Citizen science has emerged as a potent approach for environmental monitoring, leveraging the collective efforts of volunteers to gather data at unprecedented scales. Within the framework of the I-CHANGE project, MeteoTracker, a citizen science initiative, was employed to collect and analyze meteorological measurements. This paper presents the results of our analysis, demonstrating the effectiveness of citizen science in generating valuable environmental data. Through MeteoTracker, volunteers contributed to a comprehensive dataset, enabling insights into local weather patterns and trends. Our findings underscore the potential of citizen science to augment traditional monitoring efforts and inform decision-making processes in environmental research and management. The paper discusses the methodology employed, including data collection and statistical analysis techniques. Results highlight the reliability and accuracy of citizen-generated data, as well as the strengths of the MeteoTracker platform. The implications of our findings for environmental research, policy development, and public engagement are discussed, along with recommendations for enhancing the effectiveness of citizen science initiatives in environmental monitoring.

Estimating precipitation is of critical importance to climate systems and decision-making processes. This paper presents Espresso, a deep learning model designed for estimating precipitation from satellite observations on a global scale. Conventional methods, like ground-based radars, are limited in terms of spatial coverage. Satellite observations, on the other hand, allow global coverage. Combined with deep learning methods these observations offer the opportunity to address the challenge of estimating precicpation on a global scale. This research paper presents the development of a deep learning model using geostationary satellite data as input and generating instantaneous rainfall rates, calibrated using data from the Global Precipitation Measurement Core Observatory (GPMCO). The performance impact of various input data configurations on Espresso was investigated. These configurations include a sequence of four images from geostationary satellites and the optimal selection of channels. Additional descriptive features were explored to enhance the model’s robustness for global aplications. When evaluated against the GPMCO test set, Espresso demonstrated highly accurate precipitation estimation, especially within equatorial regions. A comparison against six other operational products using multiple metrics indicated its competitive performance. The model’s superior storm localization and intensity estimation were further confirmed through visual comparisons in case studies. Espresso has been incorporated as an operational product at Météo-France, delivering high-quality, real-time global precipitation estimates every 30 minutes.

Severe windstorms resulting from mesoscale convective systems (MCS) cause significant disruption to society, including widespread power outages, tree and structural damage, and transportation accidents that affect multi-state regions and metropolitan areas along their track. A derecho, defined as a long-lived, widespread severe convective windstorm, is composed of numerous downbursts (intense localized storm downdrafts) organized into clusters or families of clusters. Derechos can produce winds above hurricane force along a track that may exceed several hundred kilometers. Convective windstorm potential has been expressed as a grouping of stability parameters that are relevant for downburst generation. These include lower-to-mid-tropospheric and equivalent potential temperature (θe) lapse rates, vertical relative humidity differences, and the amount of convective available potential energy (CAPE) in the troposphere. One major environmental factor addressed in the generation of widespread severe convective winds is an elevated mixed layer and its associated instability that promotes powerful storm updrafts and downdrafts. Another critical factor is developing a rear-inflow jet into an MCS that channels unsaturated mid-tropospheric air into the leading convective storm line. Establishing an elevated, ascending front-to-rear flow originating from deep, moist convection, overlying a strong and deep outflow-induced cold pool has been found to generate and sustain a robust rear inflow jet. The North American Derecho of 29-30 June 2012 exhibits many classic progressive and serial derecho features. It remains as one of the highest-impact derecho-producing convective systems (DCS) over CONUS since 2000. This research effort enhances the understanding of the science of operational forecasting of severe windstorms through examples of employing new satellite and ground-based microwave and vertical wind profile data. During the track of the derecho from the upper Midwestern U.S. through the mid-Atlantic region on 29 June 2012, clear signatures associated with a severe MCS were apparent in polar satellite imagery, especially from the EPS METOP-A Microwave Humidity Sounder (MHS), Defense Meteorological Satellite Program (DMSP) Special Sensor Microwave Imager Sounder (SSMIS) and NASA TERRA Moderate Resolution Imaging Spectroradiometer (MODIS). In addition, morning (descending node) and the evening (ascending node) METOP-A Infrared Atmospheric Sounding Interferometer (IASI) soundings are compared to soundings from surface-based Radiometrics Corporation MP-3000 series microwave radiometer profilers (MWRPs) along the track of the derecho system. The co-located IASI and MWRP soundings revealed a pre-convective environment that indicated a favorable volatile tropospheric profile for severe downburst wind generation. An important outcome of this study will be to formulate a functional relationship between satellite-derived parameters and signatures, and severe convective wind occurrence. In summary, a comprehensive approach to observational data analysis involves both surface- and satellite-based instrumentation. Because this approach utilizes operational products available to weather service forecasters, it can feasibly be used for monitoring and forecasting local-scale downburst occurrence within derecho systems, as well as larger-scale convective wind intensity associated with the entire DCS.

