English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Smart Ice Cloud Sensing (SMICES) is a small-sat concept in which a primaryradar intelligently targets ice storms based on information collected by alookahead radiometer. Critical to the intelligent targeting is accurateidentification of storm/cloud types from eight bands of radiance collected bythe radiometer. The cloud types of interest are: clear sky, thin cirrus,cirrus, rainy anvil, and convection core.  We describe multi-step use of Machine Learning and Digital Twin of theEarth's atmosphere to derive such a classifier. First, a digital twin ofEarth's atmosphere called a Weather Research Forecast (WRF) is used generatesimulated lookahead radiometer data as well as deeper "science" hiddenvariables. The datasets simulate a tropical region over the Caribbean and anon-tropical region over the Atlantic coast of the United States. A K-meansclustering over the scientific hidden variables was utilized by human expertsto generate an automatic labelling of the data - mapping each physical datapoint to cloud types by scientists informed by mean/centroids of hiddenvariables of the clusters. Next, classifiers were trained with the inputs ofthe simulated radiometer data and its corresponding label. The classifiers of arandom decision forest (RDF), support vector machine (SVM), Gaussian na\"ivebayes, feed forward artificial neural network (ANN), and a convolutional neuralnetwork (CNN) were trained. Over the tropical dataset, the best performingclassifier was able to identify non-storm and storm clouds with over 80%accuracy in each class for a held-out test set. Over the non-tropical dataset,the best performing classifier was able to classify non-storm clouds with over90% accuracy and storm clouds with over 40% accuracy. Additionally both sets ofclassifiers were shown to be resilient to instrument noise.

Using Unsupervised and Supervised Learning and Digital Twin for Deep Convective Ice Storm Classification