Aloha, good afternoon, everyone.

It is one oh two PM on Tuesday, October 20th, 2026.

I call the Committee on Energy Environment and Sustainability Meeting to order.

I'd

like to welcome Vice Chairs Los Santo Stam, Council Members Toba and Tupola.

Uh, pursuant to section ninety-two-three point seven, white revised statutes.

This meeting will be conducted as a remote meeting by interactive conference technology with the following procedures in effect for the meeting.

Members of the public will be allowed to provide oral testimony on all items on the agenda in person in the council chamber and remotely when each agenda item is taken up.

As both English and Hawaiian are official languages of the state of Hawaii, pursuant to Article 15 section four of the Hawaii State Constitution and section one-thirteen of the Hawaii revised statutes.

Members of the public may testify in either language.

For oral testimonies offered in Olilo Hawaii, additional time as may be necessary will be allowed for the testifier to provide an English translation of their testimony if they so wish.

Before testifying, each person shall state their name.

For persons who are testifying by telephone only, I will identify you by the last three digits of your phone number.

When your number is called, please listen for the prompt to unmute and press star six.

When I call for anyone else standing by remotely, please utilize the zoom raise hand button to indicate your desire to speak.

For those joining us by telephone only, please press star nine to indicate your desire to speak.

Some friendly reminders and tips.

Therefore, I'll call for objections.

And if there are any, I will call for a roll call vote.

Which I believe today we just have info.

Briefing, written testimonies, including the testifiers' address, email address, and phone number will be available to the public as described on the posted agenda.

As a courtesy, please turn off all cell phones for the duration of this meeting.

And so that brings us members to agenda item number one.

This is an informational briefing and update on flood mitigation services by the department of emergency management and members joining us in the council chamber.

We have Dr.

Randall A.

Collins, Director of Department of Emergency Management, to give a brief presentation, which is online is Departmental Communication 654 2026.

Director, I'll go ahead and turn uh the time over to you as you're pulling up your presentation online.

And I believe it has been distributed to members as well.

Well, good afternoon.

Thank you, Chair Wire, and appreciate the opportunity to give this uh informational update briefing.

I know that uh this council has been very supportive of this project insofar as there was a resolution last year to develop an uh AI uh enhanced early warning detection system for all hazards, and then on top of that, during the budget process, allocated $300,000 to us for uh being able to kind of kick that off with a uh special project uh focused on the flood, largely based on our lessons learned from CONALO.

And so we're really excited to be able to tap into that funding as well as the other funding allocated uh that DEM has allocated through the Urban Area Security Initiative project as well.

So with that, I'll get straight into the uh presentation.

First and foremost, as I mentioned, this is going to be an all-hazards uh early warning detection system infused by AI, which is will allow us to uh really enhance the way that we read the data that uh we can uh receive back from any new instrumentation that we get out there.

So our our big project is really uh going to be about getting with the other uh stakeholders who already have flood monitoring systems and other hazard monitoring systems out uh throughout Oahu, leveraging those, understanding where the gaps are, and then uh start determining which types of monitors and gauges we may need to add to that particular system, how we bring all that data back to us into a system that can do the data analytics that we need and do the machine learning that we need uh to ultimately give us the uh decisions that uh and information that we need to make decisions from the emergency operations center.

Some of our current gaps is that we work in multiple systems.

We have we don't have data analytics, uh we don't really have any historical uh comparisons or comprehension of what these types of or what similar uh forecast patterns really resulted in, and then we also uh don't know the corollaries, the amount of rain that comes down versus the uh amount of uh uh flooding that comes up, uh whether that be in the reservoirs or in the streams or even in just standing water flood zone areas.

And then we also have limited instrumentation, as I mentioned before, we want to add

to that infrastructure.

So, with that, our focus is really going to be on instrumentation, and uh starting with a series of seminars and workshops, we'll bring together all of our uh stakeholders.

Uh and first and foremost, I'd like to really call out uh true, uh, the true initiative, which is tech readiness user evaluation and nonprofit here in Oahu who helped DEM really lay the groundwork and the foundational analysis of our AI uh policies and procedures and really studied what our needs were within DEM, and that really set the stage for everything else that we're going to do with AI, both with this particular program and other uh modernization uh initiatives.

But with that, we definitely recognized um that we need to bring together all of our stakeholders.

So true will be involved in that.

The National Weather Service, of course, are our greatest partners here.

USGS, who owns a lot of the uh infrastructure now, UH Manoa, uh an AI company called EM1, as well as Hohonu, who's done presentations here before as well, and then um the Pacific Disaster Center, H.

HDOT has a lot of flood monitors out there, and basically anybody that has a camera system that we are able to tap into any of the uh government agencies that has a camera system that we're able to tap into, those are the types of infrastructure that we want to tap into.

So we'll bring them all together and we'll have a discussion about what each one of those entities can do or what their perspective is in terms of what our needs are, so that we can then move into securing that instrumentation and then moving into the detection phase.

So we'll get that instrumentation out to the hot spots that we identify with our stakeholders and then move into validation and really start seeing the data come in and how we need to manage that to help us share the orientation of what's happening out in the field, and then that results obviously in action in so far as that we're able to see that data helps us make decisions within the EOC, helps us to communicate that out in public messaging to the public.

So again, our process is to organize, get our stakeholders aligned and one cohesive strategy that we can buy off on this particular strategy and this particular mechanism in terms of what infrastructure we buy, what computer systems we buy, and how it powers that information and how we receive that information into the emergency operations center.

The next piece is that design and build, get the information or get the flood monitors and the weather gauges and other instrumentation out there, as well as build the nucleus of computer information, computer systems that we're going to need to receive that data and do the data analytics, and then we'll move into testing and piloting.

We will have a pilot phase, especially focusing on the flood aspects first.

That is our first goal, and really start to put those flood monitors and rain gauges out there and start to see that data analytics come to life and then bring uh information back to us, and then ultimately move into the operational and improvement role, always monitoring the system, always understanding how it can improve and understanding at the beginning of going live and at the end of going live there.

Well, hopefully, there is no end going live.

It will live on to tell us the information.

But that being said, as the as the AI has the opportunity to do the machine learning, it will start recognizing weather patterns and really start developing those corollaries for

us.

Here's a Gantt chart.

Obviously,

it's really hard to read on the screen, and I'm sure it's still pretty uh if you you saw it in person, it's pretty uh small font.