While the surface of the Earth plays a key role in weather forecasting through its interaction with the atmosphere, in ensemble numerical weather predictions the uncertainty on the surface is only represented with perturbations in the parameterisations representing the surface processes. Data representing the surface, such as the land cover, are not perturbed. As fully data-driven forecasts without parameterisations are growing in importance, sampling the uncertainty on the land cover data brings a new way of making ensemble forecasts. Our work describes a method of generating ensemble land cover maps for numerical weather prediction. The target land cover map has the ECOCLIMAP-SG labels, used in the SURFEX surface model, and therefore is expected to have all relevant labels for surface-atmosphere interactions. The method translates the ESA WorldCover map to ECOCLIMAP-SG labels and resolution using auto-encoders. The land cover ensemble members are obtained by sampling the land cover probabilities in the output of the neural network. This paper builds upon the work done in a companion paper describing the high-resolution version of ECOCLIMAP-SG, called ECOCLIMAP-SG+, used for the training and evaluation of the neural network. The output map presented here, called ECOCLIMAP-SG-ML, improves upon the ECOCLIMAP-SG map in terms of resolution (from 300~m to 60~m), overall accuracy (from 0.41 to 0.63) and the ability to produce ensemble members.

ECOCLIMAP-SG+ is a new 60~m land use land cover dataset, which covers a continental domain, and represents the 33 labels of the original ECOCLIMAP-SG dataset. ECOCLIMAP-SG is used in HARMONIE-AROME, the numerical weather prediction model used operationally by Met Éireann and other national meteorological services. ECOCLIMAP-SG+ was created using an agreement-based method to combine information from many maps to overcome variations in semantic and geographical coverage, resolutions, formats, accuracies, and representative periods. In addition to ECOCLIMAP-SG+, the process generates an agreement score map, which estimates the uncertainty of the land cover labels in ECOCLIMAP-SG+ at each location in the domain. This work presents the first evaluation of ECOCLIMAP-SG and ECOCLIMAP-SG+ against the following trusted land cover maps: LUCAS 2022, the Irish National Land Cover 2018 dataset, and an Icelandic version of ECOCLIMAP-SG. Using a set of primary labels, ECOCLIMAP-SG+ outperforms ECOCLIMAP-SG regarding the F1-score against LUCAS 2022 over Europe and the Irish national land cover 2018 dataset. Similarly, it outperforms ECOCLIMAP-SG against the Icelandic version of ECOCLIMAP-SG for most of the represented secondary labels. The score map shows that the quality ECOCLIMAP-SG+ is hetereogeneous. It could be improved once new maps once they become available but we do not control when they will be available. Therefore, the second-part of this publication series aims at improving the map using machine learning.

