Laboratory for
Climatology and Remote Sensing

Abstract:

Radiation fog nowcasting remains a complex yet critical task due to its substantial impact on traffic safety and economic activity. Current numerical weather prediction models are hindered by computational intensity and knowledge gaps regarding fog-influencing processes. Machine-Learning (ML) models, particularly those employing the eXtreme Gradient Boosting (XGB) algorithm, may offer a robust alternative, given their ability to learn directly from data, swiftly generate nowcasts, and manage non-linear interrelationships among fog variables. However, unlike recurrent neural networks XGB does not inherently process temporal data, which is crucial in fog formation and dissipation. This study proposes incorporating preprocessed temporal data into the model training and applying a weighted moving-average filter to regulate the substantial fluctuations typical in fog development. Using an ML training and evaluation scheme for time series data, we conducted an extensive bootstrapped comparison of the influence of different smoothing intensities and trend information timespans on the model performance on three levels: overall performance, fog formation and fog dissipation. The performance is checked against one benchmark and two baseline models. A significant performance improvement was noted for the station in Linden-Leihgestern (Germany), where the initial F1 score of 0.75 (prior to smoothing and trend information incorporation) was improved to 0.82 after applying the smoothing technique and further increased to 0.88 when trend information was incorporated. The forecasting periods ranged from 60 to 240 min into the future. This study offers novel insights into the interplay of data smoothing, temporal preprocessing, and ML in advancing radiation fog nowcasting.

Abstract:

Abstract The Atacama Desert is the driest non-polar desert on Earth, presenting precarious conditions for biological activity. In the arid coastal belt, life is restricted to areas with fog events that cause almost daily wet–dry cycles. In such an area, we discovered a hitherto unknown and unique ground covering biocenosis dominated by lichens, fungi, and algae attached to grit-sized (~6 mm) quartz and granitoid stones. Comparable biocenosis forming a kind of a layer on top of soil and rock surfaces in general is summarized as cryptogamic ground covers (CGC) in literature. In contrast to known CGC from arid environments to which frequent cyclic wetting events are lethal, in the Atacama Desert every fog event is answered by photosynthetic activity of the soil community and thus considered as the desert's breath. Photosynthesis of the new CGC type is activated by the lowest amount of water known for such a community worldwide thus enabling the unique biocenosis to fulfill a variety of ecosystem services. In a considerable portion of the coastal Atacama Desert, it protects the soil from sporadically occurring splash erosion and contributes to the accumulation of soil carbon and nitrogen as well as soil formation through bio-weathering. The structure and function of the new CGC type are discussed, and we suggest the name grit–crust. We conclude that this type of CGC can be expected in all non-polar fog deserts of the world and may resemble the cryptogam communities that shaped ancient Earth. It may thus represent a relevant player in current and ancient biogeochemical cycling.

Abstract:

The Atacama Desert is considered as one of the driest places on Earth. At the coastline, however, small-scale fog oases harbor a specialized vegetation and fauna, living from moisture by fog, which is used by humans to feed water demands of industrial projects. To date, knowledge about fog and low stratus (FLS) clouds as well as their physical properties is limited in that only local observations or spatial products from satellites with coarse resolutions are available generally failing to capture local patterns resulting from the complex topography. Consequently, we provide the first climatology of FLS with 30m spatial resolution based on over 400 Landsat scenes acquired since 1986. The new product provides valuable estimates of FLS optical and micro-physical properties. FLS over the Pacific Ocean featured cloud optical depth values around 13.5 declining over land to 4.2. Effective radii were around 5.3 μm. Liquid water path was between 71.0 − gm 2 over the Ocean and 14.9 − gm 2 over land surfaces. The climatologies of the new Landsat product were successfully validated against those of the MODIS cloud property product over homogeneous surfaces. Over areas with heterogeneous topographies, the new product outperforms existing ones with coarse spatial resolutions if compared against in situ measurements. This shows the general need for cloud products with high spatial resolutions in areas where the development of small scale clouds is favored e.g., by a complex topography leading to systematical biases in existing retrievals.

Abstract:

Coherent wind doppler lidar (CWDL) is a cost-effective way to estimate wind power potential at hub height without the need to build a meteorological tower. However, fog and low stratus (FLS) can have a negative impact on the availability of lidar measurements. Information about such reductions in wind data availability for a prospective lidar deployment site in advance is beneficial in the planning process for a measurement strategy. In this paper, we show that availability reductions by FLS can be estimated by comparing time series of lidar measurements, conducted with WindCubes v1 and v2, with time series of cloud base altitude (CBA) derived from satellite data. This enables us to compute average maps (2006–2017) of estimated availability, including FLS-induced data losses for Germany which can be used for planning purposes. These maps show that the lower mountain ranges and the Alpine regions in Germany often reach the critical data availability threshold of 80% or below. Especially during the winter time special care must be taken when using lidar in southern and central regions of Germany. If only shorter lidar campaigns are planned (3–6 months) the representativeness of weather types should be considered as well, because in individual years and under persistent weather types, lowland areas might also be temporally affected by higher rates of data losses. This is shown by different examples, e.g., during radiation fog under anticyclonic weather types.

