here with me again is colleen keller and geoffrey thomas, les abend. and michael kay are both here. colleen, you were first. you were instrumental in finding air france flight, and you did. tell me what you would do here at this particular juncture. >> well, what we should be doing, don, what we're doing now is trying to keep track of all the different areas where the aircraft could potentially have gone down. we still have possibilities that it could be farther north, it could have crashed farther south. you know, we keep hunting around in these different areas as new evidence comes in. the approach that we used in the air force search is called applying basian search theory. what this does is allows you to keep track of all the evidence you have and its uncertainties. and at any one time you can refer to a probability map to where the evidence points would be the highest likelihood areas to go search. that's where we would recommend day to day you put your effort. right now it looks like they're chasing a very hot clue that puts them in this northern area, but they're coming up emp

let's dig deeper right now with former navy operation research analyst colleen keller. she helped with the search for air france flight 447 which crashed back in 2009. also joining us cnn's richard quest. he's in new york as is colleen. joe johns just reported about this international effort to find the wreckage of the plane. so how do countries on a practical basis actually coordinate this rather complex investigation? and as we speak right now, it's daylight in perth, australia, and surveillance aircraft are getting ready to take off to fly to that area 1500 miles at sea. >> well, wolf, they should be using something equivalent to the incident command system which puts people in charge and has a whole staff underneath that person where all the data can come in and be consolidated in a single place, logistics can be applied to the problem, then new assets requested if necessary. i'm sure the australians are using this process. >> richard, how worried are authorities in australia right now and elsewhere, u.s., australians, others are looking that horrible weather could h

back with me now, my panel, colleen keller, geoffrey thomas, les abend and michael kay. geoffrey, to you first. you say the orange objects may be easier to locate again than previous mystery items that we have seen. why is that? >> well, if the orange objects are identified as from mh 370, then the university of western australia tracks all the current data, all of these -- where this area is large eddies in that area. and they can then track back to march 8th and tell the searchers exactly where that bit of debris came from. if they have the exact location of where it is picked up, on the exact day, they can track it back to where they came from on march the 8th. and that will then give the ocean shield exact coordinates to travel to. but of course, having said, that it won't be a pinpoint in the sea. it will be an area or probably maybe 30 or 40 square miles. that's kind of the accuracy you're talking about. that's a rough figure. and it can be with the pinger locator. the distance of being about a mile to pick up the ping from the black boxes. you've got, again, be ve

we ask colleen keller. she helped track down air france flight 447. david glee is joining us. it seems not all of the countries are happy about sharing the information. and lawsuits are coming. who better to talk about that subject is arthur wolk. a firm that specializes in crash and litigation. the investigation has focused a lot on the pilot of flight 370. former fbi profiler mary ellen o'toole can talk about how investigators are digging into the background. and the cockpit and final hours of flight, crucial. david sucui is joining me. as we are now ten days away from the flight recorder batteries running out, here is what i want to know. why is this taking so long? specifically, i'm still looking for an answer to a question i asked a guest here a week ago. are geo politics partly to blame? on friday, the search shifted to the southern indian ocean that had previously been the focus. the new area is several hundred miles northeast of the region previously targeted and four times as large as search teams were focused one day prior. this shift was due to radar data with the c

. >> i want to thank you both, colleen keller, steven wallace, for your expertise on this.e you'll be back, this has been quite the story. so appreciate it. >>> when we return, the search area is now 35,000 square miles. the "uss blue ridge" is searching. we'll talk to them next. at. soon, the world's most intelligent servers, designed by hp, will give ups over twice the performance, using forty percent less energy. multiply that across over a thousand locations, and they'll provide the same benefit to the environment as over 60,000 trees. that's a trend we can all get behind. (voseeker of the sublime.ro. you can separate runway ridiculousness... from fashion that flies off the shelves. and you...rent from national. because only national lets you choose any car in the aisle... and go. and only national is ranked highest in car rental customer satisfaction by j.d. power. (natalie) ooooh, i like your style. (vo) so do we, business pro. so do we. go national. go like a pro. i've quit for 75 days. 15 days, but not in a row. for the first time, you can use nicorette... even if y

. >>> joining me now, colleen keller, former operations research analyst for the u.s.y, a licensed pilot who helped find air france 447 when it crashed into the atlantic ocean in 2009. steven wallace is former director to for the faa's office of accident investigation, also a cnn aviation analyst. we are rejoined by chad sweet and richard aboulafia. thank you all for being here. we're going to solve the thing right now. right? so they have widened is is search area, something that is almost impossible it seems to me just in terms of how big it is. how do they winnow down, colleen, where you look in this gigantic area? >> something we were talking about earlier is a way of using negative information to contain the search area. by that i mean if there's a radar or something else that should have detected the aircraft in a certain area and did not, you can say the aircraft never made it to that area. we were talking about the radars on the coast, india, pakistan, some of those countries up there in that northern arc, if we could get them to look at their radar tapes and s