The cause of the systematic daily march of barometric pressure in the tropics, notably the late morning and late evening peaks seen almost every day at all locations, was a puzzle that persisted through the nineteenth and much of the twentieth centuries. The efforts to explain the physics of the prominent 12-hour solar tidal variation helped inspire some of the earliest developments in theoretical atmospheric dynamics and ultimately led in the 1960’s to a satisfactory dynamical theory for the atmospheric tides. These important theoretical developments followed the observational discoveries, which date to the late 18th and early 19th centuries, of the surprising character of the barometric daily march and of its resolution into solar and lunar period cycles. These important, if simple, discoveries emerged primarily from the efforts of European scientists to systematically study the environment in remote areas of the globe. The two key figures in initially advancing the scientific community’s understanding of the character of barometric tides were the great German polymath Alexander von Humboldt (1769-1859) and the French naturalist Robert de Lamanon (1752-1787), who each made their discoveries on their most famous and colorful scientific expeditions of their respective careers. This paper examines the history of the early observations of the barometric tide.

Rainfall retrieval using geostationary satellites provides critical means to the monitoring of extreme rainfall events. Using the relatively new Himawari 8 meteorological satellite with three times more channels than its predecessors, the deep learning framework of “convolutional autoencoder” (CAE) was applied to the extraction of cloud and precipitation features. The CAE method was incorporated into the Convolution Neural Network version of the PERSIANN precipitation retrieval that uses GOES satellites. By applying the CAE technique with the addition of Residual Blocks and other modifications of deep learning architecture, the presented derivation of PERSIANN operated at the Central Weather Bureau of Taiwan (referred to as PERSIANN-CWB) expands four extra convolution layers to fully use Himawari 8’s infrared and water vapor channels, while preventing degradation of accuracy caused by the deeper network. The development of PERSIANN-CWB was trained over Taiwan for its diverse weather systems and localized rainfall features, and the evaluation reveals an overall improvement from its CNN counterpart and superior performance over all other rainfall retrievals analyzed. Limitation of this model was found in the derivation of typhoon rainfall, an area requiring further research.

The mass composition of Particulate Matter (PM) with an aerodynamic diameter of 2.5 microns (PM2.5) in San Joaquin Valley (SJV) is dominated by ammonium nitrate (NH4NO3) which is a secondary pollutant. The goal of this research was the investigation of the relationship between emissions, meteorology and PM2.5 concentrations in Fresno for the winter season. It was found that the location of sites near emission sources such as freeways compared with residential sites strongly affected measured PM2.5 concentrations. It was found that although long-term trends showed declines in both emissions and PM2.5 concentrations there was substantial variability between the years in the PM2.5/emissions relationship. Much of the yearly variation in the relationship between emissions and PM2.5 concentrations can be attributed to yearly variations in weather, such as atmospheric stability, precipitation frequency and average wind speed. There are moderate correlations between PM2.5 concentrations and temperature differences between nearby surface stations at varying elevations which explains some of the daily and seasonal variation in PM2.5. Occurrence of precipitation was related to low PM¬2.5 although the higher wind speeds and lower atmospheric stability associated with precipitation likely explain some of the low PM2.5 as well as washout of PM.

Diablo winds are dry and gusty north-northeasterly downslope windstorms that affect the Northern California. On the evening of October 8, 2017, Diablo winds contributed to the ignitions and rapid spread of the “Northern California Firestorm”, including the Tubbs Fire, which burned 2,800 homes in Santa Rosa, resulted in 22 fatalities and $1.2B USD in damages. We analyzed 18 years of data from a network of surface meteorological stations and show that Diablo winds tend to occur overnight through early morning in Fall, Winter and Spring. Conditions similar to Diablo winds occur commonly in the Sierra Nevada Mountains in addition to the area North of the San Francisco Bay Area including the Northern Coastal Ranges. Diablo winds are characterized by high wind speeds and low relative humidity, but they tend to neither be warmer than climatology nor do they have a higher gust factor, or ratio of wind gusts to mean wind speeds, than climatology.

