spAbundance model fitting
39 views
Skip to first unread message
Zachary Spicer
Feb 10, 2025, 3:29:33 PMFeb 10
to spOccupancy and spAbundance users
Hi all,
I am working on multi-species abundance models for 5 plethodontid species in spAbundance. My data set is comprised of 64 sites each with paired independent transects approximately 10m apart (128 total study areas). Each site was visited at least 3 times and around 90 were visited a 4th time in one season. I have 5 focal species, the most abundant of which is over 1,300 indviduals, the next most abundant has 580, the rest range from 135-250 individuals. My study area spans an area approximately 60km x 100km in the blue ridge mountains.
I can get the Negative binomial Nmix model to converge, but I am hoping to be able to use the spatial N mixture model. When I run sfMsNMix using the default parameters laid out in the Nmix vignette the model is far from converging.
I am about to begin the process of model fitting, but I thought it might be a good idea to get some insight on whether this is a fool's errand. I know these complex models are very data hungry. Is there a way to determine if the sfMsNmix model is too complex for my data before I go down the rabbit hole of working on priors and more iterations to get model convergence. I have pasted my sfMsNmix results below if that is helpful.
Thanks,
Zac
Call:
sfMsNMix(abund.formula = abund.formula, det.formula = det.formula,
data = data.Nmix, inits = ms.inits, priors = ms.priors,
tuning = ms.tuning, n.neighbors = 15, n.factors = n.factors,
n.batch = 1600, batch.length = 25, n.omp.threads = 1, verbose = TRUE,
n.report = 400, n.burn = 20000, n.thin = 20, n.chains = 3)
Samples per Chain: 40000
Burn-in: 20000
Thinning Rate: 20
Number of Chains: 3
Total Posterior Samples: 3000
Run Time (min): 33.1083
----------------------------------------
Community Level
----------------------------------------
Abundance Means (log scale):
Mean SD 2.5% 50% 97.5%
(Intercept) 2.2357 0.4664 1.2997 2.2477 3.0591
scale(elev) 1.1569 0.8721 -0.7132 1.1960 2.7778
I(scale(elev)^2) -2.0228 0.6687 -3.1955 -2.0921 -0.5040
scale(NDVI) 0.1743 0.2168 -0.2705 0.1784 0.6117
scale(sol_rad) -0.3076 0.1842 -0.6670 -0.3068 0.0677
scale(distance_to_stream) -0.4764 0.2939 -1.0513 -0.4813 0.1147
scale(elev):scale(distance_to_stream) 0.4572 0.3009 -0.1774 0.4787 0.9974
Rhat ESS
(Intercept) 1.2898 56
scale(elev) 1.0774 1889
I(scale(elev)^2) 1.1241 406
scale(NDVI) 1.1621 337
scale(sol_rad) 1.0042 410
scale(distance_to_stream) 1.2308 166
scale(elev):scale(distance_to_stream) 1.0337 136
Abundance Variances (log scale):