Abstract:

Fog plays a major role in many ecological aspects and it influences human life in various ways. In this study, a temporally highly resolved and spatially explicit anal- ysis of variations in fog occurrence was conducted for Europe and links to general weather conditions were investigated. To this end, a high-resolution fog product based on Meteosat Second Generation data was developed. Characteristic fog distri- butions were identified by applying a Self Organizing Map approach to the dataset. It was found that the resulting fog patterns are primarily determined by terrain char- acteristics. Simultaneous occurrences between these patterns and the predominant general weather situations were analyzed. The results show that the general weather situations can be categorized into three main groups, each responsible for the forma- tion of a different group of fog patterns. Additionally, distinct regional differences could be identified in the diurnal and annual fog frequency cycles and the derived region-specific frequency variations were used to draw conclusions about the fog types prevailing in these regions.

Abstract:

Fog has a substantial influence on various ecosystems and it impacts economy, traffic systems and human life in many ways. In order to be able to deal with the large number of influence factors, a spatially explicit high-resoluted data set of fog frequency distribution is needed. In this study, a hybrid approach for fog retrieval based on Meteosat Second Generation (MSG) data and ground truth data is presented. The method is based on a random forest (RF) machine learning model that is trained with cloud base altitude (CBA) observations from Meteorological Aviation Routine Weather Reports (METAR) as well as synoptic weather observations (SYNOP). Fog is assumed where the model predicts CBA values below a dynamically derived threshold above the terrain elevation. Cross validation results show good accordance with observation data with a mean absolute error of 298 m in CBA values and an average Heidke Skill Score of 0.58 for fog occurrence. Using this technique, a 10 year fog baseline climatology with a temporal resolution of 15 min was derived for Europe for the period from 2006 to 2015. Spatial and temporal variations in fog frequency are analyzed. Highest average fog occurrences are observed in mountainous regions with maxima in spring and summer. Plains and lowlands show less overall fog occurrence but strong positive anomalies in autumn and winter.

Abstract:

The mountain cloud forest of Taiwan can be delimited from other forest types using a map of the ground fog frequency. In order to create such a frequency map from remotely sensed data, an algorithm able to detect ground fog is necessary. Common techniques for ground fog detection based on weather satellite data cannot be applied to fog occurrences in Taiwan as they rely on several assumptions regarding cloud properties. Therefore a new statistical method for the detection of ground fog in mountainous terrain from MODIS Collection 051 data is presented. Due to the sharpening of input data using MODIS bands 1 and 2, the method provides fog masks in a resolution of 250 m per pixel. The new technique is based on negative correlations between optical thickness and terrain height that can be observed if a cloud that is relatively plane-parallel is truncated by the terrain. A validation of the new technique using camera data has shown that the quality of fog detection is comparable to that of another modern fog detection scheme developed and validated for the temperate zones. The method is particularly applicable to optically thinner water clouds. Beyond a cloud optical thickness of ? 40, classification errors significantly increase.

Abstract:

Operational nowcasting techniques for sea fog over the Yellow Sea rely on data from weather satellites because ground-based observations are hardly available. While there are several algorithms for detecting low stratus (LST) that are applicable to geostationary weather satellite data, sea fog retrieval is more complicated. These schemes mostly need ancillary data such as Cloud Optical Thickness (COT) and Droplet Effective Radius (DER). To retrieve the necessary parameters for sea fog detection over the Yellow Sea, the Comprehensive Analysis Program for Cloud Optical Measurement (CAPCOM) scheme developed by Kawamoto et al. (2001) was adapted to the Japanese Multifunctional Transport Satellites (MTSAT) system-Japanese Advanced Meteorological Imager (JAMI). COT and DER values were then retrieved for 64 cases over the Yellow Sea (= 85,000 LST pixels) and compared with the COT and DER products from the MYD06/MOD06, CAPCOM-MODIS (Moderate Resolution Imaging Spectroradiometer) and CloudSat (cloud radar). Results showed that the COT and DER values retrieved from JAMI were satisfactory. The MTSAT-2 JAMI data delivered better COT values than the MTSAT-1R JAMI data, due to the re-calibration of MTSAT-2 JAMI’s visible (VIS) band in 2011. Similarly, improvements were seen in DER retrieval, even though the VIS re-calibration primarily affects COT retrieval. By comparing the difference in stratus thickness calculated by MTSAT-1R and MTSAT-2, the COT and DER retrieved from MTSAT-2 JAMI can be used in ground fog retrieval schemes. These values exhibit less bias, especially in cases involving high cloud top and thin cloud thickness. Both the COT and DER retrievals from MTSAT-2 JAMI offer potential as reliable parameters for Yellow Sea fog detection.