The high-resolution bistatic lidar developed at the Physikalisch-Technische Bundesanstalt (PTB) aims to overcome the limitations of conventional monostatic lidar technology which is widely used for wind velocity measurements in wind energy and meteorology applications. Due to the large measurement volume of a combined optical transmitter and receiver tilting in multiple directions, monostatic lidar generally has poor spatial and temporal resolution. It also exhibits large measurement uncertainty when operated in inhomogeneous flow, for instance, over complex terrain. In contrast, PTB’s bistatic lidar uses three dedicated receivers arranged around a central transmitter, resulting in an exceptionally small measurement volume. The coherent detection and modulation schemes used allow the detection of backscattered, Doppler shifted light down to the scale of single aerosols, realising the simultaneous measurement of all three wind velocity components. This paper outlines design details and the theory of operation of PTB’s bistatic lidar and provides an overview of selected comparative measurements. The results of these measurements have shown that the measurement uncertainty of PTB’s bistatic lidar is well within the measurement uncertainty of traditional cup anemometers, while being fully independent of its site and traceable to the SI units. This allows its use as a transfer standard for the calibration of other remote sensing devices. Overall, PTB’s bistatic lidar shows great potential to universally improve the capability and accuracy of wind velocity measurements, such as for the investigation of highly dynamic flow processes upstream and in the wake of wind turbines.

In this paper we illustrate a new, simple and complementary ground-based methodology to retrieve the vertically resolved atmospheric precipitation intensity through a synergy between measurements from the National Aeronautics and Space Administration (NASA) Micropulse Lidar network (MPLNET), an analytical model solution and ground-based disdrometer measurements. The presented results are obtained at two mid-latitude MPLNET permanent observational sites, located respectively at NASA Goddard Space Flight Center, USA, and at the Universitat Politècnica de Catalunya, Barcelona, Spain. The methodology is suitable to be applied to existing and/or future lidar/ceilometer networks with the main objective of either providing near-real time (3h latency) rainfall intensity measurements and/or to validate satellite missions, especially for critical light precipitation (<3 mm hr⁻¹).

Unhealthy concentrations of ozone in the Uinta Basin, Utah, can occur after sufficient snowfall and a strong atmospheric anticyclone creates a persistent cold pool that traps atmospheric ozone and its precursors emitted from oil and gas operations. The winter-ozone system has two clear outcomes—occurrence or not—that is well understood by domain experts and supported by archives of atmospheric observations. Rules of the system can be formulated in natural language (“sufficient snowfall and high pressure leads to high ozone"), lending itself to analysis with a fuzzy-logic inference system. This method encodes human expertise as machine intelligence in a more constrained manner than alternative, more complex inference methods such as neural networks, increasing user trustworthiness of our model prototype before further optimization. Herein, we develop an ozone-forecasting system, Clyfar, based on knowledge of system dynamics and informed by an archive of meteorological conditions and ozone concentration. The inference system demonstrates proof-of-concept despite rudimentary tuning. We describe our framework for predicting future ozone concentrations if input values are drawn from numerical weather prediction forecasts as a proxy for observations as the system’s initial conditions. Our model is computationally cheap, allowing us to sample uncertainty with substantially more ensemble members than in traditional NWP. We evaluate hindcasts for one winter, finding our prototype demonstrates promise to deliver useful guidance for users concurrent with optimization of system parameters using machine learning.

Marine cold air outbreaks (MCAOs) are large-scale phenomena in which cold air masses are advected over open ocean. It is well-known that these events are linked to the formation of polar lows and other mesoscale phenomena associated with high wind speeds, and that they therefore in some cases represent a hazard to maritime activities. However, it is still unknown whether MCAOs are generally conducive to higher wind speeds than normal. Here this is investigated by comparing the behaviour of ocean surface wind speeds during MCAOs in three atmospheric reanalysis products with different horizontal grid spacings, along with case studies using a convection-permitting numerical weather prediction model. The study regions are the Labrador Sea and the Greenland–Iceland–Norwegian (GIN) Seas, where MCAOs have been shown to be important for air–sea interaction and deep water formation. The main findings are: 1) Wind speeds during the most extreme MCAO events are stronger than normal and higher than wind speeds during less severe events; 2) The peak times of MCAO usually occur when baroclinic waves pass over the regions; and 3) Reanalyses with grid spacings of more than 50 km appear to underestimate winds driven by the large ocean–atmosphere energy fluxes during MCAOs. It is also shown that while the strong wind episodes during MCAOs generally last for just a few days, MCAOs can persist for up to 50 days. These findings demonstrate that it would be worthwhile to forecast MCAOs, and that it might be possible to do this beyond the standard weather forecasting range of up to 10 days.