Mean SD 2.5% 50% 97.5%
(Intercept) 0.8106 2.2663 0.0503 0.2995 4.3226
scale(elev) 6.9361 31.7475 0.6292 3.7506 27.5210
I(scale(elev)^2) 2.3157 3.3848 0.3020 1.3785 11.0255
scale(NDVI) 0.2076 0.3428 0.0337 0.1280 0.8570
scale(sol_rad) 0.1569 0.2785 0.0250 0.0955 0.6447
scale(distance_to_stream) 0.3764 1.2891 0.0547 0.2072 1.5229
scale(elev):scale(distance_to_stream) 0.3383 0.6351 0.0342 0.1706 1.6767
Rhat ESS
(Intercept) 2.8545 110
scale(elev) 1.6855 1097
I(scale(elev)^2) 1.0703 2011
scale(NDVI) 1.0497 3000
scale(sol_rad) 1.1115 3000
scale(distance_to_stream) 1.1845 3000
scale(elev):scale(distance_to_stream) 1.2103 282
Detection Means (logit scale):
Mean SD 2.5% 50% 97.5% Rhat ESS
(Intercept) -3.0694 0.5182 -3.9089 -3.1361 -1.8233 1.1351 90
scale(day) 0.2786 0.2271 -0.1546 0.2767 0.7366 1.0021 3000
I(scale(day)^2) -0.2821 0.1665 -0.6250 -0.2782 0.0367 1.0044 2507
scale(atemp) 0.2043 0.2353 -0.2667 0.2044 0.6539 1.0291 1538
scale(gtemp) -0.2200 0.2270 -0.6663 -0.2245 0.2341 1.0060 2830
scale(hum) 0.6656 0.2079 0.2576 0.6596 1.0860 1.0938 1631
scale(sal.precip_24h) -0.1231 0.1703 -0.4613 -0.1209 0.2040 1.0002 2462
scale(sal.precip_3day) -0.0021 0.2026 -0.4090 0.0009 0.4112 1.0138 2312
scale(sal.avg_temp) 0.0928 0.1883 -0.2762 0.0914 0.4686 1.0620 1676
scale(sal.dry_days) -0.1666 0.2028 -0.5796 -0.1632 0.2528 1.0037 2776
Detection Variances (logit scale):
Mean SD 2.5% 50% 97.5% Rhat ESS
(Intercept) 1.0038 1.8353 0.0760 0.5278 5.1931 1.0565 655
scale(day) 0.2727 0.4058 0.0439 0.1620 1.1514 1.0340 3000
I(scale(day)^2) 0.1196 0.1803 0.0214 0.0731 0.4995 1.0760 2591
scale(atemp) 0.2096 0.3380 0.0317 0.1228 0.8643 1.0195 2841
scale(gtemp) 0.2395 0.3504 0.0353 0.1465 1.0140 1.0249 3000
scale(hum) 0.2094 0.4421 0.0315 0.1186 0.8248 1.0497 3000
scale(sal.precip_24h) 0.1202 0.1825 0.0218 0.0751 0.5028 1.0103 3000
scale(sal.precip_3day) 0.1695 0.2385 0.0277 0.1022 0.7326 1.0221 3000
scale(sal.avg_temp) 0.1512 0.2566 0.0251 0.0879 0.6318 1.0795 3000
scale(sal.dry_days) 0.1893 0.2714 0.0309 0.1149 0.7876 1.0090 3000
----------------------------------------
Species Level
----------------------------------------
Abundance (log scale):
Mean SD 2.5% 50%
(Intercept)-cyl 2.7520 0.7872 1.7474 2.5440
(Intercept)-mon 2.5644 0.5138 1.4730 2.7389
(Intercept)-ore 2.2067 0.3308 1.4622 2.2515
(Intercept)-wil 1.9682 0.5110 1.0095 1.9149
(Intercept)-yon 2.2842 0.4629 1.3544 2.2946
scale(elev)-cyl -0.5582 0.5634 -1.6120 -0.6146
scale(elev)-mon 3.7423 1.1682 2.2286 3.3064
scale(elev)-ore 3.0557 0.4530 2.0734 3.0542
scale(elev)-wil 0.7029 0.3779 -0.0366 0.7013
