Morning,

good afternoon, good evening. Uh first

of all, I'd like to thanks my colleagues

uh to join us today as speakers and

especially I I would to thank everybody

that uh spend your time here with us

engaging with this uh so important

thematic topic on geographic

segregation.

Before moving uh forward uh I would like

to briefly introduce the expert group on

integration of statistical and jatial

information uh expert group that now I'm

greatly have the honor to serve as a

co-chair. My name is Claud Stanner. I

work from the Brazilian Institute of

Geography and Statistics.

at the

next please.

Next slide please.

>> So this is the H staring group of the

expert group on integration statistical

spatial information. Uh we have a

representative from all around the the

world. uh this is um only the the term

group but you have also members from

more than 40 countries uh spread all

around the uh the world. This is a

statistical commission expert group.

This expert group is uh designing to

support uh the work of the the

commission the work of the member states

in terms of uh integration of

statistical special information in all

kind of statistical uh domains.

uh the the the expert group is uh also

uh

works uh to support the the sustainable

de development goals uh uh in in many

different ways. It helps to identify

territorial inequalities. It's crucial

to really leave no uh one uh behind. So

we are here to help you. We are here to

help in every country. We're here to

help every expert group, every working

group from statistical uh domain to

implement uh better integration of

statistical JP special information to

guarantee a geographic disagregated

uh data.

Next slide please.

So I'd like to speak briefly with this

session uh the 2026 session of the

statistical commission. This year the

expert group is presented the for

endorsement the second edition of the

global statistical joatial uh framework.

This is our main uh work our main

publication our main framework. uh the

GSTs provide a complete guidance in how

to integrate statistical injection

information. This second edition

has a lot of improvements in terms of

language in terms of usability of the

the framework but not change the the

sensors principles of the framework. It

remains the same. But I invited

everybody to to download the second

edition from the strategic commission

site and show uh how we reorganizing the

the framework and how it can provide

guidance

on integration and statistical and

geographic information. We also

presented uh a first of a paper on

geographical disagregation. This this

paper is a base of this webinar here. So

I also invited everybody to to see this

paper on the statistical commission uh

websites. This year we also present the

our strategy toward 2030 and how you can

imagine we can accelerate the the

implementation of integration

statistical space information. how we

can imagine the this this process.

Next, please.

Uh before I move uh forward, I'd like to

show you a very brief example and how

crucial geographic disagregated data

are. uh you as you see in this uh

picture uh there is a a a hard boundary

between a very wealthy area and a very

poor area that you call it as L area in

Brazil that call it favlla that that may

be a problem in terms of statistical

production next please

that's another picture the same

situation this other situation is in

lamei in in it's a real beach you have a

very of community close to the beach and

a very poor community on the hills

behind the the buildings. Next, please.

In this case, we have a a problem with

statistics. If you see uh this boundary

uh on the this neighborhoods that

include a very wealth community and a

very poor community in the same area has

42.9% of it of its population with

higher education that's uh means but

next please

next

that's a kind of track of the average

why I I'm saying this next is

the the real world is if you separate

the data from the poor areas from the

Islan areas from the uh to to the very

wealthy communities the reality is the

wealthy commun the non-Islan area of

this part of the city has 49.9%

of the it with high and has only has

only 1.3% % uh of this population with

high education. So that's the re

reality. If you don't have the

geographic segregation, if you don't

have uh this the right geograph you you

can see the reality you and we will

leave a lot of people behind.

Uh

next please.

So uh now I'd like to invite my

colleague Josh from the US sense below

that he will talk a little bit more

about the expert group expert group

working and and we'll conduct a debate

on this session. Josh the floors is

yours.

>> Fantastic. Thank you Cladio and thank

you for that example. I think that

really brings home exactly what we were

talking about the importance of that

disagregation of data uh using the

smallest geographies that we can to

produce uh relevant statistics and how

often the those uh

uh actionable statistics just get

are not able to be produced given the

geographies we use today and how

important it is because decisions happen

at the regional and municipal scales as

we know and these national averages just

as Cladio said uh really wash out um and

hide the local disparities and

vulnerable populations and so most

policy implementation and service

delivery happens as we know at the

regional, municipal, community levels

where decision makers need locally

relevant data and without that localized

evidence of that local data data

governments risk targeting resources

inefficiently. We I'm sure we all have

examples in our own communities where

communities most in need miss this uh

important uh resource allocations. And

so to put it simply, if we want to leave

no one behind, we must ensure that our

data systems leave no place behind.

And so we have I I want to just take a

moment to to acknowledge the many

colleagues and fellow travelers that we

have grappling with these issues and and

recognize the amazing work that's been

done to date already um by the working

group on geospatial indicators of the

inter agency and expert group on SG

indicators. We we have uh Mary Smith who

co-headed and championed the paper

that's brought us here today uh from the

working group on geospatial indicators.

