Implementing CBOSS

hcroze's picture
Sat, 2007-07-21 06:52 by hcroze · Forum/category:

In the context of CBOSS (the Cell-Based Observational Sampling System), we need to get cracking on outfitting the team of Maasai Scouts with handheld GPS units and mobile phones. I've started looking seriously on the web, since the local Garmin dealer was quoting with a mark-up of 250% and seems to be uncertain about availability of models (says the basic eTrex will not be any longer supplied as of end-July -- no news of this at http://www.garmin.com/). We need 15 units (one each for the scouts, one for David Sitonik for training, one spare).

So, at least five questions we can address in this topic:

  1. Data scope and collection protocols: How can we make best use of the nascent CBOSS proposal to organise the scouts to gather systematic and reliable data on elephant occupancy of the ecosystem.
  2. Ideal GPS model: best combination of functionality (track capacities, waypoints), ruggedness, battery life, size, etc.
  3. Ideal mobile phone model: low cost, minimun feature set, possibility of GPS/GPRS tracking...
  4. Sourcing & Shipping: most efficient, cost-effective route to get the units here.
  5. Training & deployment: training the scouts and getting them reporting systematically.

If you have comments on any one of these four items, please click on its link and add the comment under the appropriate heading (they've come out in decending order - sorry about that). Thanks.

AttachmentSize
Garmin _ Compare_eTrex_toprange.pdf163.06 KB
Garmin _ Compare_eTrex_midrange.pdf173.46 KB
Garmin _ Compare_eTrex_lowerrange.pdf166.92 KB
hcroze's picture

1. Data scope and collection protocols

Thu, 2007-08-02 18:39 by hcroze

To cover a number of data-related topics, for example: What is the breadth and depth of data that we can expect the Maasia Scouts to collect? How to we ensure a reliable sample frame? How do we account and correct for sampling biases? How can we expand the basic protocol (eles, signs) to include usage patterns such as trails as an input to ecosystem planning and management? And, I'm sure, many more.

_________
HC

hcroze's picture

GPS/GSM collars

Sun, 2007-08-12 06:58 by hcroze

One component of CBOSS will be advantitious deployment of GPS/GSM collars (see note below in reply to this intro). The National Geographic Society has provided us with a grant that will defray part of the costs. Iain Douglas-Hamilton has offered to help with some technical advise to get us back into tracking mode: we haven't put a collar onto an Amboseli elephant since the 1980s. Meanwhile, our Tanzanian colleague, Alfred Kikoti, has collared at least ten Amboseli eles on the other side of the border. Several meetings have been held with Alfred to discuss identities and data sharing.

_________
HC

Collars: towards a strategy for deployment

Wed, 2010-03-31 11:33 by Keith

I agree with Pili's points (below) and in addition would propose collaring families (and males?) that we suspect are using known corridor areas, esp those at risk of closure, so that we can get a good idea of the use of those potential choke points. Also, possibly families and males that appear to use areas of HEC, either pastoralist or agricultural conflict.

There's my fitty cents worth for now.
K

On Wed, Mar 31, 2010 at 9:39 AM, Phyllis Lee wrote:

Dear H

.....It is indeed time to start prioritising families and males... I have a couple of suggestions to start the ball rolling (sorry it's by email)

C needs to decide which groups will be most traumatised by matriarch loss and we could target those to see where they go now.

Males - we need to pick up some known family bulls c. 28-20 and see where they go. Preferably guys we don't know that well. An analysis of the males most seen in the last year will tell us who we DON'T need.

hcroze's picture

Report to USFWS

Mon, 2010-03-29 11:37 by hcroze

The following is an extract from ATE terminal report to USFWS.
...the prolonged drought and the extremely poor condition of the elephants over the past year have delayed deployment. One collar, on Vicky of the VA family had already been deployed in late December 2007 (see http://elephanttrust.org/node/439) and it provided an excellent picture of movements to the north (see map to right) compiled from two-hourly fixes of the VAs over two years. The VAs have clearly defined two clear corridors to the north and northeast and two likely ones to the northwest. They have also demonstrated the effectiveness of the Selenkay Community Conservation Area as a dispersal refuge for Amboseli elephants.

The battery has now expired and the collar will be removed when her 14-month calf is at least three years old. After the rains began, a second collar was put onto Kate, an adult female from the KB-2 family, which has already begun to confirm corridor use to the southeast.

The project now has six collars ready for deployment, one satellite GPS, and five GPRS/GSM models. Two will be put on family groups to the south and west, and two on wide-ranging bulls.

