Sharing and utilizing hydro-meteorological data in MAHASRI

MASUDA, Kooiti1Frontier Research Center for Global Change

Japan Agency for Marine-Earth Science and Technology (JAMSTEC)Showa-machi, Kanazawa-ku, Yokohama 236-0001, Japan

e-mail: masuda@jamstec.go.jp A personal (not official), experimental (not to be sustained) web site

http://mahasri.world.coocan.jp/

MAHASRI/HyARC WorkshopDa Nang, Viet Nam, 5-7 March 2009

Global environment: constraints to the human society(nearly one-way relationshipwhen the human population was small)

1. Background In What world do we live?

Global environmental issue: feedback (human - nature - human)

Climate change isa serious issue toecosystem, becauseit is compoundedwith human directinfluence (land usechange, fisheries,etc.).

Climate change isa serious issue tothe human society,because it iscompounded withthe issue of the limitof energy resourceswhich it can use.

We must live in a changing climate. We must adapt to it.The society needs to prepare for likely variability of climate.The likely change is warming in the global average.Regional changes will be more complicated.Cases of both “too much water” and “too little water” are likely to become more frequent.

But also, as far as human activity is changing climate, we have a hope to reduce, in other words mitigate, the change.We must shift energy resources to renewable ones:solar radiation, winds, water flow, biomass, etc.... parts of the energy cycle of the climate system(exception: geothermal energy, tidal energy)Not concentrated, not uniform, intermittent.Local societies must have knowledge of their availability.

Adaptation and mitigation to climate change

We wish to anticipate climate change.But we cannot predict some factors of climate change (e.g. volcanic eruption) years ahead.Also we think that human behaviour may change with the knowledge of climate change.So scientists decided to cut the loop and tentatively think the following way. • Climate is a physical system of the atmosphere and the ocean, and the concentration of carbon dioxide is an external factor.• The concentration is determined by the global carbon cycle coupled with the global ecosystem, where anthropogenic emission is an external factor.• The emission is determined by the socio-economics of the world.... Climate simulations designed this way are called climate change projections rather than predictions.

Climate change projection [skipped]

Climate change projectionn [skipped]

2. Needs to data, information and knowledge about climate and hydrological cycle

* Adaptation to climate, including reduction of damages due to disasters* Utilization of renewable resources

Why do we need to share/exchange information internationally?(in addition to the operational exchange coordinated by WMO etc.)

* Atmosphere, ocean and hydrological cycle are borderless.* Climate is changing, so we want to learn about different climatic conditions.

Non-experts need to know something, and experts should help them.

* Additional viewpoints are helpful in developing ideas and reducing errors.An example of the area where collaboration seems fruitfulEffects of winter monsoon + mesoscale systems + complex topography in producing local scale distribution of clouds and precipitation

Central VietnamNov.-Dec. Heavy rain (dry in Laos)Jan.-Feb. Light rain, foggy weather

JapanWinter Snowfall in NW side, dry in SE sideSummer “Yamase” foggy weather in NE side

Situation (e.g. Coriolis parameter) is not the same!

* We need to evaluate applicability of technology developed elsewhere.

Privacy and security (anti-terrorism etc.)Serious with high-resolution data (< O(10 m) ?)

Intellectual property rights (copyright, patent, database right of EU etc.)

Information and knowledge are intrinsically “public goods”(non-exclusive).To motivate creation of knowledge, governments grant rights to the creators.The rights have limited terms and the knowledge go to public domain after that.Creation of knowledge actually rely on the public domain very much.

I think that environmental data are better to be put in the public domain.But the question of how to fund creation of public good remains.•The government -- How we can get support by taxpayers?•Volunteers – can we follow the model of open source software?

Recognition of the originator (author, government agency, etc.)(An element of the notion of copyright, but different from property right)Scientists recognize authors of scientific papers. But how about those of data?It is not practical to manually track back to the origins of complex products.We should establish information systems which remember the origins.

World Data Centershttp://www.ngdc.noaa.gov/wdc/Being reorganized intoWorld Data Systemhttp://www.world-data-system.org/under International Science Council (http://www.icsu.org/)

Initially established to support International Geophysical Year (1957-58).Managed data of Global Atmospheric Research Program (1970s).

USA … Meteorology (NOAA NCDC), Snow and ice (Colorado Univ.) etc.Russia … Meteorology (RIHMI Obninsk) etc.China … Meteorology (CMA NCDC), Glaciology (CAEERI Lanzhou) etc.Japan, India … geomagnetism etc. but not one related to climate/hydrology. Germany … Climate (MPI Hamburg)

I think that Asia should have an active center (or a series of centers) to participate in the system.

3. Factors needed to be considered with respect to data sharingA legacy that contributed to free and open data exchange

4. GEWEX Asian Monsoon Experiment (--March 2005)4 regional sub-projects. Intensive Observing Period: Summer 1998 (Siberia Summer 2000)http://www.hyarc.nagoya-u.ac.jp/game/ ... home pagehttp://gain-hub.mri-jma.go.jp/ ... data portal (K. Takahashi)http://hydro.iis.u-tokyo.ac.jp/GAME-T/GAIN-T/ ... GAME-Tropics data (Y. Agata)

Example of data products

Operational meteorological data(mainly daily precipitation)1997 - 2002

5. Data management in MAHASRI

•Experimental observations (including prototypes of future operational products)• Making data available … responsibility of each observing project• Common catalog [not yet designed]• Arrangement to be used in AMY (2008-2009) Reanalysis [feasible?]• Are there a data center who can take over the data from finished projects?