Um but especially I wanted to recognize

the team behind the small area estimates

primer, the geospatial small uh small

estimates primer. Um we have colleagues

from Esqua today who've done amazing

work uh building on these concepts. Um

and many other resources for example

from the inter secretariat uh working

group on household surveys. Um, so I

hope what we present today, our expert

group's contributions are helpful to the

community and all the work that's been

done uh to provide additional resources

to help solve these issues and and

concepts. Um, because we know that

disagregation is hard, right? That's why

we're all here today. That's why so many

people are working on this. Uh, first,

many statistical administrative data

sets were never designed for spatial

integration. uh you know they may have

been been designed decades ago uh before

the geospatial capabilities that we have

today uh were really implemented and so

geocoding is often incomplete or

inconsistent and makes reliable link

linkage across data sets very difficult

and second as we move to finer

geographic detail uh with our geospatial

capabilities today privacy and

confidentiality risks increase I think

we all can think of examples of that. Uh

and this requires stronger governance

and statistical disclosure protections

to maintain public trust which we then

need to balance with appropriate data

disclosure. We need that data.

Uh and third administrative boundaries

rarely align with real world social,

economic or environmental processes. the

our administrative boundaries are units

for dissemination

uh are are often

take

you know long time to adjust once

they're implemented they tend to remain

stable and we know that social economic

and environmental processes are anything

but stable they're constantly changing

and this can distort the analysis and

complicate uh historical comparisons his

comparisons over time. And so together,

these challenges show that disagregation

is not simply about producing smaller

geographic units. It requires building

integrated coordinated national data

systems. And that's where what brings us

to uh our contribution uh this expert

group's contribution to the discussion.

The GSGF 2.0 you know and the global

statistical geospatial framework

addresses these systemic challenges by

providing a common set of principles for

integrating statistical and geospatial

information. Uh the GSGF does not

replace national systems. It helps them

work together ensuring consistent

geocoding interoperability

and protection of privacy and

confidentiality.

And by establishing location as a shared

integration key, the GSGF allows data

from surveys, administrative sources,

and earth observation to be linked

reliably. The five principles of the

GSGF create that structural foundation

needed to produce highquality

location-based statistics at scales that

support real decision-making. Um, this

is a great uh graphic from our

colleagues in ESQA. Um the GSGF enables

integration by providing consistent

approaches for assigning geo codes or

coordinates to official statistics and

administrative data. Uh allowing data

sets from different sectors such as

health, environment, economic statistics

to be linked through location.

Um this integrated foundation is what

ultimately supports advanced methods

like spatial disagregation we'll be

discussing today grid-based statistics

and small area estimation. Uh thanks

again Esqua for the graphic illustrating

how implementation of the GSG enables

the aggregation of geocoded unit level

data to a variety of relevant

geographies. Now despite all the

progress that we've discussed um and all

the different teams working many

countries still face capacity

constraints uh skills infrastructure and

institutional coordination uh can be

lacking

scaling pilot projects into sustainable

national production systems can be a

major hurdle. uh integrating

non-traditional data sources such as

earth observation can require new

quality frameworks, statistical

validation and new new uh technical

staff. So privacy and confidentiality

again still remain major considerations

when producing disagregated

location-based statistics. So addressing

these challenges requires collaboration

across agencies, international guidance

and shared implementation. Um so but we

are seeing a global shift toward hybrid

data ecosystems combining surveys

administrative data uh earth observation

and geospatial infrastructure. Um many

countries are increasingly adopting

grid-based or location enabled

statistical dissemination models. Uh

there's a growing recognition that

integration is not simply technical but

that it requires governance uh

institutional coordination and shared

standards. And these changes are helping

transform fragmented data systems into

cohesive national evidence frameworks.

This evolution is moving geospatial

disagregation from experimental projects

toward mainstream statistical

production.

Now, these challenges and opportunities

are not theoretical. They're being

addressed in real time by the

organizations represented on this panel,

and they're going to share practical

lessons from implementing these

approaches in their national context.

and are looking forward to seeing these

presentations.

Um we are going to start uh our panel

discussion with Miss Mary Smith of

statistics Ireland uh co-chair of the IA

SDGs of WGI the working group on

geospatial information. uh and the

question we have posed to her this

morning is how does geospatially enabled

data disagregation strengthen SG

indicator production and how does this

work support the priorities of the IAG

SDGs. Uh I believe Mark is going to take

the screen from me

and Mary will begin her presentation in

just a moment.

Thank you, Josh. Mark, are

>> Thank you, Mary.

>> Are you able to share the slides?

>> Should just be coming through now.

>> Okay. Thank you.

>> Thank you, Josh and Clauddio. And um as

Josh said um I represent the working

group on geospatial information. I'm the

co-chair along with Sandra Moreno of

Colombia and Mark is UN GGGIM is the

secretariat.

This presentation focuses on the paper

the WGI published last year on rescuing

the SDGs with geospatial information.

Next slide please Mark.

So the working group on geospatial

information or WGI for short is a

working group of the UN inter agency and

expert group on SG indicators. So that's

the IAG for short and we're a working

group of the IAG and we we are focused

on bringing geography to the SG

indicators.

Some of the uh key resources we have is

the SG the geospatial road map which

acts as a guide for statistical offices

in implementing geospatial information

and the reporting process of the STG

indicators.

