1 00:00:09,630 --> 00:00:10,463 - [Joshua] Good morning. 2 00:00:10,463 --> 00:00:11,590 My name's Joshua Blouin 3 00:00:11,590 --> 00:00:14,270 and I'm a master's student at the University of Vermont. 4 00:00:14,270 --> 00:00:17,050 And this morning I'd like to discuss the first chapter 5 00:00:17,050 --> 00:00:18,290 of my thesis: 6 00:00:18,290 --> 00:00:22,130 Modeling Moose Habitat Suitability by Age, Sex 7 00:00:22,130 --> 00:00:23,870 and Season in Vermont, 8 00:00:23,870 --> 00:00:27,313 based on GPS Radio-collar Data and LIDAR Imagery. 9 00:00:31,350 --> 00:00:34,170 So I'd like to start by just quickly introducing 10 00:00:34,170 --> 00:00:36,960 some of the moose population trends over the years 11 00:00:36,960 --> 00:00:38,620 here in the state. 12 00:00:38,620 --> 00:00:41,660 So due to time constraints, I won't go into great detail 13 00:00:41,660 --> 00:00:44,220 but just know that moose were excavated from Vermont 14 00:00:44,220 --> 00:00:46,420 by the late 1800s. 15 00:00:46,420 --> 00:00:48,910 But by the early 200s, they were exceeding 16 00:00:48,910 --> 00:00:52,440 both social and biological carrying capacity, 17 00:00:52,440 --> 00:00:55,240 particularly in the Northeast corner of Vermont. 18 00:00:55,240 --> 00:00:57,810 And so a permitted hunt was put into place 19 00:00:57,810 --> 00:00:59,840 by the Department of F and Wildlife 20 00:00:59,840 --> 00:01:04,000 to purposefully drop that population down a bit. 21 00:01:04,000 --> 00:01:06,830 This was successful, but managers were noticing 22 00:01:06,830 --> 00:01:11,373 that there was continued decline of the population. 23 00:01:17,520 --> 00:01:19,730 This continued decline initiated 24 00:01:19,730 --> 00:01:21,770 Vermont's first research project 25 00:01:21,770 --> 00:01:24,970 between 2017 and 2019. 26 00:01:24,970 --> 00:01:28,570 126 moose were fixed with GPS radio collars 27 00:01:28,570 --> 00:01:31,560 and wildlife management units E1 and E2 28 00:01:31,560 --> 00:01:33,850 in the Northeast corner of Vermont. 29 00:01:33,850 --> 00:01:35,870 And the purpose of this research was to really 30 00:01:35,870 --> 00:01:38,630 examine mortality and productivity rates 31 00:01:38,630 --> 00:01:40,463 for Vermont's largest moose herd. 32 00:01:41,610 --> 00:01:44,900 So this study and the results from Jacob DeBow's thesis work 33 00:01:44,900 --> 00:01:49,900 at UVM showed that heavy infestations of winter (silence) 34 00:01:52,110 --> 00:01:57,110 we're a driving force for mortality in our study area. 35 00:01:57,260 --> 00:02:02,260 And when I say heavy infestation 17 and 2019 36 00:02:02,850 --> 00:02:06,470 126 moose were fixed with GPS radio collars 37 00:02:06,470 --> 00:02:09,440 and wildlife management units E1 and E2 38 00:02:09,440 --> 00:02:11,740 in the Northeast corner of Vermont. 39 00:02:11,740 --> 00:02:13,130 And the purpose of this research 40 00:02:13,130 --> 00:02:15,080 was to really examine mortality 41 00:02:15,080 --> 00:02:18,333 and productivity rates for Vermont's largest moose herd. 42 00:02:20,640 --> 00:02:22,620 And so with this, we're beginning to really 43 00:02:22,620 --> 00:02:25,460 understand what mortality and reproductive rates 44 00:02:25,460 --> 00:02:29,000 look like for Vermont's largest moose population 45 00:02:29,000 --> 00:02:31,460 but an understanding of the relative suitability 46 00:02:31,460 --> 00:02:34,790 of habitats in a landscape will really provide a foundation 47 00:02:34,790 --> 00:02:38,420 for a population management here in the state 48 00:02:38,420 --> 00:02:39,660 and is particularly important 49 00:02:39,660 --> 00:02:41,893 for a species in decline, such as moose. 