneous RMSE of 1.88 °C, MAE of 1.35 °C, nighttime instantaneous RMSE of 2.47 °C, and MAE of 1.83 °C. The corresponding R² was 0.99 for daily average, 0.98 for daytime instantaneous, and 0.95 for nighttime instantaneous. Analysis showed that land surface temperature (LST) was the most important factor contributing to model accuracy, followed by solar declination and DSR, which implied that DSR should be prioritized when estimating T_a. Particularly, these results outperformed most models presented in previous studies. These findings suggested that RF could be used to estimate daily instantaneous T_a at unprecedented accuracy and temporal scale with proper training and very dense station data. The estimated dataset could be very useful for local climate and ecology studies, as well as for nature resources exploration. Full article

Land-surface characteristics (LSCs) and land-soil moisture conditions can modulate energy partition at the land surface, impact near-surface atmosphere conditions, and further affect land–atmosphere interactions. This study investigates the effect of land-surface-characteristic parameters (LSCPs) including albedo, leaf-area index (LAI), and soil moisture (SM) on hot weather by in East China using the numerical model. Simulations using the Weather Research and Forecasting (WRF) Model were conducted for a hot weather event with a high spatial resolution of 1 km in domain 3 by using ERA-Interim forcing fields on 20 July 2017 until 16:00 UTC on 25 July 2017. The satellite-based albedo and LAI, and assimilation-based soil-moisture data of high temporal–spatial resolution, which are more accurate to match fine weather forecasts and high-resolution simulations, were used to update the default LSCPs. A control simulation with the default LSCPs (WRF_CTL), a main sensitivity simulation with the updated LSCP albedo, LAI and SM (WRF_CHAR), and a series of other sensitivity simulations with one or two updated LSCPs were performed. Results show that WRF_CTL could reproduce the spatial distribution of hot weather, but overestimated air temperature (Ta) and maximal air temperature (Tamax) with a warming bias of 1.05 and 1.32 °C, respectively. However, the WRF_CHAR simulation reduced the warming bias, and improved the simulated Ta and Tamax with reducing relative biases of 33.08% and 29.24%, respectively. Compared to the WRF_CTL, WRF_CHAR presented a negative sensible heat-flux difference, positive latent heat flux, and net radiation difference of the area average. LSCPs modulated the partition of available land-surface energy and then changed the air temperature. On the basis of statistical-correlation analysis, the soil moisture of the top 10 cm is the main factor to improve warming bias on hot weather in East China. Full article

Mapping of near-surface air temperature (Ta) at high spatio-temporal resolution is essential for unbiased assessment of human health exposure to temperature extremes, not least given the observed trend of urbanization and global climate change. Data constraints have led previous studies to focus merely on daily Ta metrics, rather than hourly ones, making them insufficient for intra-day assessment of health exposure. In this study, we present a three-stage machine learning-based ensemble model to estimate hourly Ta at a high spatial resolution of 1 × 1 km², incorporating remotely sensed surface skin temperature (Ts) from geostationary satellites, reanalysis synoptic variables, and observations from weather stations, as well as auxiliary geospatial variables, which account for spatio-temporal variability of Ta. The Stage 1 model gap-fills hourly Ts at 4 × 4 km² from the Spinning Enhanced Visible and InfraRed Imager (SEVIRI), which are subsequently fed into the Stage 2 model to estimate hourly Ta at the same spatio-temporal resolution. The Stage 3 model downscales the residuals between estimated and measured Ta to a grid of 1 × 1 km², taking into account additionally the monthly diurnal pattern of Ts derived from the Moderate Resolution Imaging Spectroradiometer (MODIS) data. In each stage, the ensemble model synergizes estimates from the constituent base learners—random forest (RF) and extreme gradient boosting (XGBoost)—by applying a geographically weighted generalized additive model (GAM), which allows the weights of results from individual models to vary over space and time. Demonstrated for Israel for the period 2004–2017, the proposed ensemble model outperformed each of the two base learners. It also attained excellent five-fold cross-validated performance, with overall root mean square error (RMSE) of 0.8 and 0.9 °C, mean absolute error (MAE) of 0.6 and 0.7 °C, and R² of 0.95 and 0.98 in Stage 1 and Stage 2, respectively. The Stage 3 model for downscaling Ta residuals to 1 km MODIS grids achieved overall RMSE of 0.3 °C, MAE of 0.5 °C, and R² of 0.63. The generated hourly 1 × 1 km² Ta thus serves as a foundation for monitoring and assessing human health exposure to temperature extremes at a larger geographical scale, helping to further minimize exposure misclassification in epidemiological studies. Full article

