[HOT] Interim Report: Typhoon Haiyan (Yolanda) Damage Assessment

Banick, Robert Robert.Banick at redcross.org
Thu Feb 13 17:35:13 UTC 2014


@ Pierre and Mikel

I concur with everything Dale says and want to add that if these comments can be packaged in a concise form — bullets or short paragraphs — they'll be easier to absorb into the final report. There's been some great feedback off this list as well as on it and I want to make sure that input goes in here.

Pierre, I appreciate your concern about the lack of attention given to HOT and the OSM community as a whole in this report. Obviously we are on this listserv and are HOT Members, so we care quite a lot about these things. However, this report was more about results than agency.

If you would like to take the lead on writing a short description of HOT and its role (1 sentence, maybe 2) that can be worked into the introductory remarks and submitting it as a pull request, that can be accommodated right away. I too am eager to see that HOT receives the credit it deserves here. I think there's also scope to mention HOT as a likely partner for some of the process improvements described in the recommendations section.

@Mark

It seems there's quite a lot of chatter about this lately. I think it would be good to have a broader discussion on this proposed app and start documenting requirements, perhaps in the next Technical Working Group meeting? Having some concrete working documents would help us cohere as a community and put forward proposals for funding.




Robert Banick | Field GIS Coordinator | International Services | Ì American Red Cross<http://www.redcross.org/>
2025 E Street NW, Washington, DC 20006
Tel 202-303-5017 | Cell 202-805-3679 | Skype robert.banick

From: "Dale.kunce at gmail.com<mailto:Dale.kunce at gmail.com>" <Dale.kunce at gmail.com<mailto:Dale.kunce at gmail.com>>
Date: Thursday, February 13, 2014 9:14 AM
To: "mikel_maron at yahoo.com<mailto:mikel_maron at yahoo.com>" <mikel_maron at yahoo.com<mailto:mikel_maron at yahoo.com>>
Cc: Pierre Béland <pierzenh at yahoo.fr<mailto:pierzenh at yahoo.fr>>, Robert Banick <robert.banick at redcross.org<mailto:robert.banick at redcross.org>>, "hot at openstreetmap.org<mailto:hot at openstreetmap.org>" <hot at openstreetmap.org<mailto:hot at openstreetmap.org>>, Clay impact <clay.westrope at impact-initiatives.org<mailto:clay.westrope at impact-initiatives.org>>, "Kunce, Dale" <dale.kunce at redcross.org<mailto:dale.kunce at redcross.org>>
Subject: Re: [HOT] Interim Report: Typhoon Haiyan (Yolanda) Damage Assessment


@Pierre and Mikel,

Your comments are well received.

I think we do need to write a little bit on the exact role and nature of how and why the OSM community did during the Haiyan activation.

As you all know we are extremely proud and motivated supporters of OSM and HOT. We want to see OSM continue to be a huge resource for the humanitarian community. As Mikel points out we do think that OSM could be a great resource with just a few relatively small investments.

Another thing to keep in mind is that this is the interim report, the full report will come out later that will take an even deeper look at the data. I encourage HOT folks that were involved in the activation to suggest changes on github to clarify things. Please do keep in mind though that this is a Red Cross/REACH/USAID report and so not all suggested changes may be incorporated.

Thanks again to the continued support of the wonderful volunteers and HOT members who were amazing and helped out in a huge way.


@Mark

We (Red Cross) also identified the need for an app that would make both data collection for assessment and editing OSM from the field. This is suggested in the report and is one of the investments we are looking at long term.

Dale


On Thu, Feb 13, 2014 at 2:54 PM, Mikel Maron <mikel_maron at yahoo.com<mailto:mikel_maron at yahoo.com>> wrote:
Hi

> We should be careful to make the appropriate analysis and not demotivate the OpenStreetMap community who made
> such a huge effort for this activation.

This report, I believe, is the first thorough investigation ever of OSM data from a disaster activation. That gives us really concrete analysis to digest and learn from. Certainly not everything in a response is within HOT's control, but we need to accept that we can and must improve what we do, and take feedback seriously. That is precisely how HOT has grown from an outrageous idea back in 2005, to an indispensable component of disaster response in 2014.

Our mission is to be effective in disaster response, and that means acknowledging our gaps, even when that might be hard. We made a huge effort, and have gotten tremendous kudos for what we've accomplished for the Haiyan response, and we can be secure in our pride in that. We are also comfortable, strong and motivated to adapt ourselves to well researched critique and analysis.

And i n any case, worth noting that this is the quite positive take away...

"The assessment found that modest investments in technology, business processes and pre-disaster activities could make OSM a strong platform for damage assessment data and analysis in future disasters."

> The Red Cross study makes the assumption that the objective of the OpenStreetMap crowdsourcing was to make detailed
> assesments.
...
> The Red Cross study points to the fact that in most cases the limitation in damage assesment was the
> imagery that seemed to show undamaged buildings when in reality they had sustained damage.

I'm not sure if this assumption is made or not, but I agree that the report is not totally clear on the context and objective in which HOT was engaged in damage assessment. A single sentence or two on that context in the executive summary could help. As well, inclusion of the points on imagery in the executive summary would be more representative of the conclusions.