50 00:02:45,310 --> 00:02:47,920 So what do we know about moose habitat in Vermont? 51 00:02:47,920 --> 00:02:51,690 Well, we know that a habitat suitability index model 52 00:02:51,690 --> 00:02:53,690 was developed in the 1980s by the 53 00:02:53,690 --> 00:02:55,480 U.S. Fish and Wildlife Service 54 00:02:55,480 --> 00:02:58,560 which was then applied to two wildlife management units 55 00:02:58,560 --> 00:03:01,940 here in Vermont nearly 20 years ago. 56 00:03:01,940 --> 00:03:05,560 And although this HSI really informed management 57 00:03:05,560 --> 00:03:07,500 about habitat used by moose 58 00:03:07,500 --> 00:03:10,930 and made use of the best available information at the time 59 00:03:10,930 --> 00:03:12,530 it did have several limitations 60 00:03:12,530 --> 00:03:14,620 and some of those limitations were 61 00:03:14,620 --> 00:03:16,240 the fact that it was based on assumptions 62 00:03:16,240 --> 00:03:19,490 from Northern Minnesota about habitat use 63 00:03:19,490 --> 00:03:22,950 and the fact that there were no radio collar mooose here 64 00:03:22,950 --> 00:03:24,153 in Vermont. 65 00:03:26,300 --> 00:03:29,550 We proposed to create this upgraded HSI model 66 00:03:29,550 --> 00:03:33,230 for Vermont moose based on really these three new sources 67 00:03:33,230 --> 00:03:38,230 of information, GPS radio-collar locations of moose. 68 00:03:38,680 --> 00:03:41,850 So actually having radio collared animals on the landscape 69 00:03:41,850 --> 00:03:46,850 these contemporary maps of composition using NLCD 70 00:03:47,150 --> 00:03:51,010 and these extremely fine scale examinations 71 00:03:51,010 --> 00:03:53,710 of habitat structure using LIDAR 72 00:03:53,710 --> 00:03:56,560 which I'll talk a little bit more about here in a second. 73 00:03:58,360 --> 00:04:01,720 The overarching objectives of this paper were to 74 00:04:01,720 --> 00:04:04,930 develop contemporary habitat, suitability models 75 00:04:04,930 --> 00:04:08,460 by age, season, and sex for our moose, 76 00:04:08,460 --> 00:04:11,933 as well as mapping that suitability across our study area. 77 00:04:14,810 --> 00:04:17,890 Our overall process for creating 78 00:04:17,890 --> 00:04:19,880 these resource utilization functions 79 00:04:19,880 --> 00:04:21,790 and these utilization distributions 80 00:04:21,790 --> 00:04:24,210 to really examine habitat use 81 00:04:24,210 --> 00:04:25,860 within an individual's home range 82 00:04:25,860 --> 00:04:29,610 is outlined by a (indistinct) 83 00:04:29,610 --> 00:04:31,340 And the first step in the process 84 00:04:31,340 --> 00:04:36,340 is to filter our GPS collar data by these two seasons. 85 00:04:36,870 --> 00:04:39,700 So we examined habitat use annually, 86 00:04:39,700 --> 00:04:41,240 but by the dormant 87 00:04:41,240 --> 00:04:42,670 and the growth season, 88 00:04:42,670 --> 00:04:44,790 as well as by age. 89 00:04:44,790 --> 00:04:49,080 So young adult or mature adult and sex 90 00:04:49,080 --> 00:04:52,460 and ultimately this filtering of GPS locations resulted 91 00:04:52,460 --> 00:04:57,460 in the examination of 74 moose and over 40,000 GPS location. 92 00:05:01,400 --> 00:05:03,760 So the next step is to calculate the home ranges 93 00:05:03,760 --> 00:05:06,990 for each individual across the years and seasons 94 00:05:06,990 --> 00:05:11,520 which we estimated using 95% Colonel home ranges. 