Abstract:

The vertical distribution of liquid water content (LWC) in natural fog and low stratus is a crucial variable in many applications, e.g. the development of satellite based retrievals of ground fog. Unfortunately, there is very little data concerning fog LWC-profiles, mainly due to the lack of suitable operational instrumentation. A novel ground-based 94 GHz FMCW cloud radar could fill this gap if radar reflectivity Z could be converted to LWC by using appropriate Z–LWC relations. However, this relation strongly depends on drop size distribution (DSD) and is hardly known for natural fog types. In this sensitivity study, the influence of the DSD on the Z–LWC relation in different types and life cycle stages of natural fogs is analyzed using a radiative transfer code (RTC) and published fog drop size distributions. It could be shown that there is a direct but nonlinear relationship between LWC and radar reflectivity. The proportionality factor of the Z–LWC equation in particular reveals specific ranges for the different life cycle stages. If a proper classification of fog life cycle in the field is possible, the results could be used to properly convert Z to LWC.

Abstract:

As part of an interdisciplinary research programme, the spatial and temporal variability of precipitation in southern Ecuador has been investigated since January 2002. The study site is located at the northern margin of the Podocarpus National Park in the vicinity of Loja, about 500 km south of Quito, at altitudes ranging from 1800 to 3200 m.a.s.l. Due to its low density, the conventional rainfall station network fails to register the highly variable spatial distribution of rain, whereas contributions by fog are not accounted at all. Hence, for the first time in a tropical montane forest setting, a weather radar was used, covering a radius of 60 km and reaching from the Amazon Basin to the coastal plains of the region. Furthermore, a network of sampling stations supplies data about the altitudinal gradient of fog and rainwater inputs. The precipitation distribution in the study area proves to be far more variable than previously thought and is strongly coupled to the orographic characteristics and the special topographical setting of the landscape. Maxima in precipitation occur especially in the eastern parts of the radar range on slopes exposed to advected moisture from the Amazon Basin, whereas the highest crests of the Andes receive less precipitation. The study area has two cloud condensation levels, occurring at 1500–2000 and 2500–3500 m.a.s.l., respectively. At 1800–2000 m.a.s.l., fog is estimated to contribute an additional input of 5% of conventionally measured rainfall, increasing to about 35% at the highest measurement station (3200 m.a.s.l.). In contrast to some other tropical mountains, there seems to be no maximum zone of water input, although the gradient remains positive up to the highest altitudes. The unusual precipitation distribution is thought to reflect the contrasting climatological influences operating in the study area.

Abstract:

Recent work on bryophyte diversity in lowland forests of northern South America has suggested the existence of a new type of cloud forest, the ‘tropical lowland cloud forest’ (LCF). LCF occurs in river valleys in hilly areas with high air humidity and morning fog, and is rich in epiphytes. We explored epiphyte abundance and canopy microclimate of LCF in a lowland area (200–400 m asl) near Saül, central French Guiana. We analyzed the vertical distribution of epiphytic cover and biomass on 48 trees, in LCF and in lowland rain forest (LRF) without fog. Trees in LCF had significantly more epiphytic biomass than in LRF; mean total epiphytic biomass in LCF was about 59 g/m2, and 35 g/m2 in LRF. In all height zones on the trees, total epiphyte cover in LCF exceeded that in LRF, with ca 70 percent mean cover in LCF and ca 15 percent in LRF. During both wet and dry seasons, mean diurnal relative air humidity (RH) was higher in LCF than in LRF, and persistence of high RH after sunrise significantly longer in LCF. We suggest that the prolonged availability of high air humidity in LCF and the additional input of liquid water through fog, enhance epiphyte growth in LCF by shortening the desiccation period and lengthening the period of photosynthetic activity of the plants.

Abstract:

A method is presented for fog and low stratus detection from daytime satellite imagery based on Meteosat 8 SEVIRI (Spinning-Enhanced Visible and Infra-Red Imager) data. With its excellent spatial, spectral and temporal resolutions, this imagery is an ideal basis for operational fog monitoring. The scheme utilizes a range of pixel-based and novel object-oriented techniques to separate fog and low stratus clouds from other cloud types. Fog and low stratus are identified by a number of tests which explicitly and implicitly address fog/low stratus spectral, spatial and microphysical properties. The scheme's performance is evaluated using ground-based measurements of cloud height over Europe. The algorithm is found to detect low clouds very accurately, with probabilities of detection (POD) ranging from 0.632 to 0.834 (for different inter-comparison approaches), and false alarm ratios (FAR) between 0.059 and 0.021. The retrieval of sub-pixel and temporal effects remain issues for further investigation.

Abstract:

Automated detection of fog and low stratus in nighttime satellite data has been implemented on the basis of numerous satellite systems in past decades. Commonly, differences in small-droplet emissivities at 11?m and 3.9?m are utilized. With Meteosat SEVIRI, however, this method cannot be applied with a fixed threshold due to instrument design: The 3.9?m band is exceptionally wide and overlaps with the 4?m CO2 absorption band. Therefore, the emissivity difference varies with the length of the slant atmospheric column between sensor and object. To account for this effect, the new technique presented in this paper is based on the dynamical extraction of emissivity difference thresholds for different satellite viewing zenith angles. In this way, varying concentrations of CO2 and column depths are accounted for. The new scheme is exemplified in a plausibility study and shown to provide reliable results.