scale(elev)-yon 0.7645 0.3936 0.0116 0.7453
I(scale(elev)^2)-cyl -3.4200 0.5355 -4.6695 -3.3846
I(scale(elev)^2)-mon -1.9940 0.6129 -3.1687 -1.8278
I(scale(elev)^2)-ore -1.4842 0.3059 -1.9437 -1.5160
I(scale(elev)^2)-wil -1.2739 0.3791 -2.1047 -1.2536
I(scale(elev)^2)-yon -3.3259 0.4879 -4.2719 -3.3144
scale(NDVI)-cyl 0.2027 0.1617 -0.1502 0.2074
scale(NDVI)-mon -0.0652 0.1351 -0.3096 -0.0696
scale(NDVI)-ore 0.3104 0.1909 -0.1251 0.3122
scale(NDVI)-wil 0.0151 0.2207 -0.4478 0.0258
scale(NDVI)-yon 0.4244 0.1364 0.1737 0.4232
scale(sol_rad)-cyl -0.3300 0.1071 -0.5426 -0.3317
scale(sol_rad)-mon -0.1709 0.1513 -0.4532 -0.1599
scale(sol_rad)-ore -0.3191 0.1820 -0.6749 -0.3200
scale(sol_rad)-wil -0.2175 0.1747 -0.5540 -0.2189
scale(sol_rad)-yon -0.4947 0.1230 -0.7265 -0.4955
scale(distance_to_stream)-cyl -0.4561 0.1706 -0.7819 -0.4561
scale(distance_to_stream)-mon -0.6901 0.2946 -1.1937 -0.6813
scale(distance_to_stream)-ore -0.7399 0.3214 -1.3609 -0.7120
scale(distance_to_stream)-wil -0.5604 0.2737 -1.1234 -0.5657
scale(distance_to_stream)-yon 0.0020 0.1774 -0.3384 -0.0005
scale(elev):scale(distance_to_stream)-cyl 0.7750 0.3612 0.1641 0.7441
scale(elev):scale(distance_to_stream)-mon 0.4700 0.2176 0.0375 0.4941
scale(elev):scale(distance_to_stream)-ore 0.4659 0.3803 -0.5585 0.5011
scale(elev):scale(distance_to_stream)-wil 0.2405 0.3630 -0.5774 0.2732
scale(elev):scale(distance_to_stream)-yon 0.3881 0.3107 -0.2614 0.4017
97.5% Rhat ESS
(Intercept)-cyl 4.6624 5.3984 7
(Intercept)-mon 3.2500 2.9131 5
(Intercept)-ore 2.7427 1.6625 17
(Intercept)-wil 3.1767 1.8864 29
(Intercept)-yon 3.2285 1.0836 40
scale(elev)-cyl 0.5055 4.5339 23
scale(elev)-mon 5.6788 13.1529 2
scale(elev)-ore 3.9103 3.7523 12
scale(elev)-wil 1.4233 1.5305 48
scale(elev)-yon 1.5401 1.4529 175
I(scale(elev)^2)-cyl -2.4651 1.9087 49
I(scale(elev)^2)-mon -1.1626 5.5500 2
I(scale(elev)^2)-ore -0.7170 1.1363 10
I(scale(elev)^2)-wil -0.6160 1.3861 66
I(scale(elev)^2)-yon -2.4489 1.0880 312
scale(NDVI)-cyl 0.4961 1.3512 78
scale(NDVI)-mon 0.2146 2.0668 29
scale(NDVI)-ore 0.6663 1.2783 42
scale(NDVI)-wil 0.4010 1.3030 83
scale(NDVI)-yon 0.7023 1.1812 190
scale(sol_rad)-cyl -0.1091 1.1798 173
scale(sol_rad)-mon 0.1470 1.6420 22
scale(sol_rad)-ore 0.0234 1.5342 69
scale(sol_rad)-wil 0.1391 1.2850 90
scale(sol_rad)-yon -0.2326 1.3280 112
scale(distance_to_stream)-cyl -0.1240 1.2235 78
scale(distance_to_stream)-mon -0.0301 2.6950 8
scale(distance_to_stream)-ore -0.1743 2.2875 13
scale(distance_to_stream)-wil -0.0097 1.0321 60
scale(distance_to_stream)-yon 0.3684 1.2429 92
scale(elev):scale(distance_to_stream)-cyl 1.5721 1.3447 61