The surface and near-surface air temperature observations are primary data for glacio-hydro-climatological studies. The in situ air temperature (T_a) observations require intense logistic and financial investments, making it sparse and fragmented particularly in remote and extreme environments. The temperatures in Himalaya are controlled by a complex system driven by topography, seasons, and cryosphere which further makes it difficult to record or predict its spatial heterogeneity. In this regard, finding a way to fill the observational spatiotemporal gaps in data becomes more crucial. Here, we show the comparison of T_a recorded at 11 high altitude stations in Western Himalaya with their respective land surface temperatures (T_s) recorded by Moderate Resolution Imagining Spectroradiometer (MODIS) Aqua and Terra satellites in cloud-free conditions. We found remarkable seasonal and spatial trends in the T_a vs. T_s relationship: (i) T_s are strongly correlated with T_a (R² = 0.77, root mean square difference (RMSD) = 5.9 °C, n = 11,101 at daily scale and R² = 0.80, RMSD = 5.7 °C, n = 3552 at 8-day scale); (ii) in general, the RMSD is lower for the winter months in comparison to summer months for all the stations, (iii) the RMSD is directly proportional to the elevations; (iv) the RMSD is inversely proportional to the annual precipitation. Our results demonstrate the statistically strong and previously unreported T_a vs. T_s relationship and spatial and seasonal variations in its intensity at daily resolution for the Western Himalaya. We anticipate that our results will provide the scientists in Himalaya or similar data-deficient extreme environments with an option to use freely available remotely observed T_s products in their models to fill-up the spatiotemporal data gaps related to in situ monitoring at daily resolution. Substituting T_a by T_s as input in various geophysical models can even improve the model accuracy as using spatially continuous satellite derived T_s in place of discrete in situ T_a extrapolated to different elevations using a constant lapse rate can provide more realistic estimates. Full article

The objective of this research was to develop a robust statistical model to estimate climatologies (2002–2017) of monthly average near-surface air temperature (Ta) over Mongolia using Moderate Resolution Imaging Spectroradiometer (MODIS) land surface temperature (LST) time series products and terrain parameters. Two regression models were analyzed in this study linking automatic weather station data (Ta) with Earth observation (EO) images: Partial least squares (PLS) and random forest (RF). Both models were trained to predict Ta climatologies for each of the twelve months, using up to 17 variables as predictors. The models were applied to the entire land surface of Mongolia, the eighteenth largest but most sparsely populated country in the world. Twelve of the predictor variables were derived from the LST time series products of the Terra MODIS satellite. The LST MOD11A2 (collection 6) products provided thermal information at a spatial resolution of 1 km and with 8-day temporal resolution from 2002 to 2017. Three terrain variables, namely, elevation, slope, and aspect, were extracted using a Shuttle Radar Topography Mission (SRTM) digital elevation model (DEM), and two variables describing the geographical location of weather stations were extracted from vector data. For training, a total of 8544 meteorological data points from 63 automatic weather stations were used covering the same period as MODIS LST products. The PLS regression resulted in a coefficient of determination (R²) between 0.74 and 0.87 and a root-mean-square error (RMSE) from 1.20 °C to 2.19 °C between measured and estimated monthly Ta. The non-linear RF regression yielded even more accurate results with R² in the range from 0.82 to 0.95 and RMSE from 0.84 °C to 1.93 °C. Using RF, the two best modeled months were July and August and the two worst months were January and February. The four most predictive variables were day/nighttime LST, elevation, and latitude. Using the developed RF models, spatial maps of the monthly average Ta at a spatial resolution of 1 km were generated for Mongolia (~1566 × 10⁶ km²). This spatial dataset might be useful for various environmental applications. The method is transparent and relatively easy to implement. Full article