Vicky and the VAs
Vicky and the VAs, 2008-09

hcroze's picture

Blurb on collars for Born Free

Sun, 2007-08-12 07:02 by hcroze

Winnie Kiiru asked for a couple of paragraphs of background information for a Born Free event in November 2007 in which wildlife artist Angela Chidgey will donate the proceeds of sales to AERP, specifically for elephant tracking collars. (Thanks, Winnie!)

Herewith a draft, based on the rather terse extraction from the proposal:

Trunk calls: mobile telephones and elephant tracking

The growing network of mobile telephone cells is not only improving life in populated areas: it's helping elephant conservation in the bush.

New satellite-based technology to track large animals is now available in Africa. In Kenya, for example, using the growing commercial mobile telephony services, wildlife researchers can tag animals with small communication devices that tell them where the animals go, when, and what they are doing.

The devices make use of GPS and GSM technologies. GPS, for the Global Positioning System of 24 communications satellites orbiting more than 12 miles above the Earth that provide geographic fixes for devices on the ground. GSM, for the Global System for Mobile communications, the most popular standard for mobile phones in the world, used by more than two billion people in 212 countries and territories.

Getting the devices onto an animal involves immobilizing it with narcotics shot from the ground or from helicopter, and then fitting the animal with a collar. For an elephant, it's a major exercise, one that is highly invasive, nearly akin to a selective cull of one individual from a family group.

AERP policy has been for a number of years to avoid at all costs betraying the trust that has been built up between researchers and the elephants over more than three decades of peaceful proximity. The project still feels that such trust is priceless.

However, on occasion it is necessary to immobilize and treat an animal that has been wounded unnaturally by spearing, poison arrow or gunshot. In such a case, since the animal is down anyway, AERP proposes to take advantage of the opportunity to put a GPS/GSM-enabled collar on the patient (if the attending vet judges that the collar would not be inimical to the animal’s recovery).

The strategy would be to have up to ten collars on-hand and ready for deployment. Since AERP staff attends all KWS immobilizations of injured elephants to ensure accurate identification and to take measurements for growth-age studies, it would be a simple matter to attach a collar while the vet is treating the sleeping ele.

The data from the collared individuals will provide valuable additional data on where they and their families go, particularly into the surrounding Maasai group ranches and into Tanzania, in order to establish preferred dispersal areas and the main routes used to get to them. The information will help in negotiations for elephant corridors, as well provide early warning for possibly conflict as the elephants move towards populated areas.

_________
HC

hcroze's picture

Trail data

Wed, 2007-08-08 07:11 by hcroze

There is a recently posted Ecosystem Forum topic on Trails & Corridors. Please add your comments there.

_________
HC

hcroze's picture

5. Training & deployment

Sat, 2007-07-21 07:22 by hcroze

We intend to employ a local Maasai as a consultant to coordinate the work of the Maasai Scouts for CBOSS. Hopefully one of our past Scout trainees, who have attended the courses coordinated by Norah Njiraini in working with scouts as part of the training component of ATE's Community Outreach sub-programme.

The substance and timetable of the workplan is being worked out.

_________
HC

hcroze's picture

4. Sourcing & shipping

Sat, 2007-07-21 07:16 by hcroze

If we opt to get the GPS units and/or the mobile phones from a supplier in the USA or Europe (Amazon, West Marine, Garmin itself, Nokia itself) then there is the issue of sourcing and shipping. Sourcing locally will obviously be more expensive, but then there is the goodwill for subsequent service. Depends on the price differences. Currently under investigation.

_________
HC

hcroze's picture

3. Ideal mobile phone model

Sat, 2007-07-21 07:10 by hcroze

We also need to get the scouts mobile phones for reporting to Sitonik and Soila, both for routine monitoring of elephant presence and signs as well as for 'early warning' and alterting to impending incidents in the ecosystem.

We need first to identify ideal models. Same considerations as with the GPS: basic functionality, ruggedness, possibility of GRPS/GSM tracking (see Specifications in comment below).

_________
HC

hcroze's picture

Phone specifications

Thu, 2007-07-26 06:04 by hcroze

Rich Owings, whom I contacted through his blogs (see http://GPSTracklog.typepad.com and http://GPSTrackingSystems.biz), has asked for more clarity in our requirements. He says, "Are you looking for handheld GPS units or a cell phone tracking system or...?" So...