scale(elev):scale(distance_to_stream)-mon 0.8645 1.0158 21
scale(elev):scale(distance_to_stream)-ore 1.0578 1.6840 8
scale(elev):scale(distance_to_stream)-wil 0.8853 1.2935 66
scale(elev):scale(distance_to_stream)-yon 0.9791 1.0272 234
Detection (logit scale):
Mean SD 2.5% 50% 97.5% Rhat ESS
(Intercept)-cyl -2.9532 0.8317 -4.9480 -2.7589 -1.8511 4.6203 8
(Intercept)-mon -3.5718 0.3818 -4.2885 -3.4990 -2.8809 3.6977 4
(Intercept)-ore -3.6660 0.4323 -4.5719 -3.6285 -2.9487 4.9604 8
(Intercept)-wil -3.4031 0.4829 -4.4107 -3.3917 -2.4732 1.5670 27
(Intercept)-yon -2.7512 0.4892 -3.7183 -2.7613 -1.7965 1.0816 43
scale(day)-cyl 0.6695 0.1598 0.3626 0.6647 0.9982 1.0025 1939
scale(day)-mon -0.0795 0.0483 -0.1689 -0.0801 0.0178 1.0039 1543
scale(day)-ore 0.1517 0.0850 -0.0087 0.1519 0.3165 0.9999 2157
scale(day)-wil 0.4861 0.1445 0.2084 0.4854 0.7716 1.0009 1261
scale(day)-yon 0.1874 0.1467 -0.1067 0.1888 0.4791 1.0542 1779
I(scale(day)^2)-cyl -0.2497 0.1176 -0.4880 -0.2473 -0.0268 1.0491 564
I(scale(day)^2)-mon -0.2494 0.0625 -0.3715 -0.2496 -0.1260 1.0198 551
I(scale(day)^2)-ore -0.2447 0.0991 -0.4419 -0.2444 -0.0537 1.0188 604
I(scale(day)^2)-wil -0.3139 0.1420 -0.5996 -0.3111 -0.0344 1.0040 748
I(scale(day)^2)-yon -0.3611 0.1432 -0.6556 -0.3567 -0.0898 1.1082 597
scale(atemp)-cyl 0.0089 0.2423 -0.4780 0.0096 0.4647 1.1158 694
scale(atemp)-mon 0.0781 0.1058 -0.1228 0.0799 0.2888 1.0457 252
scale(atemp)-ore 0.4068 0.1830 0.0513 0.4032 0.7765 1.1798 391
scale(atemp)-wil 0.3119 0.2513 -0.1590 0.3046 0.8316 1.0070 699
scale(atemp)-yon 0.2555 0.2504 -0.2373 0.2539 0.7469 1.0174 1708
scale(gtemp)-cyl -0.0693 0.1781 -0.4077 -0.0729 0.2961 1.0014 708
scale(gtemp)-mon -0.0726 0.0787 -0.2275 -0.0730 0.0789 1.0413 955
scale(gtemp)-ore -0.0815 0.1449 -0.3621 -0.0824 0.2125 1.1765 676
scale(gtemp)-wil -0.6422 0.2150 -1.0809 -0.6382 -0.2533 1.1419 764
scale(gtemp)-yon -0.2835 0.1853 -0.6444 -0.2836 0.0795 1.0057 2380
scale(hum)-cyl 0.8939 0.1867 0.5364 0.8937 1.2527 2.3278 36
scale(hum)-mon 0.3837 0.0872 0.2163 0.3826 0.5539 1.0928 448
scale(hum)-ore 0.6918 0.1119 0.4833 0.6892 0.9208 1.1341 451
scale(hum)-wil 0.7927 0.1924 0.4338 0.7833 1.2045 1.0391 628
scale(hum)-yon 0.5904 0.1708 0.2621 0.5882 0.9273 1.0600 790
scale(sal.precip_24h)-cyl -0.1740 0.1028 -0.3773 -0.1746 0.0224 1.0605 790
scale(sal.precip_24h)-mon -0.1094 0.0951 -0.2939 -0.1093 0.0786 1.0084 461
scale(sal.precip_24h)-ore -0.1339 0.1494 -0.4285 -0.1322 0.1565 1.2893 325
scale(sal.precip_24h)-wil -0.0917 0.1563 -0.3989 -0.0927 0.2197 1.0172 1113
scale(sal.precip_24h)-yon -0.1212 0.1508 -0.4236 -0.1208 0.1813 1.0069 1555
