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:

Floods cause significant damage to human life, infrastructure, agriculture, and the economy. Predicting peak runoffs is crucial for hazard assessment, but it is challenging in remote areas like the Andes due to limited hydrometeorological data. We utilized a 300 km2 catchment over the period 2015–2021 to develop runoff forecasting models exploiting precipitation information retrieved from an X-band weather radar. For the modeling task, we employed the Random Forest (RF) algorithm in combination with a Feature Engineering (FE) strategy applied to the radar data. The FE strategy is based on an object-based approach, which derives precipitation characteristics from radar data. These characteristics served as inputs for the models, distinguishing them as “enhanced models” compared to “referential models” that incorporate precipitation estimates from all available pixels (1210) for each hour. From 29 identified events, enhanced models achieved Nash-Sutcliffe efficiency (NSE) values ranging from 0.94 to 0.50 for lead times between 1 and 6 h. A comparative analysis between the enhanced and referential models revealed a remarkable 23% increase in NSE-values at the 3 h lead time, which marks the peak improvement. The enhanced models integrated new data into the RF models, resulting in a more accurate representation of precipitation and its temporal transformation into runoff.

Abstract:

n tropical forests, herbivorous arthropods remove between 7% up to 48% of leaf area, which has forced plants to evolve defense strategies. These strategies influence the palatability of leaves. Palatability, which reflects a syndrome of leaf traits, in turn influences both the abundance and the mean body mass not only of particular arthropod taxa but also of the total communities. In this study, we tested two hypotheses: (H1) The abundance of two important chewer guilds (‘leaf chewers’ and ‘rostrum chewers’), dominant components of arthropod communities, is positively related to the palatability of host trees. (H2) Lower palatability leads to an increased mean body mass of chewers (Jarman-Bell principle). Arthropods were collected by fogging the canopies of 90 tropical trees representing 31 species in three plots at 1000 m and three at 2000 m a.s.l. Palatability was assessed by measuring several ‘leaf traits’ of each host tree and by conducting a feeding trial with the generalist herbivore Gryllus assimilis (Orthoptera, Gryllidae). Leaf traits provided partial support for H1, as abundance of leaf chewers but not of rostrum chewers was positively affected by the experimentally estimated palatability. There was no support for H2 as neither leaf traits nor experimentally estimated palatability affected the mean body mass of leaf chewers. The mean body mass of rostrum chewers was positively related to palatability. Thus, leaf traits and experimentally estimated palatability influenced the abundance and mean body mass of chewing arthropods on the community level. However, the data were not consistent with the Jarman-Bell principle. Overall, our results suggest that the palatability of leaves is not among the dominant factors influencing abundance and mean body mass of the community of chewing arthropod herbivores. If other factors, such as the microclimate, predation or further (a-)biotic interactions are more important has to be analyzed in refined studies.

Abstract:

There are many uncertainties in radiation fog forecast. Continuous effort is being made to improve the forecast. A supplementary and increasingly popular approach to numerical weather forecast is the forecast with machine learning (ML) algorithms. While numerical weather forecast is based on mathematical models with partial differential equations, ML algorithms take a more heuristic approach. The latter strategy calls for three steps. Precise data preprocessing is the initial step. This implies that after preprocessing, the dataset must contain the forecast-relevant information in a way that the algorithm can learn from it. This is not a trivial step because it necessitates a thorough understanding of the fundamental principles underlying radiation fog. Even when the relevant information is contained in the data, it is not always evident, especially in severely unbalanced fog datasets. The best strategy to achieve a pleasing result may therefore not be to simply feed the algorithm all the data and variables that are available. So that the appropriate dynamics may be detected by the algorithm, the data and information should be adjusted accordingly. The second step is the data splitting into training, validation and test datasets. The ability to predict fog is driven by the temporally linked process that describes the ongoing change in atmospheric state but in order to guarantee constant independence between the training, validation and test dataset, the data splitting method must consider this temporally linked information between the individual datapoints. Otherwise, the algorithm’s forecast accuracy can be based on the temporally correlated information content of the individual data points. The third step is the interpretation of the model scores. When looking at the forecast score alone, it is a very abstract number that does not directly allow a statement about the forecast performance of the model. In order to evaluate the model performance, two baselines are of relevance: algorithm complexity and dataset complexity. A baseline for algorithm complexity justifies the chosen algorithm and also classifies the model performance. A baseline for dataset complexity also classifies the model performance and enables a better comparability of different datasets. Following these principles, our current objective is to improve the ML based fog forecast with XGBoost for a forecasting period up to four hours for the station in Linden-Leihgestern (Germany). The training and evaluation are based on the Expanding Window Approach (Vorndran et al. 2022) that considers the autocorrelation of a fog time series and maintains the temporal order during both training and evaluation. The evaluation is based on a score for each of the following categories: Overall performance, fog formation, and fog dissipation. The results are set in relation to different baselines to evaluate the performance and the dataset complexity. Building on this scheme, newly preprocessed data led to an improvement in the prediction of radiation fog for the station in Linden-Leihgestern. We will present the most recent findings from our research.

Abstract:

This study addresses transpiration in a tropical evergreen mountain forest in the Ecuadorian Andes from the leaf to the stand level, with emphasis on nocturnal plant-water relations. The stand level: Evapotranspiration (ET) measured over 12 months with the Eddy-Covariance (ECov) technique proved as the major share (79%) of water received from precipitation. Irrespective of the humid climate, the vegetation transpired day and night. On average, 15.3% of the total daily ET were due to nocturnal transpiration. Short spells of drought increased daily ET, mainly by enhanced nighttime transpiration. Following leaf transpiration rather than air temperature and atmospheric water vapor deficit, ET showed its maximum already in the morning hours. The tree level: Due to the humid climate, the total water consumption of trees was generally low. Nevertheless, xylem sap flux measurements separated the investigated tree species into a group showing relatively high and another one with low sap flux rates. The leaf level: Transpiration rates of Tapirira guianensis, a member of the high-flux-rate group, were more than twice those of Ocotea aciphylla, a representative of the group showing low sap flux rates. Representatives of the Tapirira group operated at a relatively high leaf water potential but with a considerable diurnal amplitude, while the leaves of the Ocotea group showed low water potential and small diurnal fluctuations. Overall, the Tapirira group performed anisohydrically and the Ocotea group isohydrically. Grouping of the tree species by their water relations complied with the extents of the diurnal stem circumference fluctuations. Nighttime transpiration and hydrological type: In contrast to the isohydrically performing trees of the Ocotea group, the anisohydric trees showed considerable water vapour pressure deficit (VPD)-dependent nocturnal transpiration. Therefore, we conclude that nighttime ET at the forest level is mainly sourced by the tree species with anisohydric performance.