Data Integration and long-term planning of the Observing Systems as a cross-cutting process in a NMS

(1)

Bundesamt für Meteorologie und Klimatologie MeteoSchweiz

Data Integration and long-term planning of the

Observing Systems as a cross-cutting process

in a NMS

ECAC Zurich, Setpember 15 2020

Ch. Häberli

Deputy Head Climate Division/Head Meteorological Data Coordination

& E. Grüter, M. Brändli, M. Musa, M. Kube and L. von Dach

(2)

Agenda

• Challenges

• Basic Strategy, Concepts and Blueprints

• An Example: Improving the precipitation picture

• Consequences

• Summary

(3)

winter

summer

The Seamless Time Dimension

(4)

Data Procurement Strategy

• MeteoSwiss operates the backbone of the monitoring systems

in Switzerland and the nationwide data integration platform

for measurements in the atmosphere

• The data procurement and data processing strategy of

MeteoSwiss is an integration strategy.

• Aspects of integration

– various observing

technologies and platforms

– observing systems

outside MeteoSwiss

– data und meta data



comprises all aspects of „Enterprise Data Integration“

(5)

Data Procurement Strategy

nationwide integration platform

for meteorological&climate data

standards

continuity

transfer

assimilation

meta data

_control

quality

(6)

Agenda

Challenges

Basic Concepts

Solutions

Consequences

Summary

2000 Surface weather stations in Switzerland

public sector: 1628

- MeteoSwiss: 645

- other: 983

(7)

da

ta

s

o

ur

ces

sta

g

ing

a

rea

da

ta

s

to

ra

g

e

Met

ad

ata

Repo

sitory

Met

ad

at

a

Man

ag

em

en

t

to

o

ls

Database for the analysis

of point data

Calc, Check, Data transformation & integration

Data delivery and

products (on-/offline)

Weather Forecast

Data for the Public

Internet

Durable data archive

Light-ning

data

Radar

CAT

Data

Mart ..

Model Validation

Research

Climate Services

da

ta

a

na

ly

si

s

Observ.

systems

Upper

Air

data

Aviation

data

Database for

gridded data

Data extraction & interfaces (API, DB Tools, Web access, GIS, R …)

QC

Tools

VERA

Warnings/

Alarms

NinJo point

data servers

CM

SAF

INCA

NWP

T

L

T

L

Partner

Observ.

systems

Database for the processing

of point data

(8)

Ex 1: Combination Radar – surface in situ precipitation

69 Gauges

_>40mm/h

Red hues

1 Radar

138 Gauges

Here, we artificially double the

number of gauges by creating

for each existing automatic

raingauge a hypothetic random

daughter-station some km to

the east and to the north.

(9)

Ex: Combination Radar – surface in situ precipitation

radar data

+

high spatial and temporal

resolution (1km, 5min)

-

uncertainty in absolute

values, often bias

station data

+

accurate values at

stations

-

coarse resolution (15km,

24h)

-

uncertainty between the

stations

(10)

Ex: Combination Radar – surface in situ precipitation

stations

combination

radar

predict

ed

v

alues

predict

ion

err

or

(11)

Consequence  monitorig systems

• seamless time dimension:

– lasting operation

– keep the system adaptive (to new technologies, monitoring

needs...)

– multifunctional systems

• automation and modernization

• various platforms (in situ and remote sensing) 

syntactic and semantic interoperability

(12)

Consequence  data integration

• keep the integration platform adaptive (allow analyses with

data from 1755 till now) but keep 1 data source for all

applications

• allow permanently

editing

of the data set

• give highest importance to

meta data

• d

ata owner

≠ product owner

• world wide data

exchange  data stewardship/quality?

• permanent

increase of data volume

• real-time, 24h/365d, max downtime: 15‘

• continuous new requirements, technologies, regulations

(WMO, ICAO) but no classical requirement life cycle



automation and modernization

(13)

Consequence  Organisation

(14)

Consequence  Organisation

Main tasks of the cross cutting unit

• Long-term (strategic) planning of the data

procurement chains  architecture blueprints and

standards for projects

• Support for projects: fitting the projects into the

overall architecture; change management for the

architecture

• Support in daily operations: operations coordination

of the data warehouse system

(15)

Conclusion

• Maintaining a monitoring system for the

atmosphere is a permanent process and challenging

cross-cutting process with a strong Integration

aspect.

• The main challenge is to establish a rigorous

evolution process by:

1. regular analysis of the requirements (comprehensive!)

2. regular review of observation capabilities

3. gap identification and budget review

4. implementation and operation

for all parts of any „data chain“

(16)

We have achieved…

… a balanced mix of observing technologies and

platforms - including the access to weather

data from sources outside the national service

… a balanced mix between advanced,

state-of-the-art and conventional technologies

… a highly integrated data management system

(17)