95 00:05:11,520 --> 00:05:14,560 we then determined each animal's utilization distribution 96 00:05:14,560 --> 00:05:17,290 or UD within that home range. 97 00:05:17,290 --> 00:05:20,990 So the UD is a probability map that is explaining 98 00:05:20,990 --> 00:05:23,820 where the individual is most likely to occur 99 00:05:23,820 --> 00:05:26,000 within their home range boundary. 100 00:05:26,000 --> 00:05:30,540 So you think of this UD as a topographic map of sorts 101 00:05:30,540 --> 00:05:33,760 where these peaks are identifying areas 102 00:05:33,760 --> 00:05:36,470 of frequently used core areas 103 00:05:36,470 --> 00:05:41,353 while these blue valleys are identifying less used areas. 104 00:05:44,670 --> 00:05:48,400 So the next step is to create 105 00:05:48,400 --> 00:05:51,480 our resource utilization functions for each animal 106 00:05:51,480 --> 00:05:54,020 which in simple terms is really an equation 107 00:05:54,020 --> 00:05:56,690 that will relate the height of the UDs. 108 00:05:56,690 --> 00:05:59,730 So those peaks and valleys to resources 109 00:05:59,730 --> 00:06:01,450 like habitat structure 110 00:06:01,450 --> 00:06:06,400 and land cover characteristics at any given location. 111 00:06:06,400 --> 00:06:07,890 But it's important to note that rather 112 00:06:07,890 --> 00:06:12,790 than just describing the resources at exact GPS locations 113 00:06:12,790 --> 00:06:14,470 we use the focal function in R 114 00:06:14,470 --> 00:06:16,240 and created these mean values 115 00:06:16,240 --> 00:06:19,160 for each resource across two different scales. 116 00:06:19,160 --> 00:06:23,053 So the 400 meter scale another one kilometer scale. 117 00:06:26,040 --> 00:06:29,320 The underlying resources included 118 00:06:29,320 --> 00:06:30,820 that we thought were important 119 00:06:32,073 --> 00:06:34,290 for determining habitat use by moose included 120 00:06:34,290 --> 00:06:39,020 both it's composition and structure of habitats. 121 00:06:39,020 --> 00:06:42,330 So LIDAR variables were describing 122 00:06:42,330 --> 00:06:43,990 that fine scale structure 123 00:06:43,990 --> 00:06:46,140 so the height characteristics of vegetation 124 00:06:46,140 --> 00:06:50,140 across the landscape while the NLCD variables 125 00:06:50,140 --> 00:06:52,270 were describing the overall composition. 126 00:06:52,270 --> 00:06:55,290 So forest types that moose maybe using 127 00:06:55,290 --> 00:06:57,483 during these different time periods. 128 00:06:59,290 --> 00:07:02,090 So what is LIDAR and how can it potentially 129 00:07:02,090 --> 00:07:04,940 improve the mapping of habitat here in Vermont 130 00:07:04,940 --> 00:07:07,720 and particularly for moose purposes? 131 00:07:07,720 --> 00:07:10,740 So LIDAR stands for Light Detection and Ranging 132 00:07:10,740 --> 00:07:13,290 and it's an act of sensor technology. 133 00:07:13,290 --> 00:07:16,010 So this means an airborne sensor directs a beam 134 00:07:16,010 --> 00:07:18,150 of light towards earth surface. 135 00:07:18,150 --> 00:07:21,900 And this light is reflected off of the ground surfaces 136 00:07:21,900 --> 00:07:25,233 or vegetative structure back to the sensor. 