scale(sal.precip_3day)-cyl -0.0265 0.1353 -0.2893 -0.0265 0.2445 1.0133 833
scale(sal.precip_3day)-mon 0.0564 0.0948 -0.1257 0.0568 0.2458 1.0789 371
scale(sal.precip_3day)-ore 0.2050 0.1141 -0.0157 0.2021 0.4345 0.9999 602
scale(sal.precip_3day)-wil -0.0614 0.1971 -0.4725 -0.0603 0.3164 1.0411 765
scale(sal.precip_3day)-yon -0.1807 0.2022 -0.6036 -0.1737 0.1937 1.0419 1530
scale(sal.avg_temp)-cyl 0.0659 0.1866 -0.2935 0.0645 0.4419 1.3567 252
scale(sal.avg_temp)-mon 0.0636 0.0969 -0.1313 0.0637 0.2515 1.0253 573
scale(sal.avg_temp)-ore 0.0005 0.1237 -0.2367 -0.0002 0.2359 1.0161 573
scale(sal.avg_temp)-wil 0.2319 0.2112 -0.1573 0.2257 0.6895 1.0861 505
scale(sal.avg_temp)-yon 0.1089 0.1952 -0.2668 0.1078 0.5121 1.0321 984
scale(sal.dry_days)-cyl -0.4136 0.1581 -0.7265 -0.4122 -0.1074 1.0429 1259
scale(sal.dry_days)-mon -0.2696 0.0785 -0.4281 -0.2678 -0.1219 1.0497 779
scale(sal.dry_days)-ore 0.0403 0.1071 -0.1734 0.0420 0.2459 1.0490 1428
scale(sal.dry_days)-wil -0.1746 0.1864 -0.5429 -0.1738 0.1841 1.0629 935
scale(sal.dry_days)-yon -0.0361 0.2101 -0.4581 -0.0351 0.3786 1.0104 1722
----------------------------------------
Spatial Covariance
----------------------------------------
Mean SD 2.5% 50% 97.5% Rhat ESS
phi-1 0.0024 0.0032 2e-04 0.0006 0.0111 16.5100 7
phi-2 0.0019 0.0020 1e-04 0.0012 0.0069 7.8297 9
phi-3 0.0046 0.0050 5e-04 0.0030 0.0176 3.0529 62
I am working on multi-species abundance models for 5 plethodontid species in spAbundance. My data set is comprised of 64 sites each with paired independent transects approximately 10m apart (128 total study areas). Each site was visited at least 3 times and around 90 were visited a 4th time in one season. I have 5 focal species, the most abundant of which is over 1,300 indviduals, the next most abundant has 580, the rest range from 135-250 individuals. My study area spans an area approximately 60km x 100km in the blue ridge mountains.
I can get the Negative binomial Nmix model to converge, but I am hoping to be able to use the spatial N mixture model. When I run sfMsNMix using the default parameters laid out in the Nmix vignette the model is far from converging.
I am about to begin the process of model fitting, but I thought it might be a good idea to get some insight on whether this is a fool's errand. I know these complex models are very data hungry. Is there a way to determine if the sfMsNmix model is too complex for my data before I go down the rabbit hole of working on priors and more iterations to get model convergence. I have pasted my sfMsNmix results below if that is helpful.