The short list on the SG indicators

looks at the indicators through a

geographical lens and we have a short

list A and B. It focuses on the

indicators where geospatial information

can provide the indicator itself. And

examples of these are environmental

indicators for example the extent of

ecosystems and so on. And um as Josh me

mentioned some these natural um

phenomenon aren't based on um grid based

systems. So we have to account for

geographical and location of particular

indicators. And short list B is

indicators where the geospatial data can

provide data disagregation and adhere to

the principle of the STGS in leave no

one behind. We have a number of reports

such as the global and complimentary

geospatial data for STGS and the land

cover reports which support the work

we're doing and this focuses on rescuing

the STGS with geospatial information.

the paper we published last year and we

look at data disagregation.

>> Next please.

>> So the rescuing paper looks at how

geospatial information can transform the

production, measurement, monitoring and

dissemination of the SG indicators.

Big ideas from the paper include um

the 2030 agenda obviously and how we can

ensure it benefits from geospatial

information

and there's a short summary for policy

makers as well and we identify key

stakeholders and ideas and empowering

the stakeholders to overcome challenges

and embrace opportunities that

geospatial information offers.

We look at the role of new data and

technologies and globally available data

sets for enabling local and national SDG

monitoring. And as um mentioned earlier,

a lot of these technologies and data are

only recently available and were not

available at the time the STG indicator

framework was developed.

Outcomes of the rescuing paper include

identifying issues such as under

reporting of the indicators and the

importance of geographic disagregation

within the statistics.

We looked at the metadata as part of the

IAGS 2025 comprehensive review of the

indicator framework last year and we

proposed quick wins in also

methodological advancements to improve

the indicator reporting process.

We provide enhanced guidance on how

geography impacts the indicators and

this looks at resolution at country and

global level. Next slide please.

So just very briefly I won't read all of

this but we understand that geospatial

information and earth observation can

provide new and consistent data sources

which enhance the locationbased

variables and support and inform

official statistics and indicators. Next

slide please.

So looking at the what and how. So um

obviously the geospatial disagregation

to enhance the reporting process and how

we are doing this is through

implementing frameworks guided by the

SDG geospatial road map which supports

statistical and geospatial actors within

the glo global indicator framework. We

promote collaboration and focus on

increasing collaboration at all levels

agencies ministries and uh peers in

other countries.

And we're focusing again on the um

incorporating the geospatial dimension

into the metadata and looking at the

indicators through a geographical lens.

We prioritize country-led and

country-owned data, but it's also

important to recognize that there are

data gaps and these data gaps can be

filled with globally available data and

these can highlight where and how

countries can invest in the future. The

work we're doing is supported by simple

and impactful storytelling through our

story maps and um the data visualization

of these this work is quite important.

Next slide.

So here's a link and a QR code to the

story maps. You can explore it. Next,

please.

and some live links here to the examples

at country level and um the particular

goals that it focuses on. So I invite

you to explore this data. If there's

anything that's relevant to your country

or organization, please get in touch

with us. Next slide.

And just um as Josh mentioned the

integration of statistics and geospatial

information expert group we are part of

this group and we're on the steering

group and we work closely with the ISGI

Josh's team on the task team on

disagregation of statistics by

geographic location. So we're working

with him on this and supporting this

work. the 2030 vision again we're

working with them on that and the global

data where there's methodology and

scientifically approved methodology for

supporting the indicators and looking

beyond 2030 we are focused on a future

oriented approach which will require

developing the geospatial reporting

indicators and we focus on

institutionalizing the capacity for

local to global Global integration and

ensuring geospatial inter information

will underpin all future development

frameworks both national and

international. So this will be a

foundation to shape sustainable

development beyond 2030. Next slide.

Thank you very much. And this is just a

very quick visual

um impact of the work we're doing

accessing public transport in Colombia

goal 11 and nine and then the story map

looking at annual temperature anomalies

with the d the world meteorological

organization. Our details are here so

get in touch with us informally if you

think there's anything we can support.

Thank you very much and thanks Mark.

Back to you Josh.

>> That is great Mary. Thank you very much.

Uh these are great story maps. I'm sure

everybody looks forward to digging into

those if they haven't already. Um the

next person we have uh next presenter

that I have on our list is uh Mr. Soel

Rastan and I believe Soil you're going

to share your slides with us.

>> Yes, I going to start sharing my slide.

>> Okay, thank you.

>> Let's do it.

I can see in the map that uh the center

of the

>> African Sahara is getting cooler.

Anyway, so it's an interesting one.

Let's go. So, you see my uh my screen

and uh this is Do you see my

presentation? Just quick.

>> Perfect.

>> So, I'm going to make it

>> a little bit bigger presentation. So,

>> perfect.

>> Um

today, I'm gonna basically um express

something. I'm I'm from uh Esqa in

Lebanon. Uh we are one of the fifth five

uh economic and social commissions. Um

this is linkable. I'm going to go into

our u uh portal and we'll show you some

examples. And I have about 20 slides.