137 00:07:27,490 --> 00:07:30,400 So, the results of this is a very dense three-dimensional 138 00:07:30,400 --> 00:07:33,720 point cloud that consists of not only an X Y coordinate 139 00:07:33,720 --> 00:07:36,680 for each point describing where it is in space 140 00:07:36,680 --> 00:07:39,740 but also the Z coordinate that's describing the height 141 00:07:39,740 --> 00:07:43,163 of vegetation on earth surface. 142 00:07:45,490 --> 00:07:48,190 So these point clouds are really cool to look at. 143 00:07:48,190 --> 00:07:51,040 So they have great depth and texture, 144 00:07:51,040 --> 00:07:53,940 and we can use this information to really filter out 145 00:07:53,940 --> 00:07:57,990 and yield really specific components of that cloud. 146 00:07:57,990 --> 00:08:01,410 So this map, for example, is a depiction of the point cloud 147 00:08:01,410 --> 00:08:04,500 in our study area, it's really showing the height 148 00:08:04,500 --> 00:08:05,540 of the vegetation. 149 00:08:05,540 --> 00:08:07,940 So if you look at the yellow and orange color schemes 150 00:08:07,940 --> 00:08:11,640 that's showing trees that are more mature or taller 151 00:08:11,640 --> 00:08:15,080 canopy structures while the blues are really showing those 152 00:08:15,080 --> 00:08:18,350 more shrubby young forest vegetation. 153 00:08:18,350 --> 00:08:21,580 So if we draw a transect across this area, 154 00:08:21,580 --> 00:08:24,710 we can see in the top left-hand corner there, 155 00:08:24,710 --> 00:08:28,550 the returns of the points for that transect 156 00:08:28,550 --> 00:08:30,170 and these points are showing 157 00:08:30,170 --> 00:08:32,580 these more individual mature trees, 158 00:08:32,580 --> 00:08:34,480 the yellow on the far left. 159 00:08:34,480 --> 00:08:37,800 And as we progressed East into that more clear-cut 160 00:08:37,800 --> 00:08:40,310 those points are becoming more concentrated 161 00:08:40,310 --> 00:08:42,130 towards the forest floor. 162 00:08:42,130 --> 00:08:45,220 So this really allows us to filter through 163 00:08:45,220 --> 00:08:48,900 these point clouds to specific height classifications 164 00:08:48,900 --> 00:08:51,270 that may be of importance to moose, 165 00:08:51,270 --> 00:08:54,700 such as a forage classification for example, 166 00:08:54,700 --> 00:08:57,220 which is identifying vegetation that is 167 00:08:57,220 --> 00:08:59,263 less than three meters in height. 168 00:09:01,730 --> 00:09:04,350 So back to our overall methods. 169 00:09:04,350 --> 00:09:06,850 So after we've related the height of the UD 170 00:09:06,850 --> 00:09:09,940 to the underlying lighter and NLCD structure 171 00:09:09,940 --> 00:09:11,830 and composition variables, 172 00:09:11,830 --> 00:09:15,240 we then wanted to examine combinations of these variables 173 00:09:15,240 --> 00:09:18,550 to determine how they may influence the predictability 174 00:09:18,550 --> 00:09:20,583 of habitat use by moose. 175 00:09:23,800 --> 00:09:24,633 To do this 176 00:09:24,633 --> 00:09:28,120 we developed nine resource utilization functions 177 00:09:28,120 --> 00:09:30,407 Four (silence) each moose home range. 178 00:09:34,200 --> 00:09:36,440 So (indistinct) predict the height of the UD 179 00:09:36,440 --> 00:09:40,483 as a function of habitat variables at different scales. 180 00:09:43,370 --> 00:09:46,080 So after we have created our model set for each moose 181 00:09:46,080 --> 00:09:50,130 we repeat this process across all moose examined. 182 00:09:50,130 --> 00:09:53,060 So 74 moos to create this population model 183 00:09:53,060 --> 00:09:56,903 of habitat use by season, sex, and age. 184 00:09:59,020 --> 00:10:01,110 So on to our key results. 