It is essential to monitor the ground temperature over large areas to understand and predict the effects of climate change on permafrost due to its rapid warming on the Qinghai-Tibet Plateau (QTP). Land surface temperature (LST) is an important parameter for the energy budget of permafrost environments. Moderate Resolution Imaging Spectroradiometer (MODIS) LST products are especially valuable for detecting permafrost thermal dynamics across the QTP. This study presents a comparison of MODIS-LST values with in situ near-surface air temperature (T_a), and ground surface temperature (GST) obtained from 2014 to 2016 at five sites in Beiluhe basin, a representative permafrost region on the QTP. Furthermore, the performance of the thermal permafrost model forced by MODIS-LSTs was studied. Averaged LSTs are found to strongly correlated with T_a and GST with R² values being around 0.9. There is a significant warm bias (4.43–4.67 °C) between averaged LST and T_a, and a slight warm bias (0.67–2.66 °C) between averaged LST and GST. This study indicates that averaged MODIS-LST is supposed to be a useful data source for permafrost monitoring. The modeled ground temperatures and active-layer thickness have a good agreement with the measurements, with a difference of less than 1.0 °C and 0.4 m, respectively. Full article

Land surface temperature (LST) is one of the sources of input data for modeling land surface processes. The Landsat satellite series is the only operational mission with more than 30 years of archived thermal infrared imagery from which we can retrieve LST. Unfortunately, stray light artifacts were observed in Landsat-8 TIRS data, mostly affecting Band 11, currently making the split-window technique impractical for retrieving surface temperature without requiring atmospheric data. In this study, a single-channel methodology to retrieve surface temperature from Landsat TM and ETM+ was improved to retrieve LST from Landsat-8 TIRS Band 10 using near-surface air temperature (T_a) and integrated atmospheric column water vapor (w) as input data. This improved methodology was parameterized and successfully evaluated with simulated data from a global and robust radiosonde database and validated with in situ data from four flux tower sites under different types of vegetation and snow cover in 44 Landsat-8 scenes. Evaluation results using simulated data showed that the inclusion of T_a together with w within a single-channel scheme improves LST retrieval, yielding lower errors and less bias than models based only on w. The new proposed LST retrieval model, developed with both w and T_a, yielded overall errors on the order of 1 K and a bias of −0.5 K validated against in situ data, providing a better performance than other models parameterized using w and T_a or only w models that yielded higher error and bias. Full article

Near surface air temperature (Ta) is one of the key input parameters in land surface models and hydrological models as it affects most biogeophysical and biogeochemical processes of the earth surface system. For distributed hydrological modeling over glacierized basins, obtaining high resolution Ta forcing is one of the major challenges. In this study, we proposed a new high resolution daily Ta estimation scheme under both clear and cloudy sky conditions through integrating the moderate-resolution imaging spectroradiometer (MODIS) land surface temperature (LST) and China Meteorological Administration (CMA) land data assimilation system (CLDAS) reanalyzed daily Ta. Spatio-temporal continuous MODIS LST was reconstructed through the data interpolating empirical orthogonal functions (DINEOF) method. Multi-variable regression models were developed at CLDAS scale and then used to estimate Ta at MODIS scale. The new Ta estimation scheme was tested over the Langtang Valley, Nepal as a demonstrating case study. Observations from two automatic weather stations at Kyanging and Yala located in the Langtang Valley from 2012 to 2014 were used to validate the accuracy of Ta estimation. The RMSEs are 2.05, 1.88, and 3.63 K, and the biases are 0.42, −0.68 and −2.86 K for daily maximum, mean and minimum Ta, respectively, at the Kyanging station. At the Yala station, the RMSE values are 4.53, 2.68 and 2.36 K, and biases are 4.03, 1.96 and −0.35 K for the estimated daily maximum, mean and minimum Ta, respectively. Moreover, the proposed scheme can produce reasonable spatial distribution pattern of Ta at the Langtang Valley. Our results show the proposed Ta estimation scheme is promising for integration with distributed hydrological model for glacier melting simulation over glacierized basins. Full article