The first one is important. This one I

going to spend something around 3 four

minutes in it. The rest are going to go

very fast and the last one is the most

important. So as long as I give you the

pictures is exactly what we're doing

here. So this is available for

everybody. You can basically go in and

you see the Arab development portal and

we have a small section in it which is

the uh geo stories in it and in the geo

stories you can see that uh we publish

stories and one or two stories once

every three to four months. Um for

example um I can go through I don't know

I mapping mapping damage okay in the

crisis what happened in Lebanon for

example I'm going to go through it very

quickly and uh

you see that we try to see the damages

happen in the southern part of Lebanon

and then we going to bring data from

multiple sources and we're going to see

the extent of the damage and see

buildings on top of it and try to

basically map it all together

and look exactly where are the damages

and validate it with the ground

truththing data and bring in satellite

images and look at them as well and

classify them and then basically present

in terms of the policy framework in

terms of the cost of the damage that

happens but this is not what I would

like to talk about today okay so let me

go back and I will show you something

else there u another thing for example I

would like to show you in terms okay

that's a good one actually uh mapping uh

schools in Lebanon and how far is for uh

uh children to walk to their schools

because the quality of education in

Lebanon we found is been declining. So

the question was is it because the

number of schools or is it because of

something else. So we mapped multiple

layers as as you can see and we

estimated the distances and at the end

of the day we categorized in terms of

the uh uh advantage disadvantaged people

and the conclusion was no 90% of the

children in Lebanon have access to

schools within 30 minute walk. So the

conclusion is we don't need to build

schools the problem is somewhere else.

Okay. So but again this is not the issue

I would like to talk about in today. So

uh one more thing I'd like to basically

uh maybe share with you which I like. Oh

here when we speak about the Arab world

we think that about the Arab world the

22 countries is about a desert with a

camel right. So this is the first thing

come into our mind. But then I asked my

team that no we need to basically

present the the 22 countries in a in a

in a different realm and then we did the

flora and fauna of the Arab region and

then we figure out it's a pretty rich um

uh rich uh uh ecosystems right and then

>> so so can I break in just for one moment

I'm sorry it it looks like there are a

number of people in the audience that

can't see your slides

and I just wanted to check in with the

audience and see if there's anyone one

that can see his slides. I'm seeing them

quite well. I don't know if it's

>> Josh. I'm wondering if the speakers can

see it but not the audience for some

reason.

>> Yeah, I'm getting a lot of thumbs up

saying they can see it. So, it's not

everyone.

>> Okay.

>> So, um

I hate to ask the audience to leave and

then come back in, but that might be the

resolution since it's not a systemic

issue. So, I'm going to go ahead and

with apologies to folks that are having

to people that are having problems

seeing the slides, I think we're going

to proceed. Um, and I encourage everyone

to maybe leave the the room and come

back in. I'm sorry for the interruption.

So, help why don't we go ahead and

proceed. Thank you.

>> No worries. And uh the slides will be

sent to anybody who would like to see

it. That's fine. Uh as long as they hear

my voice, that's okay. So anyway, so we

went and we brought uh whatever we can

in terms of how to basically represent

the ecosystems across the Arab world and

we showcase it that it is beyond the

beyond the the desert and u and the u uh

uh camel. U so again um it is a smaller

slow slower today I can understand this

one. Uh so again in terms of the power

of the data and what type of ecosystems

we do have per location and per

distance.

Then uh the fundamental things I would

like to talk about today today is about

the concept of economic landscapes on

the pulse. This is basically how can we

use um multi-layers geospecial data and

economic data I mean coming back from

statistics Canada which I retired from

there in terms of produce the pulsation

of a country. This is not a new concept

of course what we call the grided GDP.

So we took this is about two years ago.

We took Jordan as an example and then

everybody knows I mean we we took

population length. We can take

nightflight. We did not do it this time

but we took the economic sectors in

multiple dimensions and we try to

basically play with it to produce a

gritted GDP for Jordan and then we we

waited them. So the policy makers can go

there and start waiting them. Okay. And

you can see the pulse somewhere become

to be in the capital. But if you change

the weight then the pulses change. Okay.

So you can see there if you go to more

the agriculture sector then the

partition will change. Now this is what

I would like to basically present and

hence I going to go to the major

presentation which is the slide. Now let

me just get out of that

and this is the slide right you see the

slides. So

the idea here is the grided GDP and I

think this is the best tabular rasa

where GIS can basically serve

statistical commissions. Okay. So it's I

think it's a time for a new GDP

accounting and estimating framework. We

do not have one yet unless somebody

tells me that there is one as a UN

standard. We do not have similar to the

SNA similar to the system of

environmental economic accounting. So we

need a best practice guideline

and then adopt the standard. These are

the references I did a very quick is a

disclaimer the entire presentation the

one that you can see has basically

assisted but cloud anthroponic help me

and do that. So it needs some human

intelligence validation is required. So

please help everybody get that but I

asked uh cloud and I I spent something

about maybe three hours to generate

these slides and they I would like to

underline this for the last

one year the direction of the laboratory

has changed tremendously because of the

AI. So the AI is coming in terms of

helping us to generate beautiful things

if we know how to communicate with it.

So the first okay let's admit this

person here.

So usually the SNA there are some hints

into the uh grid GDP the CIA has

mentioned in it in applications and

extensions and our global geosper

framework it's a states that we can make

it happen.

Of course you either go with population

or you go with light or you combine.