185 00:10:01,110 --> 00:10:04,620 This figure is showing the nine RUF models examined 186 00:10:04,620 --> 00:10:08,240 on the Y axis and the average EIC ranking 187 00:10:08,240 --> 00:10:13,240 on the X axis for mature, adult females, young adult females 188 00:10:13,400 --> 00:10:15,193 and young adult males. 189 00:10:16,160 --> 00:10:18,950 The shade of the bars is indicating the seasons 190 00:10:18,950 --> 00:10:21,120 for dormant growth and weights 191 00:10:21,120 --> 00:10:24,070 from a separate color study that we conducted. 192 00:10:24,070 --> 00:10:25,560 We don't have time to really dive 193 00:10:25,560 --> 00:10:28,100 into the color study and what that means. 194 00:10:28,100 --> 00:10:31,020 So please just focus on the overall picture here. 195 00:10:31,020 --> 00:10:34,430 The models with the lowest AIC ranking are the top models 196 00:10:34,430 --> 00:10:36,670 for predicting habitat use. 197 00:10:36,670 --> 00:10:38,830 And as you can see our two top models 198 00:10:38,830 --> 00:10:41,740 across all seasons, sex, and ages 199 00:10:41,740 --> 00:10:44,840 were these two models that were including both NLCD 200 00:10:44,840 --> 00:10:46,250 composition variables 201 00:10:46,250 --> 00:10:49,640 as well as these lighter structure variables. 202 00:10:49,640 --> 00:10:54,363 Our top model was at the one kilometer scale. 203 00:10:57,620 --> 00:10:59,250 The results from the top model 204 00:10:59,250 --> 00:11:02,750 show the relative importance of habitat variables 205 00:11:02,750 --> 00:11:07,750 on patterns of habitat used by moose age, sex, and season. 206 00:11:08,190 --> 00:11:10,360 So you can see from this figure on the X axis 207 00:11:10,360 --> 00:11:13,740 we have the habitat variables included in the top model. 208 00:11:13,740 --> 00:11:14,910 And on the Y axis 209 00:11:14,910 --> 00:11:17,663 we have the average standardized betas. 210 00:11:18,580 --> 00:11:21,300 Again for the sake of the presentation 211 00:11:21,300 --> 00:11:23,220 let's focus on the big picture here. 212 00:11:23,220 --> 00:11:25,770 So the direction of the betas are indicating 213 00:11:25,770 --> 00:11:29,050 whether the variable was used or not used. 214 00:11:29,050 --> 00:11:33,410 And the height of the bar is indicating it's important. 215 00:11:33,410 --> 00:11:36,320 So from this it's clear that moose really are differing 216 00:11:36,320 --> 00:11:38,370 in their use of habitat structure 217 00:11:38,370 --> 00:11:42,480 and composition by sex, season, and age. 218 00:11:42,480 --> 00:11:45,710 However, there are some variables that are important 219 00:11:45,710 --> 00:11:47,370 across all moose 220 00:11:47,370 --> 00:11:51,550 primarily the forage vegetative structure 221 00:11:51,550 --> 00:11:54,260 proved to be important for all moose across 222 00:11:54,260 --> 00:11:56,803 both the dormant season and the growth season. 223 00:12:01,270 --> 00:12:03,540 So the final step was the mapping 224 00:12:03,540 --> 00:12:06,050 of our suitability across our study area. 225 00:12:06,050 --> 00:12:07,620 So we used the coefficients 226 00:12:07,620 --> 00:12:11,350 from our top ranked RUF model to create the suitability 227 00:12:11,350 --> 00:12:14,660 across patches across our landscape. 228 00:12:14,660 --> 00:12:16,410 So I think it's easiest to think 229 00:12:16,410 --> 00:12:18,770 of this as kind of a recipe. 