The cellphones are required for the usual phone functions, in particular to report to Soila or David if there is an incident that requires their attention (or that of KWS, the police, etc). It's to aid rapid response. If the GPS functions outlined above could be integrated into the phone, all well and good, bearing in mind the following:

  • There is a basic requirement of robustness: the slick little sliding jobs that impress at coffee shops won't do in the bush.
  • A Bluetooth solution between phone and GPS is probably too fiddley and complicated for the scouts, not to mention the increased battery drain.
  • The whole raft of today's mobile phone functions would probably be overkill. I'm ambivalent about the camera. On balance, a camera could be done without. If it is included willy-nilly, then it might find some use in recording details of Incidents.

Cost is obviously a consideration. If the units are less than around $200 each, we could probably squeeze out of the current budget. Otherwise, we would need donor assistance.

_________
HC

hcroze's picture

2. Ideal GPS model

Sat, 2007-07-21 07:00 by hcroze

The Garmin family first comes to mind, something of the ilk of eTrex or the GPS 60. I've downloaded some comparisons from http://www.garmin.com/ and will try to summarise over next few days. Although the local agent (Titan Aviation at Wilson Aerodrome) seems a bit disorganised, at least there is an agent). Magellan seems less promising, and I haven't found an agent yet.

Ideally it would be great if we could find a robust device that would have both mobile phone and most GPS functions (particularly tracking and waypoint recording), but no one seems to know if there is anything suitable out there yet.
_________
HC

hcroze's picture

GPS specifications

Thu, 2007-07-26 05:57 by hcroze

Rich Owings, whom I contacted through his blogs (see http://GPSTracklog.typepad.com and http://GPSTrackingSystems.biz), has asked for more clarity in our requirements. He says, "Are you looking for handheld GPS units or a cell phone tracking system or...?" So...

The basic GPS requirements the Maasai Scouts need are (a) waypoint recording (location, date and type), (b) tracking (primarily to create a periodic record of where the scout has wandered).

Concerning (a), the 'typology' would ideally be specialised, that is, instead of Golf Courses and Picnic Sites, it would include;

  • Elephant Dung, Broken Vegetation (evidence of feeding eles);
  • Tracks (with an estimate of numbers of animals);
  • Eles Seen (with a count of numbers in group);
  • Incidents (depending on the data-logging capacity the unit, this could range from a simple waypoint number with referenced annotations in a notebook detailing uncommon occurrences, for example, a poaching incident, an elephant killing a goat, a broken borehole pipe, etc, etc.

Obviously, if programming a specialised typology in the device would make it hugely expensive, then we have to devise a work-around, for example, a convention that would code the built-in iconography of the GPS to our own needs (Golf Course = dung piles, Picnic Site = broken vegetation, etc.

Concerning (b), tracking, we want to create a record of search effort expended by the scout, either by simple inspection of his trail as to where he went and where he didn't go or by creating kernel surfaces of probability using a suitable tool in ArgGIS. The idea is to distinguish between genuine nulls and no sample (i.e., were there no sightings in a particular area because there were no eles there or because the scout didn't visit that area?).
_________
HC

hcroze's picture

Tracking management system

Thu, 2007-08-23 22:00 by hcroze

We have entered into contract a local firm that is doing GPS-based fleet management, RiverCross Technologies Ltd that has worked for several years with Save the Elephants.

The RiverCross BlueTrax system can ingest and export ArcGIS shapefiles, so the client's work area can be customized with topomaps, waypoints, etc. We currently have one unit on the new project Toyota Landcruiser and on elephant from the VA family, Vicky. More information to come...

GPS suggestions

Thu, 2007-07-26 16:22 by Rich Owings

First of all, I would definitely recommend a Garmin. GPS devices are complex, but Garmin has the most intuitive interface of any manufacturer.

Second, consider reception. If you are dealing with any areas of dense canopy or other barriers to reception, consider a unit with a high-sensitivity receiver, like the new Garmin eTrex H-series (http://gpstracklog.typepad.com/gps_tracklog/2007/05/new_garmin_etre.html). A pricier solution would be the 60Cx.

Third, is there value in tracking / recording elevation? If so, a unit with a barometric altimeter may be useful. This can always be estimated from lat/long if good enough topo maps are available.

Hope this helps.

hcroze's picture

Garmin it is, then...

Fri, 2007-07-27 06:31 by hcroze

The consensus is clear: Garmin. Certainly would appear to be the best choice with regard to local support.

We have no real canopy issues in the Amboseli ecosystem -- mostlyopen bushed grassland. Also, we are not really interested in elevation: the ecosystem is generally flat and there are good topo maps at 1:50k.

The eTrex H looks fine, particularly with regard to price. Just talked to the local agents (Titan Avionics). They don't have the units in yet: two weeks. They are quoting $200 per unit, which is quite a markup. They muttered something about a 10% discount if we buy 15, but are putting the question to the new boss (Raju). I wouldn't mind a reasonalble markup for shipping, clearing, support good will etc, but 100% is a bit steep.