Okay. But none of them have this system

of national account conceptual framework

in it. Right?

Then it has been expanded and some

people started to basically use shared

social economic pathways, population

weight. Some people start to bring in

vegetation indices and CO2 emissions to

generate the graded type of a GDP. Okay.

So why the gap assumed to exist

fundamentally I think is number four.

There's a tension between accounting

precision and a spatial estimate. I

think this is we need to go beyond that.

And where are the opportunities? Of

course there is no official standard

yet. So there's a room to pioneer

especially for data steps. So that

that's what I would like to basically

underline. So we need to pioneer right

now. So some official subnational GDPs.

I'm pretty surprised. I mean there are

83 countries produce some type of a

disagregated subnational

GDPs. Okay.

Brazil was an eye-catching 5,570.

I mean how they do it I have to sit down

with my Brazilian colleagues. Do they do

the system of national accounts? Do they

have supply and use table specialized?

Do they allocate the addresses in the

business registry? I don't or do they

just multiply by a weighted population?

We need to basically set with and the US

is another one and then we have the UK

of course you have Canada, Japan, South

Korea. All of this are producing some

type of a disagregated GDP which I'm

surprised that I see that uh there are

there okay of course the European Union

they use the new culture of territorial

units of statistics if they basically

allocate to it some type of a uh added

value in terms of the u uh possession in

the in the country themselves and that's

the

summary okay of the countries that

produce

of course some global disagregators. I

mean u we have the database of

subnational economic outputs. It's an

academic research. Of course we have

OECD. They're trying their their their

contribution as well. And then we have

another global great sa uh in terms of

the aggregated uh GDP from University of

Chicago and there are many more of this

and just some of them. Okay. And these

are the tiers from m municipality to

metropolitan area to sub regional system

and to the state and provinces.

The idea is I would like to basically

bring this into the table. We would like

to activate a united nation endorsed

grided GDP accounting and estimating

framework. This is what I'm calling in

this meeting. Okay. for many reasons for

where the impact is where it matters and

where it affects.

So it should be a standardized framework

within frameworks. So we can use the SNA

and the CIA and the UNGO special

frameworks. We can amalgamate them

together. Okay. And we can do it these

days thanks to the multi power AI that

we do have. And I proposed a road map

established up to this year. we can

establish secure funding and develop it

then standardize it and by 2030 we

implement it.

My final slide is from the London group.

I know many of you you know the London

group which is the group started the

system of environmental economic

accounting. I would like to propose the

Beirut group on graded GDP accounting

and estimation. Okay, thank you so much.

And that is my slice and it's

interesting the AI generated this for me

as Beirut and it is basically the Roshia

in our capital in Beirut. Very smart

eyes here and this is where my my

contacts are. Please feel free to

communicate and thank you so much. I can

stop sharing.

That was fantastic. So, I wish we had uh

much longer to go through that much more

slowly. I'm looking forward to seeing

how this develops and to participate in

that. I'm sure you would appreciate

anyone who is on this to contact and uh

get involved. That was great. Um

next up, next uh presenter we have is

Mr. Richard Tonkan of ESCAP. And

Richard, can you share some thoughts

with us on which institutional and

government arrangements have proven most

effective in enabling sustainable

geospatial data disagregation across

countries in the Asia-Pacific region or

whatever's on your mind on this topic.

We're always glad to hear from you,

Richard. So, thank you very much for

sharing.

>> Yeah. No, thank thank you very much and

train

Okay,

>> there it is.

>> That's great.

>> Yep.

>> And and then we lost it a moment.

>> We had it.

>> Is that sharing? Okay,

>> uh I'm not seeing it. It came up and

then it disappeared again. So maybe if

you could try sharing one more time. I

don't know if anybody else can see it.

Maybe a thumbs up. I think we're on the

same boat at this point.

Not seeing it.

Thank you all.

Sorry. It's okay.

Three.

There we go.

I'm seeing it. Richard,

there we go. All right, Richard, I can

see it. It looks like other people

giving thumbs up. so others can see it.

Take it away, Richard. Thank you very

much.

I can't see my

It's Sorry. Um,

okay. So, you got it now. Yep.

Hello.

>> Yeah, we lost it, Richard. We lost it,

Richard. We had it for a moment.

>> Yeah. I tell you what, would um maybe it

makes sense if you email it to Mark

quickly and we can have Denise

>> come on for a moment.

>> Yeah. And we'll circle back to you in

just a minute, Richard. No worries.

>> These things happen. So, um,

>> jeez. Denise, I wonder if you would be

able to share.

>> I will try sharing. For whatever reason,

my camera does not not want to join the

meeting today. So, but see if I can get

the slides to share. Okay.

>> Well, Denise is let me introduce. Denise

is from the place foundation. I'm sure

you're capable of introducing yourself.

I can see your slides. I see them.

>> That's a good start.

>> And there we go. Great. And the question

we posed to Denise was why is

geospatially enabled data disagregation

at re regional and local levels

essential for informed decision making

making and for truly leaving no one and

no place behind pun intended I assume

>> pun intended indeed and a really big

topic so I'm I'm going thank you for the

intro Josh and I will

>> I guess do this through a bit of a lens

of explaining what place is uh why it

came to be uh and what we do as an

organization. So hopefully that will

help illustrate uh the answer to those

questions.

