The first version of the primitive OWL-ontology of collections of climate and meteorological data of Institute of Monitoring of Climatic and Ecological Systems SB RAS is presented. The ontology is a component of expert and decision-making support systems intended for quick search for climate and meteorological data suitable for solution of a certain class of applied problems.
Today every large meteorological center uses original meteorological models for calculation of climate and meteorological parameters, which can differ both in the level of detail and set of calculated values of physical parameters. During the reanalysis of a meteorological situation, key meteorological parameters corresponding to measurements at weather stations are usually taken into account.
The results of climatic numerical simulation,
weather forecast, or reanalysis of meteorological fields are
collections of meteorological parameters that
characterize the state of the atmosphere. They are represented by
data arrays in common formats, e.g., grib [
At Institute of Monitoring of Climate and Ecological
Systems SB RAS (IMCES SB RAS), the data
processing environment [
When using climate data from different collections
of numerous data manufacturers, the problem arises of
ambiguous identification of physical parameters from
these collections. The sense of physical parameters in
the collections agrees with physical parameters advised
by World Meteorological Organization (WMO). They
are described in the taxonomy of the WMO ontology
Codes Registry [
The ontology description of data collections in the form of a primitive (simplified) formal OWL-ontology is intended for the selection of data collections within an expert system, which can be used during solution of an applied task of an object domain.
The ontology approach selected for the solution of the problem stated consists in the following. An ontology description is constructed for an applied problem. In addition to the physical statement, the description should include the mathematical statement of the task, i.e., a mathematical model with equations. Variables, which conform the WMO classification, and limitations are described in the form of an OWL-ontology. On the one hand, the set of parameters includes common meteorological parameters, such as sea level pressure, surface pressure, air temperature and humidity, wind speed and direction, and so on. This allows comparison of the computed values with the weather station measurement results. On the other hand, both meteorological and climate models supplemented by an applied task compose a component of a more complex model, where the results of prognostic calculations by climate/meteorological parameters are used for the solution of applied problems in different fields of human activity. This, in turn, enriches collections of climate and meteorological data with values of new physical parameters. 2 Virtual data processing environment
Approaches used in the creation of the prototype
of a subject virtual data processing environment
(VDPS) for the analysis, estimation, and forecast of
the impacts of global climate changes on the natural
environment and climate of a region were mainly
developed during the design of the “Climate” web
GDS [
The GDS Web Client [
Today, the VDPS prototype combines data collec-tions (reanalyses and climate simulation results and weather station measurements) within the unified geo-portal, supports the statistical analysis of archive and required data, and provides access to the WRF and «Planet Simulator» models. In particular, a user can run a VDPS-integrated model, preprocess the results, pro-cess them numerically and analyze, and gain the results in graphical representation. The prototype provides for specialists that participate in a multidisciplinary re-search process prompt tools for integral study of climate and ecological systems on the global and regional scales. With these tools, a user that does not know programming is able of processing and graphically representing multidimensional observation and simulation data in the unified interface with the use of the web browser.
Support of the following data sets is built in the prototype: NCEP/NCAR reanalysis, ed. II, JMA/ CRIEPI JRA-25 reanalysis, ECMWF ERA-40 reanalysis,
ECMWF ERA Interim, MRI/JMA APHRODITE’s Water Resources Project data, DWD Global Precipitation Climatology Centre data, GMAO Modern EraRetrospective analysis for Research and Applications (MERRA), reanalysis of the joint Project «Monitoring atmospheric composition and climate (MACC)», NOAA-CIRES Twentieth Century Global Reanalysis, ver. II, NCEP Climate Forecast System Reanalysis (CFSR), simulation results obtained with the use of global and regional climate and meteorological models. Observation data from weather stations from the territory of the former USSR for the 20th century included in the PostGIS database are also accessible.
Data processing 1. Statistical characteristics of meteorological parameters: sample mean, variance, excess, median, minimum and maximum, and asymmetry. 2. Derived climate parameters: vegetation period duration, sum of effective temperature, Selyaninov hydrothermal coefficient. 3. Periodic variations: standard deviation, norms, aberrations, amplitudes of diurnal and annual variations. 4. Non-periodic variations: duration and repeatability of atmospheric phenomena with meteorological parameters below or above the limits specified at different time points.
Then a user can either analyze the results or continue adding new layers on the map. To study the results, the user is provided for a possibility of selecting a geographical region, scaling, getting values from all layers at a point, additionally processing earlier results (e.g., comparison between data from different layers). In addition to the direct analysis of geophysical data, a used can carry out joint researches with other user, share the results, and use proper data collections in the processing. In general, this hardware-software complex provides for distributed access, processing and visualization of large collections of geospatial data with the use of cloud technologies.
The data processing environment “Climate” developed at IMCES SB RAS limits possibilities of users by local software applications. A current task is to extend the environment by external user applications. For this, the corresponding problems should be specified in general. Below we describe one of possible classes of problems connected with decision-making.
The “Climate+” virtual information platform includes collections of meteorological and climate data. It is intended for the data representation with the use of GIS technologies. Its further development is oriented to providing researchers possibilities of using selected data sets or their parts as input data. Most collections include data related to some (not all) spatiotemporal objects of the Earth; different collections often include different sets of physical parameters. To search for required spatiotemporal objects and their meteorological and clime characteristics, it was necessary to create a corresponding expert system on the basis of a knowledge base on spatial objects of the data collections and their parameters.
In this work, we discuss questions of creation of a
knowledge base for the expert system. The main
problem which has been solved is substantiation of the
reduction problem solution [
The OWL DL language [
The taxonomy of physical parameters allows forming sets of properties of spatiotemporal objects of a domain for solution of specific applied tasks. This taxonomy is used in the OWL-ontology for T-box construction.