Two technical issues with the eTrex H. One is that it doesn't seem to have a USB interface. Not a terminal condition, but it means we'll need to use serial-USB converters. The other is that the battery life is listed as 17 hours, which is a bit low compared to the 22-32 hrs rated for other units (GPS60, 28 hrs!). Again, perhaps not a terminal excluder, since it's somewhat better than the low value of 12 hrs I've seen for others.

The GPS 60 was also the next on my list of possibilities. Long battery life, USB interface, rugged. It is pricey ($194 on the Garmin website, $475 (!!) quoted by Titan. I shall write the director about a possible discount on that one, too.

_________
HC

PS: This just in. Titan has quoted $140 for the eTrex H, Four week delivery time.

Hans's picture

eTrex vs. GPSMAP 60C

Fri, 2007-07-27 07:37 by Hans

That's a mapping GPS (I own its brother, the 76C, same electronics in a different case), and meanwhile obsolete through the newer Garmin GPSMAP 60Cx.

But since we don't need mapping (no loadable maps available for the area), it is actually not the right device.

Haven't looked at the eTrex model range, but isn't there an eTrex that does exactly what we need, including a USB port?

17 hours is plenty, particularly since the device doesn't have to be on all the time. I guess that would last almost a week in normal operation.

Hans-Georg

hcroze's picture

eTrex family of GPSs

Fri, 2007-07-27 13:49 by hcroze

Good point, Hans. To get a better view of all the eTrex's, I made three compare sheets (attached above under this topic intro) from the Garmin website, for the lower, middle and top range units. It's not till you get to the middle range and the over-$200 territory that there's a USB. You also start to get overkill of features, including compass, micro SD card slot, that sort of thing.

Need to study further, but looks like, if we can live without a USB interface and ensure a ready supply of batteries, then the basic xTrex H looks best for the money. If we really can live without a USB interface and battery life is an issue, then the Summit or Venture would seem best. If USB is essential, then the Venture Cx would be the choice (given that the Venture HC, also a USB, only has 14 hours of endurance). The USB would basically cost an extra $100 per unit, whereas, I think I have a USB-serial cable lying around here somewhere. Anyone know how seamlessly the cable works?

See what you think.

_________
HC

More eTrex considerations

Fri, 2007-07-27 14:35 by Rich Owings

Several more things to consider:

  • eTrex units bearing a "C" have a color screen and will be MUCH more legible in bright sunlight.
  • The basic eTrex in the past has not come with a cable, which must be purchased separately. It appears that this is also true of the new basic eTrex H.
  • It is possible to make your own maps for Garmin receivers. You would need someone with a degree of tech savvy, interest and time. Then you would need a mapping receiver like the Venture HC or Cx.
  • If you can live with the serial interface and monochrome screen, the basic eTrex H looks like it could be the best unit for you. Maybe you can get one and try it out before buying all of them.

Hans's picture

Serial connection, etc.

Fri, 2007-07-27 14:46 by Hans

I know from own experience and reports that not all USB-to-serial adapters work with the Garmin receivers. But I have used one myself, and that worked very well. We may have to test and try one or two. Fortunately they are not expensive, and once they work, they always work.

Mapping is probably not needed, as the researchers know the area very well. I've looked into map-making, but as you say, it's not for the faint-hearted.

The is a funny organization that sells you the special Garmin plugs. I've used them myself to make Garmin cables. But in our situation this is probably not a good way to do it. Fortunately we need only one or two of these cables, if I'm not mistaken.

I agree with the eTrex H choice. I suppose it can record the track, but am not absolutely sure.

Hans

eTrex H

Mon, 2007-07-30 15:13 by Rich Owings

My experience with serial:USB adapters is that the name brands tend to work fine.

You're right; you don't need a cable for each, unless each unit will be used with a separate computer at a separate site. Otherwise, one or two would do fine.

The eTrex H will record up to 10,000 track points and 10 tracks. It will also record 50 waypoints.

HTH

Hans's picture

Tracking

Thu, 2007-07-26 06:36 by Hans

A little ambiguity crept in here. The word tracking is often used for systems with the capability that the GPS device can be tracked down (particularly against the knowledge and will of the bearer). One technical solution for this is that the GPS receiver is combined with a mobile phone or satellite transmitter and reports its position that way.

For now we don't need that, particularly since such tracking devices are much more expensive, although it would be nice to stick such a device onto an elephant.