230 00:12:18,770 --> 00:12:22,310 So different combinations of these habitat variables 231 00:12:22,310 --> 00:12:25,110 determine the habitat suitability score 232 00:12:25,110 --> 00:12:28,800 for each 10 meter squared patch on the landscape. 233 00:12:28,800 --> 00:12:31,860 So some patches have just the right ingredients 234 00:12:31,860 --> 00:12:34,430 and indicated by these coefficients 235 00:12:34,430 --> 00:12:38,410 to make highly suitable areas for moose on the landscape. 236 00:12:38,410 --> 00:12:42,100 While other areas may be lacking certain ingredients 237 00:12:42,100 --> 00:12:44,470 which is really dropping those suitability scores 238 00:12:44,470 --> 00:12:45,693 or pulling them down. 239 00:12:48,940 --> 00:12:51,210 So these are two maps that are examples 240 00:12:51,210 --> 00:12:52,600 of habitat suitability 241 00:12:52,600 --> 00:12:55,890 for mature female moose across our study area. 242 00:12:55,890 --> 00:12:59,350 So the map on the right is showing habitat suitability 243 00:12:59,350 --> 00:13:03,890 across our study area, using NLCD composition variables only 244 00:13:03,890 --> 00:13:06,770 while the map on the left is showing the suitability 245 00:13:07,610 --> 00:13:09,870 for our top model, which was incorporating 246 00:13:09,870 --> 00:13:13,250 both structure and composition variables. 247 00:13:13,250 --> 00:13:14,790 And from the shade of these maps, 248 00:13:14,790 --> 00:13:17,840 we can see that if we use the, 249 00:13:17,840 --> 00:13:20,570 just the NLCD composition variables 250 00:13:20,570 --> 00:13:23,700 the overall suitability tends to be a bit higher. 251 00:13:23,700 --> 00:13:26,130 And for the top model on the left, 252 00:13:26,130 --> 00:13:29,460 there appears to be more defined areas 253 00:13:29,460 --> 00:13:34,460 of high suitability and low suitability. 254 00:13:40,010 --> 00:13:43,400 So how can these HSI maps potentially help moose 255 00:13:43,400 --> 00:13:44,710 in the region? 256 00:13:44,710 --> 00:13:46,340 So, as I mentioned in the beginning, 257 00:13:46,340 --> 00:13:48,640 declines and moose health survival 258 00:13:48,640 --> 00:13:51,010 and fecundity due to the impacts of winter ticks 259 00:13:51,010 --> 00:13:53,230 is really threatening their stability 260 00:13:53,230 --> 00:13:57,510 and persistence both here in Vermont, but also regionally. 261 00:13:57,510 --> 00:14:00,830 And the lack of information on the relative suitability 262 00:14:00,830 --> 00:14:04,900 or quality of habitats is really limiting moose management 263 00:14:04,900 --> 00:14:06,700 at a fairly critical time. 264 00:14:06,700 --> 00:14:10,820 As the population may continue to show a future declines. 265 00:14:10,820 --> 00:14:12,430 And although direct approaches 266 00:14:12,430 --> 00:14:14,770 to managing winter tick abundance are limited 267 00:14:14,770 --> 00:14:16,370 at this point in time 268 00:14:16,370 --> 00:14:19,570 the manipulation of existing habitats may ultimately 269 00:14:19,570 --> 00:14:23,960 promote moose population persistence on the landscape. 270 00:14:23,960 --> 00:14:26,040 So generally speaking, our results 271 00:14:26,040 --> 00:14:27,870 from these HSI maps showed 272 00:14:27,870 --> 00:14:31,150 that female moose are actively using areas 273 00:14:31,150 --> 00:14:33,820 with proportionally more regenerating forest 274 00:14:33,820 --> 00:14:36,320 or that forge LIDAR structure 275 00:14:36,320 --> 00:14:38,230 and canopy structure 276 00:14:38,230 --> 00:14:39,690 while males are actively 277 00:14:39,690 --> 00:14:43,310 using higher elevation mixed forest types. 