So place is a nonprofit data institution

which might be a term that you haven't

heard before. Um it's a fairly novel

concept. It's something that is built

around a data trust, but it was founded

really on an understanding that these

days high quality mapping data is

critical for any country to be working

with whether you're working at a

national level or you're working down at

that local level and in particular place

really felt that some of the greatest

need in the world was in our rapidly

expanding urban spaces um particularly

for those which are undermapped or under

represented throughout the world. So

what we do as an organization is we form

a relationship or a partnership with

government agencies across the world. So

we do not work uh in a country that we

don't have a formalou to work with the

the national mapping agency or the

relevant agency. Um we work with that

country I guess not in a traditional uh

you know contractor type of way but in a

way that helps them build both capacity

and also get hold of the data. So, I've

heard people like Mary and Josh talk

already today about where we've still

got incredibly existing gaps in the data

that we need, whether that's about local

transport uh or energy or roads or

people. Uh what place does then is we

will actually work to come into the

country. We we'll do that either

ourselves or work with other companies

to collect high resolution imagery. Um I

think what's interesting and why it's

really relevant to be talking about this

these days is the technology shift

that's happened particularly in the last

5 years means that the way we can

collect uh information data and

knowledge has really really shifted and

imagery is a critical part now uh I've

heard people already reference earth

observation data which is another way

you can describe imagery here too but

it's a really rich and incredibly

important resource uh for most people

and most countries now to work with to

help understand their population and

their environment.

Um what we do then and this I think is

really important is uh in terms of

accessibility any data that place

collects is retained sovereignly and

owned by the country that we work with.

Um but really importantly getting it out

to people who may be statisticians, they

may be uh researchers, trying to unlock

that data and make sure that it can be

used by the people across the world that

need to use it is critically important

and that's where the place trust comes

into play here. So place trust uh is is

given a copy a licensed copy in

perpetuity to steward that copy for use

um by trusted users across the world as

well as ensuring that the countries

retain that sovereignly owned data as

well. So it's a novel structure. Um I

guess now you're wondering where are

those gaps we're trying to fill? Well,

we're obviously across Africa. Uh so

you'll see a number of those here and

I'll flip through a lot of slides here

quite quickly at this point. Um and

we're working across more land states.

And I'll just pause here because I think

to go back to uh some of the points here

around the the really important aspect

of the disagregated data at local level

when you're down at small island state

level often satellite imag

uh data collection methods don't work

particularly well because the very small

scales uh that you're working with. So

it's really critical to be able to get

down to as as low a disagregation as you

can for many of these places in order

for them to understand the nuances and

differences about their population and

their environments too. So this data

approach I guess is absolutely critical

when you're down into small spaces.

Um so small island states and also

recently in Bangladesh has been where

place has been as well.

Um I will say that we only collect uh

high resolution or positionally accurate

imagery. So I think it was Josh that's

mentioned something earlier about how um

geocoding has not always been super

accurate in the way data has been

collected. So imagery and the way we

collect it uh is doing that positional

accuracy is absolutely critical. We

collect two types of data. So we look at

high resolution aerial photography and I

think this is one of those things that's

really important is we also collect what

we call place ground or street view

imagery uh which gives you a very very

different and rich data source to help

understand the landscape and the people

and the population that you might be

working in.

So I'm going to jump through images

really rapidly now at this point. What

else can you do with that imagery? This

allows you to very rapidly create things

like digital surface models or digital

elevation models to understand your

landscape. It allows you to understand

topographic products. Um I haven't got

time but I I would play this word for

you uh if I could. I don't I don't think

it's going to work right now but 3D

modeling obviously gives you a very very

rich understanding and very quick

understanding of an urban landscape.

And I think this is what I mentioned

earlier with the goai and machine

learning tools that we have now. Imagery

gives you a very very very rapid way of

getting statistical information that

used to take months for you to be able

to understand. So whether that's

understanding where the buildings are,

how big they are, um being able to

rapidly use the imagery to create land

use classification very quickly, uh

looking at population census. So one of

the big partners of place has been world

pop out of University of Southampton who

have been refining further how they use

the place imagery to do that population

census which had traditionally been

satellite imagery but now with both

aerial and also that ground level street

feature. What that allows you to start

doing is to really start collecting all

of that feature information that you

might want to understand about an urban

landscape and importantly where that is.

So buildings become really important

here. How many windows are inside that

building? What condition is the building

in? All starts to be able to give you

rich information to start estimating the

condition, the living conditions of the

population that might live there.

Um the other one that it critically does

is we recognize text. So this allows

countries too to rapidly build data sets

that they may traditionally have had

taken them months if not years to be

able to build if they can build them at

all. Um but the imagery rapidly gives

them that that capacity to do so.

So I guess that's a really very quick

quick kind of jump through of what place

does and I'll finish by saying um going

back to the the why is this data really

important from a localized perspective.