for more, joined by colleen keller, senior analyst at metron which provides advanced air traffic management products and services and investigated the june follow-up 29th accident happenance of air fraunings and editor of flying magazinence an editor of flying magazine . >> a new flight path was programmed into the on-board computer as this latest report would have it, this happened just before the last control? >> good evening, antonio. this is something that i wrote about last friday. it's something that apparently, the investigators were not looking seriously at and until i came out with the piece and there were a number of pieces that followed on after that. the point is that when you program something in to the computer that they call the flight management system or the f.m. 4 -- f.m. 4 s, it can drive the autopilot to make deviation based upon what the computer programming language is. they flew to a series of way points like checkpoints in the sky that are virtual. there is no checkpoint. nothing that's other than a computer recognition of where they are in space, and they went to ve

back with me now my team of experts, david soucie, les abend, miles o'brien and colleen keller.k you two were having a fascinating discussion about a potential theory you think may have happened. >> les and i have been talk a lot about dropping down to 12,000 feet. taking some assumptions into that it did go down to 12,000 feet. if they were using a flight change button it was preset to that for emergencies. it could have been press and gone down. if there was a mechanical failure on board the aircraft. so i'm thinking along the lines of here, and miles keep me honest on this because i know there are some holes in here and we'll probably get there. but the only thing that makes sense to me, most probable and confidence i have in the data, this aircraft had some mechanical failure on board, namely the lithium batteries. if the lithium batteries started on fire. remember we talked about and mary has brought this up before it has the capability of putting out those fires that are going on. but there's another hazard that comes in with that. if you put out the fire, the lithium batt

i want to bring in colleen keller, a senior analyst with metro, inc. she helped in the search for air france flight 447. your firm has been hired. let's talk about this new search area and the new debris spotted by the five planes, all different colors. first of all, i want to get your take on this. >> well, it was an interesting curveball, carol. we'll just have to go with the flow here. this is the nature of search. you know, we started with it could be anywhere in the -- in that hemisphere of the world and we've narrowed it down to the southern indian ocean now. this is just -- you know, when new data comes in you have to roll and move to the new area. it's not throwing away efforts that say that we've searched in the south. we actually -- you know, we never picked up a piece of debris so to look at it that way, we really -- we never saw the airplane down there. the fact that new information has come up and moved us up to the north, it's good from a logistical sense that we're closer to land. we can get the towed ping gers out there and i know that

. >> let's bring down colleen keller. senior analyst that helped with search for flight 447. tell us about the search underway. this is an area they are searching that is a whole lot bigger than the search in the atlantic for the air france plane that went down. >> we are covering an enormous area with a large number of search assets in a multiple day search here. what they are doing right now. you might wonder how they pick the size of search areas they cover with different aircraft. i'm sure they are using a process that is outlined in the international air nautical search manual which takes into account visibility, sea state, size of target, altitude and speed of search aircraft. most are probably using visual search looking out the window, trying to pick up the signature, just a visual of pieces of floating debris. and the manualual will tell them how long they have to stay on station, how close they think they need to get before they can see something. and that is how these search areas are laid out. >> i want you to stand by. i think we have re-connected with tom forema

. >> joining us is hank whitmore and underwater search expert colleen keller. let me go first to you, hank. the new information tonight about the words coming out of that cockpit. i don't know, is it profoundly different although i think there is probably a lot of frustrations on everybody's part that the malaysian government led us to believe it was one set of words and now it's something different and we were told it was the co-by hot who uttered them and now we don't know who resulterred them. make any difference at all in this investigation? >> i don't think there is any substantive difference in the way the words are used. the only thing that seems to make a difference is the fact that they don't have -- there is a changes in the words. and it seems like if the tape said one thing, that's what it would say. it sounds like a substitution or a taper a tape that was cut off. and they discovered the tail end of it it's hard to understand what the significance could be. >> the one other thing i don't quite understand about it is that 1:19 a.m. is when we were t

let's talk to colleen keller, a special analyst, a defense contract or who helped search for flight had atlantic. we know the australians have said that they spotted a gray or green circular object in this general vicinity. also, an orange rectangular object. how tough should it be for the ships to find these objects and determine whether or not they are wreckage from the plane? >> it's -- it's hard to imagine but i've seen pictures of the seas. they are seeing some pretty high seas there, wolf, with wind-whipped waves and low visibility. it can be pretty difficult. anyone who has been on the ocean can attest to that. would you think that the colors would help and i heard that they dropped some flares so there's smoke or fire marking these objects. so it shouldn't be too difficult but then again, with the heavy seas, they might have their work made out for them. >> how quickly could these objects float in a different direction given the ocean currents? >> i would hope that they have the drift buoys in the water. the coast guard routinely drops these kinds of buoys when they have a man o

us now, the oceanographer from the research center at new south wales and also joining us is colleen kellertron which helped search for air france flight 447. colleen, what do you make of these large clusters, items, whatever they are seen in the satellite images. do you think it could be the plane? >> you know, wolf, this is such a remote part of the ocean that seeing large groups of objects like this, what could consider to be a drifting pattern of aircraft parts, i think it's very encouraging to see these objects close together. the simulations that we were just discussing there that my colleague is doing predicted that the debris would stay fairly close together, clumped together as it drifted around. so this is consistent with the models and i'm very encouraged by the recent information. >> let me bring eric in. you know the indian ocean well, eric. if this plane crashed, let's say after 19 days, and there's debris from the plane, wreckage from the plane, you know the waters, you know the currents, you know the weather. would they all generally be in the same area presumably or would t