278 00:14:43,310 --> 00:14:46,880 And so these result in maps of habitat suitability 279 00:14:46,880 --> 00:14:49,370 is really providing a means 280 00:14:49,370 --> 00:14:51,900 of informing management activities 281 00:14:51,900 --> 00:14:54,380 such as the restoration or alteration 282 00:14:54,380 --> 00:14:57,670 of habitats to ultimately benefit moose 283 00:14:57,670 --> 00:15:00,025 as well as policies around land use 284 00:15:00,025 --> 00:15:04,663 that may contribute to population recovery. 285 00:15:07,020 --> 00:15:10,300 So our findings from chapter one are very informative 286 00:15:10,300 --> 00:15:12,620 but several questions still remain 287 00:15:12,620 --> 00:15:16,010 such as, how does habitat selection relate 288 00:15:16,010 --> 00:15:18,940 to this issue of winter tick abundance? 289 00:15:18,940 --> 00:15:20,780 So tomorrow I'll be giving a talk 290 00:15:20,780 --> 00:15:23,250 on the second chapter of my thesis 291 00:15:23,250 --> 00:15:24,410 that is really focusing 292 00:15:24,410 --> 00:15:26,650 on patterns of habitat selection 293 00:15:26,650 --> 00:15:29,710 during two critical winter tick life stages 294 00:15:29,710 --> 00:15:34,710 and how those patterns of selection may relate to fitness. 295 00:15:35,710 --> 00:15:40,283 And so I hope to see you there tomorrow afternoon. 296 00:15:42,030 --> 00:15:46,550 With that I would like to end my talk here today. 297 00:15:46,550 --> 00:15:49,800 I just wanna give a shout out to some organizations 298 00:15:49,800 --> 00:15:51,340 that made this study possible 299 00:15:51,340 --> 00:15:54,190 such as the Silvio O. Conte Wildlife Refuge 300 00:15:54,190 --> 00:15:55,890 and all of the kinds staff there 301 00:15:55,890 --> 00:15:58,930 particularly refuge managers, Steve Agius. 302 00:15:58,930 --> 00:16:02,160 I also want to thank VELCO for their continued support 303 00:16:02,160 --> 00:16:04,450 that really made this project possible 304 00:16:04,450 --> 00:16:07,433 especially during the capturing process. 305 00:16:08,390 --> 00:16:10,350 There's a handful of people I wanna quickly thank 306 00:16:10,350 --> 00:16:12,390 particularly my advisors, Jen Murdock 307 00:16:12,390 --> 00:16:15,610 and Terry Donovan and folks from the 308 00:16:15,610 --> 00:16:16,750 Vermont fish and wildlife, 309 00:16:16,750 --> 00:16:19,480 such as Cedric Alexander, Scott darling, 310 00:16:19,480 --> 00:16:22,180 Mark Scott and Tony Smith. 311 00:16:22,180 --> 00:16:25,400 I also would like to thank the UVM spatial analysis lab 312 00:16:25,400 --> 00:16:26,760 and Sean MacFadden 313 00:16:26,760 --> 00:16:29,370 for all of his hard work dealing 314 00:16:29,370 --> 00:16:32,160 with these large LIDAR datasets. 315 00:16:32,160 --> 00:16:33,650 So thank you for tuning in 316 00:16:33,650 --> 00:16:36,930 and I'll be happy to answer any questions folks might have. 317 00:16:36,930 --> 00:16:37,763 Thank you. 318 00:16:39,950 --> 00:16:41,337 - [Chairwoman] Alan Thompson asked, 319 00:16:41,337 --> 00:16:42,927 "How does LIDAR differentiate 320 00:16:42,927 --> 00:16:45,807 "between vegetation and any other substrate?" 321 00:16:49,700 --> 00:16:51,070 - [Joshua] That's a really good question. 