We talked about municipal councils and

local city councils which place will

also work quite a lot with those are

often incredibly

resource poor entities in trying to

implement change on the ground. So I

think looking for novel ways to get them

the data they need um in a format that

they can also understand and use to be

able to help them underpin their

decision- making is absolutely critical

because they're asking questions like

this. They're asking how green is my

city so I know where I can plant more

trees. They're asking if I make a road

change or I'm going to build more

buildings, who is going to be impacted

by that infrastructure and how can I

make sure that it's a positive impact,

not a negative one. They're looking

increasingly because of climate change

at who's at risk of floods. Uh how do I

save them? How do I move them in my

city? How do I build protection for

them? Um, and they're looking

importantly as as our world heats, how

resilient is things like our housing,

which are the buildings that are most at

risk, which are the populations that are

most at risk. Um, so I think it's really

important to look at how we localize

this data so that those councils and

those local decision makers have got the

tools that they need to be able to

answer some of these questions and

protect their both their population and

their environments. And with that, I

will finish and give the screen back.

That was that was really great, Denise.

And I really love how you wrapped that

up with the environment, society,

housing, economics, everything that

we've been talking about. Uh great

presentation. Thank you very much for

that contribution. Um I think we want to

continue and move on and I will just

announce to everyone at this point our

intent is to finish with all the

speakers and still hold a bit of a

discussion. Uh it looks like uh again

these have been fantastic presentations

and we're closing up on the top of the

hour. I'm hoping the speakers will be

able to stay on with us a bit even if we

extend past the hour. Uh we understand

if you need to drop off but uh I know we

the organizers will be here. Uh we

invite everybody to stay along with us

as we we wrap up this session. So, we

may run a bit over 10:00 a.m. and I

encourage everybody to stay with us. Um,

looks like our present next presenter is

here, Richard Tonkan again to talk about

institutional governance arrangements uh

in the Asia-Pacific region. Thank you,

Richard. Take it away.

>> Thanks. Thanks so much, Josh. And yeah,

apologies for the the slide sharing

issues. Um so yeah know I just wanted to

kind of answer this question kind of

around the institutional and governance

arrangements really by drawing on some

of ESCAP's experiences working with

countries kind of in the Asia- Pacific

region kind of on uh developing

disagregated uh uh statistics.

So yeah and I think that the short

answer to answer the question you said

really is sustainable geospatial

disagregation. It's

not about individual tools or data sets

kind of and it it really is about how

the statistical and geospatial

institutions work together in practice.

Uh so what I want to do is just

highlight a number of things that we've

consistently seen which really make a

big difference using a few examples uh

from from countries to illustrate them.

Uh so of course

Denise and others have already stress

why why this matters. Uh so we know

national averages hide important

regional and local disparities

for decision makers. It's not enough to

know just how much but we need to know

where and who. Uh so soon as we turn our

focus towards geospatially enabled

statistics then these institutional

questions become unavoidable. Who holds

the data? Who maintains the reference

layers? how different systems connect

and I think certainly what what I've

found what my colleagues in ESCAP have

found many countries struggle not

because they don't have the data but

because systems have evolved separately

uh so we see some of these issues that

have been highlighted repeatedly kind of

inconsistent geocoding incomplete

metadata

legal institutional frameworks that

weren't designed for routine

cross agency data sharing

uh infrastructure that isn't optimized

for integrated workflows and of course

skills gaps as well. Um I think really

importantly none of these are purely

technical problems. I think they're all

really symptoms of institutional

arrangements that haven't yet adapted to

the demands for disagregated data.

Uh so to me this is where the the GSGF

is particularly useful and very it's

great to see the the updates to the GSGF

being uh launched kind of the the

statistical commission.

Uh so in my experience one of the key

values of the GST is it provides a

shared reference point for NSOs and uh

uh national geospatial information

agencies both in terms of assessing

geostical capacity and but identifying

priority actions to uh to address those

capacity gaps.

Um I can I I mention it every time I I

speak with you Josh but one one

initiative that's been uh particularly

influential for me kind of is this high

level seminar kind of on the integration

of statistical and geospatial

information uh which we organized with

UND and as Norway kind of in Bangkok uh

the other year. At that seminar we had

large number of countries from across

the world using the GSGF self assessment

to have very practical conversations

about roles, responsibilities,

priorities uh not in the abstract but in

relation to real national needs. Um that

was really instrumental uh for certainly

for us at ESCAP but I think also for

other partners to help understand where

follow-up support supports can be most

effective.

Um

so I just want to pick up briefly on a

couple of country experiences uh on work

that's come out uh following uh that

that GSGF uh self assessment. So in

Kyrgyzstan for example led to the

development of a joint at national

roadmap for geosis integration uh

clarifying roles between the national

statistical office and the land agency

and that was also accompanied by the

development of a national geostical data

inventory and associated metadata

uh which again turned out to be a really

critical enabler for producing uh

priority of national STG indicators.

uh in Indonesia um again we've seen uh

close collaboration between the Cisco

office research agencies and sector

ministries

uh uh combined with investment in shared

data infrastructure which has really

enabled

uh the production of rice estimate rice

production estimates integrating survey

and earth observation uh data which has

allowed the production of estimates that

are not only more geographically

granular but also more timely and more

cost effective.