Thanks,
Zac
Call:
sfMsNMix(abund.formula = abund.formula, det.formula = det.formula,
data = data.Nmix, inits = ms.inits, priors = ms.priors,
tuning = ms.tuning, n.neighbors = 15, n.factors = n.factors,
n.batch = 1600, batch.length = 25, n.omp.threads = 1, verbose = TRUE,
n.report = 400, n.burn = 20000, n.thin = 20, n.chains = 3)
Samples per Chain: 40000
Burn-in: 20000
Thinning Rate: 20
Number of Chains: 3
Total Posterior Samples: 3000
Run Time (min): 33.1083
----------------------------------------
Community Level
----------------------------------------
Abundance Means (log scale):
Mean SD 2.5% 50% 97.5%
(Intercept) 2.2357 0.4664 1.2997 2.2477 3.0591
scale(elev) 1.1569 0.8721 -0.7132 1.1960 2.7778
I(scale(elev)^2) -2.0228 0.6687 -3.1955 -2.0921 -0.5040
scale(NDVI) 0.1743 0.2168 -0.2705 0.1784 0.6117
scale(sol_rad) -0.3076 0.1842 -0.6670 -0.3068 0.0677
scale(distance_to_stream) -0.4764 0.2939 -1.0513 -0.4813 0.1147
scale(elev):scale(distance_to_stream) 0.4572 0.3009 -0.1774 0.4787 0.9974
Rhat ESS
(Intercept) 1.2898 56
scale(elev) 1.0774 1889
I(scale(elev)^2) 1.1241 406
scale(NDVI) 1.1621 337
scale(sol_rad) 1.0042 410
scale(distance_to_stream) 1.2308 166
scale(elev):scale(distance_to_stream) 1.0337 136
Abundance Variances (log scale):
Mean SD 2.5% 50% 97.5%
(Intercept) 0.8106 2.2663 0.0503 0.2995 4.3226
scale(elev) 6.9361 31.7475 0.6292 3.7506 27.5210
I(scale(elev)^2) 2.3157 3.3848 0.3020 1.3785 11.0255
scale(NDVI) 0.2076 0.3428 0.0337 0.1280 0.8570
scale(sol_rad) 0.1569 0.2785 0.0250 0.0955 0.6447
scale(distance_to_stream) 0.3764 1.2891 0.0547 0.2072 1.5229
scale(elev):scale(distance_to_stream) 0.3383 0.6351 0.0342 0.1706 1.6767
Rhat ESS
(Intercept) 2.8545 110
scale(elev) 1.6855 1097
I(scale(elev)^2) 1.0703 2011
scale(NDVI) 1.0497 3000
scale(sol_rad) 1.1115 3000
scale(distance_to_stream) 1.1845 3000
scale(elev):scale(distance_to_stream) 1.2103 282
Detection Means (logit scale):
Mean SD 2.5% 50% 97.5% Rhat ESS
(Intercept) -3.0694 0.5182 -3.9089 -3.1361 -1.8233 1.1351 90
scale(day) 0.2786 0.2271 -0.1546 0.2767 0.7366 1.0021 3000
I(scale(day)^2) -0.2821 0.1665 -0.6250 -0.2782 0.0367 1.0044 2507
scale(atemp) 0.2043 0.2353 -0.2667 0.2044 0.6539 1.0291 1538
scale(gtemp) -0.2200 0.2270 -0.6663 -0.2245 0.2341 1.0060 2830
scale(hum) 0.6656 0.2079 0.2576 0.6596 1.0860 1.0938 1631
scale(sal.precip_24h) -0.1231 0.1703 -0.4613 -0.1209 0.2040 1.0002 2462
scale(sal.precip_3day) -0.0021 0.2026 -0.4090 0.0009 0.4112 1.0138 2312
scale(sal.avg_temp) 0.0928 0.1883 -0.2762 0.0914 0.4686 1.0620 1676
scale(sal.dry_days) -0.1666 0.2028 -0.5796 -0.1632 0.2528 1.0037 2776
Detection Variances (logit scale):
Mean SD 2.5% 50% 97.5% Rhat ESS
(Intercept) 1.0038 1.8353 0.0760 0.5278 5.1931 1.0565 655