What we need, and where the word tracking comes in, is that the device stores its own track for later offloading into a computer. That's all that was meant when the word tracking was used here. Most handheld GPS receivers do that anyway.

Hans

hcroze's picture

HGM on batteries

Sat, 2007-07-21 07:29 by hcroze

In an email, 21-Jul, HGM writes:

"One afterthought: if you order GPS devices, think of a few quick-chargers and enough Sanyo "eneloop" rechargeable cells.

"I recommend to buy a few good ultra-quick chargers. Even though they are more expensive, they'll redeem some of their price by not needing double sets of batteries and by being able to fully recharge batteries during the limited times of electricity.

I think you need only buy as many batteries as the devices can hold plus very few spares, maybe two sets, if they are all kept in a centralized place like [ERH]."

hcroze's picture

Solar battery chargers

Fri, 2007-07-27 04:35 by hcroze

The Sanyo eneloop batteries (where did they get that name?) certainly appear to be among the best batteries around. A charger in the office at ERH would be only a partial (and necessary) solution. The Maasai Scouts are out in places with very few plugs in walls (actually, with very few walls at all). For them, we really need solar chargers. Need to search for a source of AA solar chargers.

+++++++++++++++++++++++++++++++
On 14-Aug-07, Steve Gold wrote:

"...your question of eneloops, I have no experience with this so I referred it to Bruce. He says:

" 'I did a quick web search and it appears that these are NiMH batteries (standard sizes AA, C,etc). The difference seems to be a lower self discharge rate. This would be useful to allow longer storage in a charged state. Solar charge regulators are by and large designed to work with Lead Acid Batteries (flooded and sealed). I am not aware of any commercially available solar charge regulators that are designed for other types of batteries. Bruce Roush, VP Technical Services, Solar Depot, 707-766-7727, xt 114.'

"There may be a way of charging these, but it might take some digging around. Perhaps Mark Eddy has some knowledge of this. You are charging the Nicads at higher voltages, so you might try one battery..."
++++++++++++++++++++++++++++++

Did more searching and found at least 4 solar chargers taking off from treehugger (seems a good site, but the way), ranging from ca. $20 to $170(!). Guess we will invest in a number of the lower-priced one for the Scouts.
_________
HC

Hans's picture

eneloop

Wed, 2007-08-15 07:59 by Hans

A little technical background info:

eneloop accumulators are modified NiMH (Nickel Metal Hydride) accumulators. Although they have a slightly higher voltage than ordinary NiMH accumulators, they can still be charged with any normal NiMH charger. Don't fall into the trap to buy expensive "eneloop chargers". (eneloops, like any NiMH cell, cannot be charged with a NiCd charger, except for slow chargers—see below.)

You probably want a more expensive quick charger that charges them in around two to three hours. Be very wary of time-controlled chargers, which will destroy your rechargeable batteries soon.

The other, very cheap solution is always a slow charger. When the charging current is low enough that a full charge takes 10 hours or more, the rechargeable batteries are immune from overcharging, so in that case you can use very cheap chargers, and the combination is totally idiot-proof and can be left connected in places where generators are switched off and on. I actually recommend that, but the charging times from totally empty to totally full can easily exceed 24 hours with many types of cheap, slow chargers. It's still nice to know that you can't destroy the accumulators that way.

Solar chargers are more of a problem. Very small ones that fulfill the 10 hours condition are cheap, but it would take two full days or more of sunshine to charge the accumulators from totally empty to totally full. Still this could be a good solution, if the users recharge one set of accumulators every day while using another. Assuming that the accumulators are still half full after one day of use, they could be recharged to full in one day with a cheap, weak solar charger.

The downside is that you need two sets of accumulators, if you want to use the GPS every day. If you need the GPS only every other day, the cheap solar charger would be the way to go. My only remaining worry would be high temperatures when you put such a charger out in the sun on a day with no wind with the accumulators inside. The accumulators will begin to take damage when their temperature goes above some limit, perhaps around 60°C, I don't know exactly.

Hans

hcroze's picture

HGM on devices

Sat, 2007-07-21 07:05 by hcroze

Email from HGM, 21-Jul:

"Personally I don't know such devices, but I'm pretty sure they exist. Motorola used to have one (based on some version of Linux, by the way), and other manufacturers surely have them too.

"They would not have the same rich functionality as the better Garmin receivers, but they may still have all the functions we need. That would have to be checked carefully.

"Note also that receivers that just track and take waypoints are very cheap, small, and light, like the eTrex family from Garmin. Some are worn like a watch, like the Garmin Foretrex (https://buy.garmin.com/shop/shop.do?cID=144 ).

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.