Paul/Will- LIDAR is new enough that it still takes judgement, and ground-truthing. One of the ways that LIDAR users try to improve their analysis is to have random (or systematic) plots of high density LIDAR (cost- and data storage-prohibitive for large areas). High density LIDAR has about 4 pie-plate sized 'hits' per square meter and can be depended on to obtain accuracy in the .5 to 1.5 foot height data. As to ground level accuracy, LIDAR can be confused when going in and out of swamp, as Paul explained. He's correct in his comment that there are other sources available to determine ground level data, although to get good accuracy the two layers need calibration and adjustment. I am glad that NPS folks were interested in using your climbing/sin-sin height data to ground-truth theirs. -Don
> From: [email protected] > To: [email protected] > Subject: [ENTS] Re: Lidar data availability and possible application > Date: Sat, 7 Mar 2009 13:11:47 -0600 > > > Will, > > I've been thinking about the excessively high 150' canopy in the swampy part > of Congaree. I can understand LiDAR underestimating height due to > reflection intensity on small limbs, but gross overestimating would have to > be due to an incorrect ground surface approximation. The cypress/tupelo > swamps have a unique condition of partial or complete standing water under > trees. Water can provide very high quality reflections such that multipath > errors could be possible. The laser could reflect off the water back to the > plane, but also reflect off the water then off the understory or large > trunks back to the water and then back to the plane, providing an estimate > of the ground level that is actually significantly beneath the actual > surface. I would suspect that if this were occurring, that the LiDAR ground > surface plot would probably not be as flat as expected there (only a few > feet of variation would be typical) and would be far lower than the known > elevation (depending upon the sample angle, up to half the difference of the > measured canopy height vs. real height). If anyone still has access to this > data, it would be simple to check it out. The flood plain levels are > probably well known. > > Paul > ----- Original Message ----- > From: "Will Blozan" <[email protected]> > To: <[email protected]> > Sent: Saturday, March 07, 2009 11:34 AM > Subject: [ENTS] Re: Lidar data availability and possible application > > > > > > Paul, Bruce, > > > > I briefly looked at the LiDAR data at Congaree last visit and found some > > glaring "errors". The highest forest canopy was generally the pine groves > > but one tupelo/cypress swamp had a 150'+ canopy over dozens of acres. That > > is simply not possible. > > > > Will F. Blozan > > President, Eastern Native Tree Society > > President, Appalachian Arborists, Inc. > > -----Original Message----- > > From: [email protected] [mailto:[email protected]] On > > Behalf Of Paul Jost > > Sent: Saturday, March 07, 2009 12:23 PM > > To: [email protected] > > Subject: [ENTS] Re: Lidar data availability and possible application > > > > > > Bruce, > > > > I don't expect LiDAR to measure tree heights very accurately, especially > > in > > free USGS data. I hope that it can help locate potential tall trees that > > can be found on the ground and then measured accurately by ENTS methods. > > It > > > > could help focus searches on large public lands that would otherwise take > > many lifetimes to effectively explore. > > > > Paul > > > > ----- Original Message ----- > > From: "Bruce Allen" <[email protected]> > > To: <[email protected]> > > Sent: Saturday, March 07, 2009 6:49 AM > > Subject: [ENTS] Re: Lidar data availability and possible application > > > > > > > > In the Congaree, Lidar doesn't seem to pick up the highest small > > branches. If you looked at the Park Services canopy height map it > > just doesn't seem to pick up the max height effectively. > > > > Bruce > > > > On Fri, Mar 6, 2009 at 11:32 PM, Paul Jost <[email protected]> wrote: > >> > >> Gary, > >> > >> With regards to the application of LiDAR data to our tall tree searches, > >> when I've