joined by navy research analyst, colleen keller.which crashed into the atlantic. it took two years to find it and days to find the location. key of what you learned there, colleen, is that you have to question all assumptions, right? because there was an issue there with what you assumed about what were then known as black boxes wound up being untrue. remind us. >> yeah. the black boxes are equipped with an underwater beacon that sends out an acoustic ping and we looked at historical data on black boxes in the past in crashes and we found that 90% of the time they survive the crash and they were operating correctly. we thought that was pretty good assumption. we gave very good credit to the search listening to the pinging boxes. it turns out this was the one time that it didn't work. both boxes pingers were broken or failed in the crash and they weren't sending a signal at all. we had to revisit the assumptions and consider that and that's when we came up with the map that led us to the crash. >> even when times are darkest, you hav

colleen keller helped with the air france airliner in 2009 and peter golds, a former managing directorf the national transportation safety board. colleen, you helped with the search for the air france flight 447. it took five days to find some initial debris, but then another two years to recover the flight data recorder. give us a sense of how daunting this current challenge is. >> well, if you compare it to the air france search, it's much more daunting, wolf. this area is much larger and we still don't have a very good motivation for what happened to the plane. whether it was a hijacking or some kind of mechanical failure. we knew in the air france search that we were looking at a mechanical failure that brought the plane down within four minutes of the last known point. here, we don't have that kind of information. if the airplane did go into the water or over land, we don't know how it struck the surface or really anything. we're still searching in what we call a wide area search using wide area search centers. >> the first few days were wasteded basically. they were lookinging in

ervik van seville of the climate change research center from new south wales in australia, and colleen kellerense contractor that helped with the search. thanks to both of you for joining us. so you look at these pictures, 122 image, colleen, on this latest satellite image. it was taken on sunday. yesterday they went out, they've been looking all day. apparently no one has seen anything from the air or on the sea, but what do you make of these images? >> well, wolf, until now we've seen wednesdays and tuesdays, you know, a couple objects floating in the water. but this is what we would expect. we would expect a full debris field. the real question was, what did the currents do to that debris in the last 16 days? my colleagues back in my home office back in virginia have been assembling models of the currents. they're using the hybrid coordinate ocean model and they're running similar leighs to see what a debris pattern would do over a 16-day period. to their surprise, the debris is staying fairly close together and moving in an easterly southeasterly direction so that looks consistent with wh

colleen keller, who helped locate an air france flight in the atlantic ocean, suggests a return to theld search zone if nothing's found soon. >> right now, it looks like they're chasing a very hot clue that puts them in this northern area, but they're coming up empty, which is lending credence to the fact that maybe it's not the right place to be looking. so, it might be that after a couple more days, we should be looking back south, because we did see satellite pictures of large numbers of drifting stuff in the ocean. so, they should be keeping track of all their evidence and all their clues and where they've looked to date and be looking at, okay, where's the next best place to look tomorrow? >> we're also learning more about the malaysian government reaching out to the cia and its counterparts in britain and china in the flight 370 investigation. they want to take a deeper look again at the passengers and crew. it is fueling new speculation about what may have happened to the jetliner. so, let's go to cnn's jim clancy in kuala lumpur. he's been following the investigation really fro

i'm joined by former navy operation research analyst colleen keller. yst for metron. we know she helped in the search for air france 447, which crashed in the atlantic. also with me is david funk. colleen, obviously this is encouraging news. you try not to get hopes up too high that this debris has been spotted, not just on radar or satellite but visually seen from a search plane. >> yeah, we've been waiting for a break like this. it's very important to get our hands on some debris. it's still not clear it debris from the aircraft. but we have to verify every piece of junk we see out there and check it out to discount it if it not part of the aircraft. >> does it sound hopeful to you? they had more descriptive words. they talked about color, shape and size, not just a murky image on a satellite. >> the previous sightings seemed like things that fall off ships and float around. i have to caution everybody, there's a lot of junk out there. it amazing we haven't found more, to me at least. i'm withholding judgment until we get our hands on it. let's say th

. >> keller and his girlfriend, colleen rowley, are amongst one dozen burghley victims. burgerlary victims. >> it just feels like we have been taken advantage of. >> after the fire, the building was fenced off. at the time of the burglaries, of private security guard was supposed to be on duty. say they are asking the same question. site, and security on for something like this to take place, these people do not deserve to be victimized twice. >> tonight there may be a bit of a break in this investigation. police responded to the report of a robbery, a street mugging nearby. it was a couple miles away. when police went into those woods, they actually found a cache of stolen material, including six of the seven guns taken from the man we interviewed. thosef that material, guns, have been recovered. police have still not identified any suspects. they are not commenting on what has happened. you can see the security truck is out here tonight. many more people seem to be guarding the perimeter. brad bell, abc 7 news. >> disrupted traffic around rfk stadium this afternoon. tw