•Operational observations (Some ideas of a group including myself. Maybe beyond capacity, need prioritize)(3) More extensive (in time and space) collection of daily precipitation(2) Multi-decadal daily (or more detailed) series of surface meteorological obs.

(temperature, humidity, wind speed, sunshine duration)=> evaluation of energy and water fluxes at surface application to agricultural productivity assessment evaluation of how climate models can reproduce the recent past (1) High-resolution data for process studies (AMY 2008-2009 year-round)(a) Synoptic observations at 3-hour time interval(b) Radiosonde profiles more detailed than “standard” and “significant”

6. Issues on data quality (from a research user’s viewpoint)One important way of improvement of data quality is that data are used and comments of the user are fed back to the producer. e.g. FGGE (First GARP Global Experiment, 1978-1979) Level 2b data (collection of observations), 3b data (gridded data)

Data contain not only quantitative errors but also qualitative errors (mistakes).We need multiple ways to tackle them.The most important seems to be (man-machine interactive) visualization.Situations not easily represented by numerical values need good description•Missing observations•No precipitation and trace precipitation•Documentation of units and precision (number of significant digits)

Identification of location across multiple datasets is a difficult task.(Actual move, administrative change, transliteration, units of lat./long. …)

When producing a unified data set, loss of information due to unit conversion etc. is inevitable. We should also keep the original version of the source data so that a user can trace back when needed.

Document of methods of observation is important even for standard elements, and necessary for non-standard items, e.g. ground temperature, evaporation.

7. JAMSTEC’s activity in DIAS

16

Tokyo University

Core SystemData

Infrastructure

Support System for

Establishing Data

Interoperability Arrangements

Data Preprocessing

DB

Web

Data Archive System

× ×

Data Integration and Analysis System <DIAS>

（ implementation test phase ）

ConversionSystem

＜ DIAS Practical Application Systemfor Earth Observation Data Service ＞

（ implementation phase ）

FY2011~

In DIAS, JAMSTEC developsPractical Application System for Earth Observation Data Service.

(Materials prepared by Mr. Hayata)

17

JAMSTEC provides the sample productsJAMSTEC provides the sample productsResearch Projects

JAMSTEC

Images

Digital data

Website

Descriptions

Consideration of a long-term sustainable service

Gathering source data,Metadata format,Human resources & Facilities,etc.

Ocean & Fishery Resources

Research Institutes,Public offices & Private companies

Water Cycle -Asia monsoon-

Water Cycle -Eurasian Cold Region-

EcosystemContinuous

deeping

SampleProducts

Feedback into researchers

Sample Data

Feedback

Needsanalysis

Documents

D I A S

User

Feedback into services

18

Product imageProduct image 　（Ａｓｉａｎ　Ｍｏｎｓｏｏ　（Ａｓｉａｎ　Ｍｏｎｓｏｏｎ ｎ water cyclewater cycle ））

A ． Gridded precipitation data (meso-scale) by radar and raingauge

Indonesian west SumatraEvery 30 minutes, one month (November 2006) Approximately 2.8km lattice, 80 km radius

B ． Relationship between Intraseasonal variation(Madden-Julian Oscillation) and local precipitation in Indonesia

C ． Gridded precipitation data (large scale) by raingauge

(

DAY

1-degree lat./long. grid, monthly, 1997-2002

D. Estimation of solar radiation

(Kamimera)

(Masuda)

(Hamada)

(Xu))

　 　 Global-local connection of intraseasonal oscillationDiurnal variation of precipitaiton vs.phase of Madden-Julian Oscillation

OLR & 850 hPa wind www.bom.gov.au/bmrc/clfor/cfstaff/matw/maproom/RMM/

t

1. Western hemisphere

3. Indian Ocean

5. Maritime continent

7. Western Pacific

Precipitation anomalyRed < -20%; blue > +20%

1.

5.

3.

7.

Amp: radius of circlePhase: direction of arrow

Heavy rain events in Sumatra (station data)vs. phase of Madden-Julian Oscillation

Daily precip. vs. MJO Frequency of heavy rain

1

2 3

4

567

8

Occurrence of daily precipitation in the uppermost 1 % quantile at each of 14 stations in Sumatra,sorted by the phase of MJO..

Line: Cases of MJO amplitude > 1.0

Indian Ocean

MaritimeContinent

WesternPacificAfrica

(Jun-Ichi HAMADA)

21

JAMSTEC provides the sample productsJAMSTEC provides the sample productsResearch Projects

JAMSTEC

Images

Digital data

Website

Descriptions

Consideration of a long-term sustainable service

Gathering source data,Metadata format,Human resources & Facilities,etc.

Ocean & Fishery Resources

Research Institutes,Public offices & Private companies

Water Cycle -Asia monsoon-

Water Cycle -Eurasian Cold Region-

EcosystemContinuous

deeping

SampleProducts

Feedback into researchers

Sample Data

Feedback

Needsanalysis

Documents

D I A S

User

Feedback into services