worked out the bugs as much as reasonably possible, I'll post > >> some > >> screen shots and some info on the process so that others can do the same > >> thing for FREE!! It should be in the next week or so.... > >> > >> Paul > >> > >> ----- Original Message ----- > >> From: <[email protected]> > >> To: <[email protected]> > >> Sent: Friday, March 06, 2009 8:26 PM > >> Subject: [ENTS] Re: Lidar data availability and possible application > >> > >> > >>> Thanks Paul. I am enjoying your posts. > >>> > >>> Gary > >>> Sent via BlackBerry from T-Mobile > >>> > >>> -----Original Message----- > >>> From: Paul Jost <[email protected]> > >>> > >>> Date: Fri, 06 Mar 2009 18:45:03 > >>> To: <[email protected]> > >>> Subject: [ENTS] Re: Lidar data availability and possible application > >>> > >>> > >>> > >>> Gary, > >>> > >>> I'm using free LiDAR data from the USGS product page with free FUSION > >>> software from the U. of Washington. > >>> > >>> Paul > >>> > >>> ----- Original Message ----- > >>> From: <[email protected]> > >>> To: <[email protected]> > >>> Sent: Friday, March 06, 2009 6:34 PM > >>> Subject: [ENTS] Re: Lidar data availability and possible application > >>> > >>> > >>>> Paul, > >>>> > >>>> Just joining the list after a hiatus and I am interested in what you > >>>> are > >>>> doing. I have ArcGIS 9.3 and now a Magellan Mbile Mapper 6. Did you buy > >>>> the LIDAR? > >>>> > >>>> Gary > >>>> Sent via BlackBerry from T-Mobile > >>>> > >>>> -----Original Message----- > >>>> From: Paul Jost <[email protected]> > >>>> > >>>> Date: Fri, 06 Mar 2009 18:28:27 > >>>> To: <[email protected]> > >>>> Subject: [ENTS] Re: Lidar data availability and possible application > >>>> > >>>> > >>>> > >>>> Josh, > >>>> > >>>> I've been working on the Wright Creek LiDAR data and need to finish > >>>> working > >>>> on optimizing the generation of the bare earth topographical layer from > >>>> the > >>>> LiDAR data. Preliminary work shows quite a few trees in the 140's, a > >>>> few > >>>> in > >>>> the 150's, and possibly one around 171 feet tall. This software is > >>>> pretty > >>>> cool. I can measure individual trees, color by tree height, etc. > >>>> > >>>> I need to verify if any are outliers or noise but would like to ground > >>>> check > >>>> some. I also haven't checked out much of the creek yet. Did you GPS any > >>>> of > >>>> the trees when you were there? If so, can you enter a user coordinate > >>>> system in your GPS? I'm working in State Plane for NC in feet, the form > >>>> that the LiDAR data comes in. I can give you info on how to set up user > >>>> coordinate systems in your GPS if you used it there. Otherwise, I'll > >>>> have > >>>> to find a way to convert the coordinates myself - easier to do with the > >>>> more > >>>> expensive GIS software. > >>>> > >>>> Paul > >>>> > >>>> ----- Original Message ----- > >>>> From: "Josh Kelly" <[email protected]> > >>>> To: "ENTSTrees" <[email protected]> > >>>> Sent: Monday, March 02, 2009 3:56 PM > >>>> Subject: [ENTS] Re: Lidar data availability and possible application > >>>> > >>>> > >>>>> > >>>>> Paul, > >>>>> > >>>>> Let me know how using the LIDAR data around Santeetlah goes. If you > >>>>> find any great trees, I could ground truth them next time I return to > >>>>> the area. I'm quite sure there are 160' poplars in the area (second > >>>>> growth) and maybe some 170's. > >>>>> > >>>>> Josh > >>>>> > > >>>>> > >>>> > >>>> > >>>> > >>>> > >>>> > >>>> > > >>>> > >>> > >>> > >>> > >>> > >>> > >>> > > >>> > >> > >> > >> > >> > > >> > > > > > > > > -- > > Bruce P. Allen > > Springfield, NH > > > > > > > > > > > > > > > > > > > > > > > > _________________________________________________________________ Windows Live™: Life without walls. http://windowslive.com/explore?ocid=TXT_TAGLM_WL_allup_1a_explore_032009 --~--~---------~--~----~------------~-------~--~----~ Eastern Native Tree Society http://www.nativetreesociety.org Send email to [email protected] Visit this group at http://groups.google.com/group/entstrees?hl=en To unsubscribe send email to [email protected] -~----------~----~----~----~------~----~------~--~---