Um

you and others already spoken about the

importance of small area estimation and

I just like to take this opportunity to

uh briefly highlight some of the work

that we've been doing in that area. Uh

so uh together with

uh uh with various partners uh we've uh

recently launched new practical how-to

guide on geospatial small area

estimation uh building on the SAPE

primer that was developed by the inter

intersectaria working group on household

surveys.

Uh this guide is designed to support

practitioners in using open-source tools

uh to combine geospatial data from

multiple sources to produce disagregated

estimates uh whether it's a poverty or

other priority indicators. And the

reason I mention it here uh not just

because I I I think it's an incredibly

helpful uh tool for uh for statistics

producers

uh but I found that when we've been

working with countries kind of on this

uh these very practical applications

found a very effective way for surfacing

uh some of these key issues kind of

around institutional responsibilities

around data sharing ing and uh around

around governance more broadly.

Uh so coming coming back uh at to the

question kind of addressing it directly

based on these experiences from across

the region

this want to highlight these three

institutional factors which have

consistently proven to be most important

in uh supporting sustainable geospatial

uh data disagregation. So first it's

effective collaboration uh based on

clear role definitions between national

core offices and national geospatial

information agencies.

Uh so for example in in the example I

shared in Kystan. This was formalized

through a joint road map aligned to the

GSGF which uh really clarified

responsibilities

and made collaboration routine rather

than NAD hawk. Uh second

um

uh establishing maintaining

strengthening uh national geospatial

data or national geostical data

inventory and associated metadata bring

in absolutely every single country you

kind of I've worked with in Asia Pacific

region and beyond as well kind of yeah

kind of weaknesses and limitations here

have been a really key factor. in

holding back progress on data

disagregation. So yeah, I think that's

absolutely key for me. And and I think

third is anchoring GSGF implementation

in concrete priority use cases uh

whether that's price estimation or small

area estimates of poverty or other

indicators.

uh really found that this is what makes

uh geopolitical integration real for

institutions. It connects uh all of

these issues which coming from an NSO

perspective could be quite theoretical

kind of geocoding reference layers data

sharing it connects them to decisions

which

absolutely matter and I think can be a

real enabler for progress.

Uh so so I think I'll I'll stop there.

Uh hand back to you uh Josh happy to

answer any questions anyone has in the

chat or after all the presentations.

Thank you.

>> Thank you very much Richard. I I really

appreciate those uh examples that you

gave of the GSGF being implemented in

the real world and having direct impacts

on the global community. I know people

uh who've been around this for a while

like I see uh Martin Brady in the

audience uh should feel really heartened

uh and pleased with the work that

they've done presenting uh this work to

the the global community and and uh in

our contributions. But without further

ado, I'm going to pivot to Mikuel Miiori

um from the European Commission's Joint

Research Center and his question is

where does Earth observation add the

most value for moving beyond national

aggregates. Uh Mcklly, the floor is

yours. Thank you, sir.

>> Yeah, thank you Josh and thanks

everyone.

>> And I can see your slides. I'm sorry,

you're you're good.

>> Great. Thanks Josh and everyone for for

the opportunity. So I'll try to provide

a very short answer that is through

disagregation capacity building and that

integration. This answers is of course

uh incomplete but in four slides I'll

try to transition you from your products

to harmonize the statistics. So in the

next slide we see um how national

average actually hide the special

patterns but also how special data can

uncover those patterns. This links back

to what Cladio showed at the beginning.

So this aggregation is key to map and

quantify subnational and place a

specific phenomena. A lot is known in

the domain of of statistics about the

downsides of averages. But when we

introduce geospatial data or data refer

to locations, these data are also

subject to limitation and biases

unfortunately coming from the special

data properties. Here we can see an um

an example of an earth observation

derived data to map uh human settlements

and their characteristics.

Uh this is a key coariant or ancillary

data as we will see to disagregate data.

In this image, we see uh building

typologies in in grid format. And let's

think for a moment about census planning

and reporting the average building

typology at the district level to plan

for for the census versus uh mapping the

distribution of building typologies in a

10 m grid cells like it's in this image.

Of course, um if a district has

uncertain average typology, we can have

still a lot of variation having very

tall buildings and residential buildings

where a lot of people should be

interviewed next to uh probably an

industrial facility that does not

require the same effort. So earth

observation allow uh allow for a

cost-ffective data acquisition at scale.

This is one of the properties of of data

coming from satellites but also over a

long period of time. So the acquisition

are routine over time. Some satellites

are acquisition time other down to 5

days but still very frequent and this is

really key for the concept of

sustainability and scalability of

operations.

And also with the maturity of the earth

observation sectors both private and

public players are producing an

increasing amount of data with

increasing resolution and theatic

accuracy and also commercial commercial

services are becoming more accessible.

In the next slide we see an example of

uh a product of the copernicus product

that has a space segment. So acquisition

of of satellite data but also a variety

of thematic services in this case a

product from the emergency service that

is actually a population grid. A lot is

known by now about population grids.

When we first started those activities

back in the 2000s, it was not such a

common practice. But of course, um it's

very easy to understand the difference

with reporting the average of a density

of a district versus the density of

people at 1 kilometer grid cell. So in

the next slide we can see that the full