Land degradation, desertification and tree mortality related to global climate change are in the spotlight of Remote Sensing research in recent decades, since extreme climatic events could affect the composition, structure and biogeography of forests. However, the complexity of tree mortality processes requires a holistic approach. Herein, we present the first global assessment and a historical perspective of tree mortality by reviewing both Remote Sensing and ground-based studies. We compiled 254 papers on tree mortality that make use of remotely sensed products, ground-based monitoring and climatic drivers, focusing on their spatial and temporal patterns, in the methods applied, while highlighting research gaps. Forested ecosystems, especially in the Mediterranean area, appear to have experienced often climatic-induced physiological stress under extreme droughts and warming, raising concern that forests may become increasingly vulnerable to tree mortality. Our core results indicated that international publications on tree mortality are on the increase with the main hotspots being North America (39%) and Europe (26%). Wetness indicators appear as the barometer in explaining tree mortality on a local scale, while vegetation indicators derived from multispectral optical sensors are promising for large scale assessment. We observed that almost all of the studies we have reviewed were based on less than 25 years of data and were of local scale. Longer timeframes and regional scale investigations that will include multiple tree species analysis could have a significant impact on future research.

Observations collected during cold season precipitation periods at Utquagvik, Alaska and at the mul-tidisciplinary drifting observatory for study of Arctic climate (MOSAiC) are used to statistically ana-lyze relations among the atmospheric water cycle parameters including the columnar supercooled liq-uid and ice amounts (expressed as liquid water and ice water paths, i.e., LWP and IWP), the integrated water vapor (IWV) and the near-surface snowfall rate. Data come from radar and radiometer-based retrievals and from optical precipitation sensors. While correlation between snowfall rate and LWP is rather weak, correlations coefficients between radar-derived snowfall rate and IWP are high (~ 0.8), which is explained, in part, by generally low LWP/IWP ratios during significant precipitation. Corre-lation coefficients between snowfall rate and IWV are moderate (~0.45). Correlations are generally weaker if snowfall is estimated by optical sensors, which is, in part, due to blowing snow. Correlation coefficients between near-surface temperature and snowfall rates are low (r<0.3). Results from the Alaska and MOSAiC sites are generally similar. These results are not very sensitive to the amount of time averaging (e.g., 15-minute averaging versus daily averages). Observationally-based relations among the water cycle parameters are informative about atmospheric moisture conversion processes and can be used for model evaluations.

A coupled stratospheric chemistry-meteorology model was developed by combining the Canadian operational weather prediction model Global Environmental Multiscale (GEM) with a comprehensive stratospheric photochemistry model from the Belgian Assimilation System for Chemical ObsErvations (BASCOE). The coupled model was called GEM-BACH for GEM-Belgian Atmospheric CHemistry. The coupling was made across a chemical interface that preserves time splitting while being modular, allowing GEM to run with or without chemistry. An evaluation of the coupling was performed by comparing the coupled model, refreshed by meteorological analyses every 6 hours, against the standard offline chemical transport model (CTM) approach. Results show that the dynamical meteorological consistency between meteorological analysis times far outweighs the error created by the jump resulting from the meteorological analysis increments at regular time intervals, irrespective whether a 3D-Var or 4D-Var meteorological analysis is used. GEM-BACH forecast refreshed by meteorological analyses every 6 hours were compared against independent measurements of temperature, long-lived species, ozone and water vapor. The comparison showed a relatively good agreement throughout the stratosphere except for an upper-level warm temperature bias and an ozone deficit of nearly 15%. Arguments in favor of using the same horizontal resolution for chemistry, meteorology, and meteorological analysis increments are also presented. In particular, the coupled model simulation during an ozone hole event gives better ozone concentrations than a 4D-Var chemical assimilation at a lower resolution.