scale(day) 0.2727 0.4058 0.0439 0.1620 1.1514 1.0340 3000
I(scale(day)^2) 0.1196 0.1803 0.0214 0.0731 0.4995 1.0760 2591
scale(atemp) 0.2096 0.3380 0.0317 0.1228 0.8643 1.0195 2841
scale(gtemp) 0.2395 0.3504 0.0353 0.1465 1.0140 1.0249 3000
scale(hum) 0.2094 0.4421 0.0315 0.1186 0.8248 1.0497 3000
scale(sal.precip_24h) 0.1202 0.1825 0.0218 0.0751 0.5028 1.0103 3000
scale(sal.precip_3day) 0.1695 0.2385 0.0277 0.1022 0.7326 1.0221 3000
scale(sal.avg_temp) 0.1512 0.2566 0.0251 0.0879 0.6318 1.0795 3000
scale(sal.dry_days) 0.1893 0.2714 0.0309 0.1149 0.7876 1.0090 3000
----------------------------------------
Species Level
----------------------------------------
Abundance (log scale):
Mean SD 2.5% 50%
(Intercept)-cyl 2.7520 0.7872 1.7474 2.5440
(Intercept)-mon 2.5644 0.5138 1.4730 2.7389
(Intercept)-ore 2.2067 0.3308 1.4622 2.2515
(Intercept)-wil 1.9682 0.5110 1.0095 1.9149
(Intercept)-yon 2.2842 0.4629 1.3544 2.2946
scale(elev)-cyl -0.5582 0.5634 -1.6120 -0.6146
scale(elev)-mon 3.7423 1.1682 2.2286 3.3064
scale(elev)-ore 3.0557 0.4530 2.0734 3.0542
scale(elev)-wil 0.7029 0.3779 -0.0366 0.7013
scale(elev)-yon 0.7645 0.3936 0.0116 0.7453
I(scale(elev)^2)-cyl -3.4200 0.5355 -4.6695 -3.3846
I(scale(elev)^2)-mon -1.9940 0.6129 -3.1687 -1.8278
I(scale(elev)^2)-ore -1.4842 0.3059 -1.9437 -1.5160
I(scale(elev)^2)-wil -1.2739 0.3791 -2.1047 -1.2536
I(scale(elev)^2)-yon -3.3259 0.4879 -4.2719 -3.3144
scale(NDVI)-cyl 0.2027 0.1617 -0.1502 0.2074
scale(NDVI)-mon -0.0652 0.1351 -0.3096 -0.0696
scale(NDVI)-ore 0.3104 0.1909 -0.1251 0.3122
scale(NDVI)-wil 0.0151 0.2207 -0.4478 0.0258
scale(NDVI)-yon 0.4244 0.1364 0.1737 0.4232
scale(sol_rad)-cyl -0.3300 0.1071 -0.5426 -0.3317
scale(sol_rad)-mon -0.1709 0.1513 -0.4532 -0.1599
scale(sol_rad)-ore -0.3191 0.1820 -0.6749 -0.3200
scale(sol_rad)-wil -0.2175 0.1747 -0.5540 -0.2189
scale(sol_rad)-yon -0.4947 0.1230 -0.7265 -0.4955
scale(distance_to_stream)-cyl -0.4561 0.1706 -0.7819 -0.4561
scale(distance_to_stream)-mon -0.6901 0.2946 -1.1937 -0.6813
scale(distance_to_stream)-ore -0.7399 0.3214 -1.3609 -0.7120
scale(distance_to_stream)-wil -0.5604 0.2737 -1.1234 -0.5657
scale(distance_to_stream)-yon 0.0020 0.1774 -0.3384 -0.0005
scale(elev):scale(distance_to_stream)-cyl 0.7750 0.3612 0.1641 0.7441
scale(elev):scale(distance_to_stream)-mon 0.4700 0.2176 0.0375 0.4941
scale(elev):scale(distance_to_stream)-ore 0.4659 0.3803 -0.5585 0.5011
scale(elev):scale(distance_to_stream)-wil 0.2405 0.3630 -0.5774 0.2732
scale(elev):scale(distance_to_stream)-yon 0.3881 0.3107 -0.2614 0.4017
97.5% Rhat ESS
(Intercept)-cyl 4.6624 5.3984 7
(Intercept)-mon 3.2500 2.9131 5
(Intercept)-ore 2.7427 1.6625 17
(Intercept)-wil 3.1767 1.8864 29
(Intercept)-yon 3.2285 1.0836 40