When developing the decision-making support
system on the basis of both meteorological and climate
data, the parameters should be matched. Therefore, the
WMO classification in version [
There are climatic and meteorological resources
[
In our OWL-ontology of climate information
resources, we created classes and individuals that
correspond to names of meteorological parameters, e.g., the
Meteorological_Products class and subclasses,
according to [
For the INMCM4 collection, which corresponds to
output data of the INMCM4 climate model of general
atmospheric and ocean circulation [
The OWL DL developed and formalized ontology of climate information resources describes the current state of collections of data arrays of the data processing environment as one of the main Russian information resources on climate data. Numerical data are represented by data arrays that are stored in netCDF files. The data arrays are grouped in data sets. All data arrays in a set should: (a) be received at one temporal or spatial grid; (b) cover the same time interval; (c) be received under the same simulation or observation conditions (if possible); (d) be represented by a set of netCDF files, which include the same physical parameters. The data sets are grouped in data collections. A data collection is an ensemble of data sets received by an organization within a project, but represented on different spatial or temporal grids or for different model scenarios. In particular, a collection can consists of the only data set.
The basic classes in the OWL-ontology are: Collection, Spatiotemporal_object, Organization, Data_set, Data_array, Scenario, Spatial_resolution, Physical_quantity, Physical_quantity_values, Unit, Longitudes_array, Time_step, Latitudes_array, Height_levels_array, and Times_array. The spatiotemporal system is a four-dimensional object determined by arrays of numerical values of longitudes (Longitudes_array), latitudes (Latitudes_array), height levels (Height_levels_array), and time labels (Times_array), which are subclasses of the class of the list of values of a physical parameter and, therefore, numerical arrays of one physical parameter (Physical_quantity) in certain measurement units (Unit). They can be described by: the number of members of the array of a physical parameter (has_number_of_values), its minimal value (has_minimum_value) and maximal value (has_maximum_value), or by numerical values of the parameter (has_value). A data array (Data_array) is an ordered list of numerical values of a physical parameter (Physical_quantity), as a property of the spatiotemporal system (has_spatiotemporal_system), at each 4D point (longitude, latitude, height level, and time) of the spatiotemporal system (Spatiotemporal_system). In the OWL-ontology, a data array (Data_array) is a subclass of the class Physical_quantity_values and, hence, is a numerical array of values of one physical parameter (Physical_quantity) in certain measurement units (Unit); it is described by the number of members (has_number_of_values), maximal values (has_minimum_value) and minimal values (has_maximum_value) of the physical parameter. A data array (Data_array) belongs (has_data_array) to a data set (Data_set), which differs from other data sets by the model scenario (Scenario), spatial resolution (Spatial_resolution), time step (Time_step), and belonging to one collection (Collection). A data collection (Collection) consists of (has_data_set) data sets (Data_set) and belongs (has_organization) to one organization (Organization). The OWL properties of the climate data ontology are represented in Tables 1 and 2.
Definitions of object properties are given in first three rows of Table 1; their unique identifying properties, in the fourth row; the range of definition (the first row) and range of values (the third row) are specified for each property. Definition of the data array properties are given in the first three rows of Table 2; unique identifying properties are given in the fourth row. The range of definition (the first row) and range of values (the third row) are specified for each property from the second row.
ovals; literal values are given in rectangles; the arrows show properties with unique identifiers in small rectangles, taken from Tables 1 and 2. Three arrows mean probable property cardinality higher than unity. Three overlapped ovals mean probable number of individuals of the OWL-ontology larger than unity. The individual “Data_collection” is connected by the property “has_data_set” with the individuals “Data_set”, each of which is connected by the property “has_data_array” with individuals ”Data_array”.
The domain analysis of climate numerical data arrays of the “Climate+” platform, stored as NetCDF files, allows the description of a primitive ontology of climate data of this platform in the OWL DL language. The primitive ontology is a simple and easily extended systematization of information resources required for the further work on the development of the decision-making support system.
“Climate+” platform the software has been developed for the formation of the fact-based block (A-box). An A-box has been formed for the climate data ontology using this software. Facts have been retrieved from the analysis of 80 Tb of climate data from the “Climate+” platform over 13 numerical data collections, which include 36 data sets and 793 data arrays. All the climate data collections include description of 170 spatiotemporal systems and 156 physical parameters that characterize properties of these systems.
The prototype of subject virtual data processing environment has been developed to provide for researchers, specialists, and people that make decisions an access to different geographically distributed and georeferenced resources and climate data processing services via a typical web browser. It includes a geoportal, systems for distributed storage, processing, and providing of spatial data and results of their processing. In particular, it allows the simultaneous analysis of several subject sets of climate data with the use of up-to-date statistical methods and, thus, revealing the impacts of climate changes on ecological processes and human activity. After finishing the work on the prototype, different interactive web tools are to be developed for the profound analysis of climatic variables and their derivatives provided by the subject geoportal.
spatial datasets, including observation and reanalysis data, for the spatiotemporal analysis of recent and probable climate changes, with the special focus on extreme climate phenomena in northern latitudes.
The primitive OWL-ontology of climate and meteorological collections of IMCES SB RAS is constructed; it can be used for the search and selection of data for classes of applied problems in coupled decision support systems. The matching of physical parameters of applied tasks with IMCES SB RAS collections is carried out in WMO accepted terms.
The authors thank the Russian Science Foundation for the support of this work (developing of webservices and solution of reduction problems) under the grant No16-19-10257. We also thank Russian Foundation for Basic Research (16-07-01028) for the support of work (conceptualization of domains) partially described in the sections 4, 5, 6 of the article.