While thunderstorms can pose severe risks to property and life, forecasting remains challenging, even at short lead times, as these often arise in meta-stable atmospheric conditions. In this paper, we examine the question of how well we could perform short-term (up to 180min) forecasts using exclusively multi-spectral satellite images and past lighting events as data. We employ representation learning based on deep convolutional neural networks in an “end-to-end” fashion. Here, a crucial problem is handling the imbalance of the positive and negative classes appropriately in order to be able to obtain predictive results (which is not addressed by many previous machine-learning-based approaches). The resulting network outperforms previous methods based on physically-based features and optical flow methods (similar to operational prediction models) and generalizes across different years. A closer examination of the classifier performance over time and under masking of input data indicates that the learned model actually draws most information from structures in the visible spectrum, with infrared imaging sustaining some classification performance during the night.

Environmental pollution indicated by the presence of PM2.5 particulate matter varies based on prevailing atmospheric conditions described by certain meteorological variables. Consequently, it is important to understand atmospheric behavior in areas such as the Aburrá Valley, which experiences recurrent pollution events twice a year. This study examines the behavior of specific meteorological variables and PM2.5 particulate matter in the Aburrá Valley. By using statistical analysis tools such as correlation coefficients, principal component analysis (PCA), and multiple linear regression models, the research identifies relationships between PM2.5 and daily cycles of temperature, rainfall, radiation, and wind speed and direction. Data sets were analyzed considering periods before and after the COVID-19 lockdown (pre-pandemic and pandemic, respectively), and specific pollution events were also analyzed. Furthermore, this work considers the relationships between PM2.5 and meteorological variables, contrasting the pre-pandemic and pandemic periods. This study characterizes diurnal cycles of meteorological variables and their relationship with PM2.5. There are consistent patterns among temperature, atmospheric boundary layer (ABL) height, and solar radiation, whereas precipitation and relative humidity show opposite behavior. PM2.5 exhibits similar relative frequency functions during both daytime and nighttime, regardless of rainfall. An inverse relationship is noted between PM2.5 levels and ABL height at different times of the day. Moreover, the PCA results show that the first principal component explains around 60% of the total variance of the hydrometeorological data. The second PC explains 10%, and the rest of the variance is distributed among the other three to eight PCs. In this sense, there is no significant difference between the two PCA analyses with hydrometeorological data from a pre-pandemic period and a COVID-19 pandemic. Multiple regression analysis indicates a significant and consistent dependence of PM2.5 on temperature and solar radiation across both analyzed periods. This work presents new insights that enhance the diagnosis, understanding, and modeling of environmental pollution, thereby supporting informed decision-making and strengthening management efforts.

We examine data assimilation coupling between meteorology and chemistry in the stratosphere from both weak and strong coupling strategies. The study was performed with the Canadian operational weather prediction Global Environmental Multiscale (GEM) model coupled online with the photochemical stratospheric chemistry developed at the Belgian Institute for Space Aeronomy, described in Part I. Here, the Canadian Meteorological Centre’s operational variational assimilation system was extended to include errors of chemical variables and cross-covariances between meteorological and chemical variables in a 3D-Var configuration, and we added the adjoint of tracer advection in the 4D-Var configuration. Our results show that the assimilation of limb sounding observations from the MIPAS instrument on board Envisat can be used to anchor the AMSU-A radiance bias correction scheme. Also, the added value of limb sounding temperature observations on meteorology and transport is shown to be significant. Weak coupling data assimilation with ozone-radiation interaction is shown to give comparable on meteorology whether a simplified linearized or comprehensive ozone chemistry scheme is used. Strong coupling data assimilation, using static error cross-covariances between ozone and temperature in a 3D-Var context, produced inconclusive results with the approximations we used. We have also conducted the assimilation of long-lived species observations using 4D-Var to infer winds. Our results showed the added value of assimilating several long-lived species, and an improvement in the zonal wind in the Tropics within the troposphere and lower stratosphere. 4D-Var assimilation also induced a correction of zonal wind in the surf zone and a temperature bias in the lower tropical stratosphere