322 00:16:51,070 --> 00:16:52,740 And to be perfectly honest 323 00:16:52,740 --> 00:16:57,740 I don't know the exact details of how those pulses 324 00:16:58,200 --> 00:17:02,050 or those reflections from that laser actually differentiates 325 00:17:02,050 --> 00:17:05,600 between vegetation and other substrates. 326 00:17:05,600 --> 00:17:07,020 I know there'll be returns 327 00:17:07,020 --> 00:17:12,020 from LIDAR for example, for manmade structures. 328 00:17:13,240 --> 00:17:15,270 And I think we can differentiate 329 00:17:15,270 --> 00:17:19,010 between those structures and trees, for example, 330 00:17:19,010 --> 00:17:21,760 given the amount of returns 331 00:17:21,760 --> 00:17:25,373 from that actual reflection. 332 00:17:26,990 --> 00:17:29,260 That's a question, I guess maybe 333 00:17:29,260 --> 00:17:31,920 Sean MacFadden in the UVM Spatial Lab 334 00:17:31,920 --> 00:17:32,850 would know a little bit more 335 00:17:32,850 --> 00:17:37,600 about how LIDAR actually can differentiate between those. 336 00:17:37,600 --> 00:17:38,510 Thank you. 337 00:17:38,510 --> 00:17:40,830 - I was wondering, I probably missed the slide 338 00:17:40,830 --> 00:17:45,610 where you showed the spatial configuration 339 00:17:45,610 --> 00:17:47,130 of your captures, 340 00:17:47,130 --> 00:17:49,780 but I was curious how you thought 341 00:17:49,780 --> 00:17:52,070 or if you thought that your selection 342 00:17:52,070 --> 00:17:56,630 of where you chose to capture moose drives the results 343 00:17:56,630 --> 00:17:59,620 of your model in terms of what habitats are available 344 00:17:59,620 --> 00:18:03,370 in the local landscape to those moose that you did capture. 345 00:18:03,370 --> 00:18:05,270 - [Joshua] Yeah, I think that's a great question. 346 00:18:05,270 --> 00:18:08,910 And it's one that is, it's a challenging one to address 347 00:18:08,910 --> 00:18:10,390 because we can only really work 348 00:18:10,390 --> 00:18:14,800 with what the capture crew was was able to do. 349 00:18:14,800 --> 00:18:16,800 So they showed up in January 350 00:18:16,800 --> 00:18:21,640 and over the course of three years, they captured 126 moose. 351 00:18:21,640 --> 00:18:23,807 We told them during that capturing process, 352 00:18:23,807 --> 00:18:26,647 "If you could really try to equally distribute 353 00:18:26,647 --> 00:18:29,797 "where you capture moose, that would be ideal. 354 00:18:29,797 --> 00:18:31,587 "So try to get an equal number 355 00:18:31,587 --> 00:18:36,540 "of moose and E1 wildlife management unit as E2." 356 00:18:36,540 --> 00:18:39,040 But of course there are limitations 357 00:18:39,040 --> 00:18:42,400 and the capture crew had to use logging roads 358 00:18:42,400 --> 00:18:46,440 and openings and just opportunistically color moose 359 00:18:46,440 --> 00:18:47,273 when they could. 360 00:18:47,273 --> 00:18:51,345 But given that, I think given our sample size 361 00:18:51,345 --> 00:18:54,640 and how moose are fairly mobile on the landscape 362 00:18:54,640 --> 00:18:58,340 I think it reflects fairly accurately 363 00:18:58,340 --> 00:19:00,870 their habitat use during those towing periods. 364 00:19:00,870 --> 00:19:03,370 But I think that's a really great question. 365 00:19:03,370 --> 00:19:07,270 - So these weren't just helicoptered moose 366 00:19:07,270 --> 00:19:08,420 - [Joshua] They weren't. 367 00:19:08,420 --> 00:19:09,253 - They were. 368 00:19:09,253 --> 00:19:10,890 - [Joshua] Yeah, so they were, 369 00:19:10,890 --> 00:19:13,370 all of the capturing occurred strictly 370 00:19:13,370 --> 00:19:16,023 with helicopter and hired out crew.