Btw, the report is in GitHub :) Not sure if ARC wants to respond to pull requests, but very possible to make fine tuned suggestions to there https://github.com/AmericanRedCross/OSM-Assessment

Cheers
-Mikel

* Mikel Maron * +14152835207<tel:%2B14152835207> @mikel s:mikelmaron


On Wednesday, February 12, 2014 9:25 PM, Pierre Béland <pierzenh at yahoo.fr<mailto:pierzenh at yahoo.fr>> wrote:
Hi Robert,

In the early days after Typhoon Haiyan, there were  estimations of 10,000 dead, severe damages and no communications with various remote areas. The humanitarian community as a whole needed even  rough estimates of the extent and the distribution of the damages.

The Coordination group where OCHA , US Red Cross and HOT participated right from the beginning of this intervention thought that OpenStreetMap should contribute to make Damage assesments. The objective was to make these assesments from Satellite imagery as soon as these would be available.

The Red Cross study makes the assumption that the objective of the OpenStreetMap crowdsourcing was to make detailed assesments. But you can only make rough assesments from Satellite imagery especially when atmospheric conditions restrict the quality of the images provided. And a simple classification was used (ie. damaged or destroyed).  UAV's (drones) or Aerial oblique imageries could have been used to make detailed assesments. But this was not part of the established workflow of the humanitarian community before Haiyan and such images were not available to make detailed assesments.

Once such crisis are ended, we should surely analyze our actions and plan collectively for better interventions in the future. But we should avoid to have wrong conclusions about actions taken during this crisis.

The way the report is written, it gives the impression that imprecision in evaluation of assesment is due to the use of the OpenStreetMap community.  The humanitarian community as a whole did not build before this event the capacity to react rapidly, deploy teams and provide detailed post-disaster imagery in other ways then through Satellite.

In the context of this emergency and with the imagery provided, would professionnals specialized in damage assesment have scored significantly better? Due to the limits of such assesments in the operational context of this operation, analysis should be based on the capacity to identify zones of high damages and not focus on individual houses. To my point of view, the objective of that operation following the severed damages after Typhoon Haiyan was to give an early warning to identify zones and not individual houses. This would need oblique imagery.

Thinking about a better workflow in the context of such disasters,  the capacity to have more flexibility and deploy rapidly teams when necessary to obtain either UAV imagery (drone) or aerial oblique imagery would surely give a different response, this either with the OpenStreetMap community or professionnals of damage assesment.

We surely have a workflow to build and establish the role and limits of assesments done with aerial imagery in the context of such emergency operations.

We should be careful to make the appropriate analysis and not demotivate the OpenStreetMap community who made such a huge effort for this activation.

The Red Cross study points to the fact that in most cases the limitation in damage assesment was the imagery that seemed to show undamaged buildings when in reality they had sustained damage. But this is not reflected in the Executive summary and in  the Conclusion of the study. This study should be completed with a better analysis of the type of imagery necessary to make better asssesment studies.


Pierre

________________________________
De : "Banick, Robert" <Robert.Banick at redcross.org<mailto:Robert.Banick at redcross.org>>
À : "hot at openstreetmap.org<mailto:hot at openstreetmap.org>" <hot at openstreetmap.org<mailto:hot at openstreetmap.org>>
Cc : Clay impact <clay.westrope at impact-initiatives.org<mailto:clay.westrope at impact-initiatives.org>>; "Kunce, Dale" <dale.kunce at redcross.org<mailto:dale.kunce at redcross.org>>
Envoyé le : Mercredi 12 février 2014 9h22
Objet : [HOT] Interim Report: Typhoon Haiyan (Yolanda) Damage Assessment

Dear HOT Communuity,

The American Red Cross and the REACH Initiative are pleased to present an interim assessment report on the validity of the building damages assessed through OpenStreetMap in the weeks following Typhoon Haiyan. You can find a print copy attached and a more interactive website version at the above link.

The results were unfortunately negative and underline real limitations in OpenStreetMap’s ability to capture these results in the present. Neverthless, this report identifies strong promise in the OSM model of crowdsourcing and highlights the investments needed to make that potential possible.  It’s our sincere hope that funders, NGO partners and most especially the OpenStreetMap community will rally around these investments so that OSM can play an even stronger and more operationally useful role in future disaster responses.

We are indebted to the US Agency for International Development’s Office of Foreign Disaster Assistance (OFDA) for funding this assessment and look forward to future partnerships to improve the utility of open data and OpenStreetMap in particular for disaster response.

With all the best,
Robert Banick, Dale Kunce and Clay Westrope
American Red Cross & REACH Initiative

Robert Banick | Field GIS Coordinator | International Services | Ì American Red Cross<http://www.redcross.org/>
2025 E Street NW, Washington, DC 20006\

_______________________________________________
HOT mailing list
HOT at openstreetmap.org<mailto:HOT at openstreetmap.org>
https://lists.openstreetmap.org/listinfo/hot



_______________________________________________
HOT mailing list
HOT at openstreetmap.org<mailto:HOT at openstreetmap.org>
https://lists.openstreetmap.org/listinfo/hot



_______________________________________________
HOT mailing list
HOT at openstreetmap.org<mailto:HOT at openstreetmap.org>
https://lists.openstreetmap.org/listinfo/hot




--
sent from my mobile device

Dale Kunce
http://normalhabit.com

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/hot/attachments/20140213/58de30d0/attachment-0001.html>


More information about the HOT mailing list