scale(elev)-cyl 0.5055 4.5339 23
scale(elev)-mon 5.6788 13.1529 2
scale(elev)-ore 3.9103 3.7523 12
scale(elev)-wil 1.4233 1.5305 48
scale(elev)-yon 1.5401 1.4529 175
I(scale(elev)^2)-cyl -2.4651 1.9087 49
I(scale(elev)^2)-mon -1.1626 5.5500 2
I(scale(elev)^2)-ore -0.7170 1.1363 10
I(scale(elev)^2)-wil -0.6160 1.3861 66
I(scale(elev)^2)-yon -2.4489 1.0880 312
scale(NDVI)-cyl 0.4961 1.3512 78
scale(NDVI)-mon 0.2146 2.0668 29
scale(NDVI)-ore 0.6663 1.2783 42
scale(NDVI)-wil 0.4010 1.3030 83
scale(NDVI)-yon 0.7023 1.1812 190
scale(sol_rad)-cyl -0.1091 1.1798 173
scale(sol_rad)-mon 0.1470 1.6420 22
scale(sol_rad)-ore 0.0234 1.5342 69
scale(sol_rad)-wil 0.1391 1.2850 90
scale(sol_rad)-yon -0.2326 1.3280 112
scale(distance_to_stream)-cyl -0.1240 1.2235 78
scale(distance_to_stream)-mon -0.0301 2.6950 8
scale(distance_to_stream)-ore -0.1743 2.2875 13
scale(distance_to_stream)-wil -0.0097 1.0321 60
scale(distance_to_stream)-yon 0.3684 1.2429 92
scale(elev):scale(distance_to_stream)-cyl 1.5721 1.3447 61
scale(elev):scale(distance_to_stream)-mon 0.8645 1.0158 21
scale(elev):scale(distance_to_stream)-ore 1.0578 1.6840 8
scale(elev):scale(distance_to_stream)-wil 0.8853 1.2935 66
scale(elev):scale(distance_to_stream)-yon 0.9791 1.0272 234
Detection (logit scale):
Mean SD 2.5% 50% 97.5% Rhat ESS
(Intercept)-cyl -2.9532 0.8317 -4.9480 -2.7589 -1.8511 4.6203 8
(Intercept)-mon -3.5718 0.3818 -4.2885 -3.4990 -2.8809 3.6977 4
(Intercept)-ore -3.6660 0.4323 -4.5719 -3.6285 -2.9487 4.9604 8
(Intercept)-wil -3.4031 0.4829 -4.4107 -3.3917 -2.4732 1.5670 27
(Intercept)-yon -2.7512 0.4892 -3.7183 -2.7613 -1.7965 1.0816 43
scale(day)-cyl 0.6695 0.1598 0.3626 0.6647 0.9982 1.0025 1939
scale(day)-mon -0.0795 0.0483 -0.1689 -0.0801 0.0178 1.0039 1543
scale(day)-ore 0.1517 0.0850 -0.0087 0.1519 0.3165 0.9999 2157
scale(day)-wil 0.4861 0.1445 0.2084 0.4854 0.7716 1.0009 1261
scale(day)-yon 0.1874 0.1467 -0.1067 0.1888 0.4791 1.0542 1779
I(scale(day)^2)-cyl -0.2497 0.1176 -0.4880 -0.2473 -0.0268 1.0491 564
I(scale(day)^2)-mon -0.2494 0.0625 -0.3715 -0.2496 -0.1260 1.0198 551
I(scale(day)^2)-ore -0.2447 0.0991 -0.4419 -0.2444 -0.0537 1.0188 604
I(scale(day)^2)-wil -0.3139 0.1420 -0.5996 -0.3111 -0.0344 1.0040 748
I(scale(day)^2)-yon -0.3611 0.1432 -0.6556 -0.3567 -0.0898 1.1082 597
scale(atemp)-cyl 0.0089 0.2423 -0.4780 0.0096 0.4647 1.1158 694
scale(atemp)-mon 0.0781 0.1058 -0.1228 0.0799 0.2888 1.0457 252
scale(atemp)-ore 0.4068 0.1830 0.0513 0.4032 0.7765 1.1798 391
scale(atemp)-wil 0.3119 0.2513 -0.1590 0.3046 0.8316 1.0070 699
scale(atemp)-yon 0.2555 0.2504 -0.2373 0.2539 0.7469 1.0174 1708
scale(gtemp)-cyl -0.0693 0.1781 -0.4077 -0.0729 0.2961 1.0014 708
scale(gtemp)-mon -0.0726 0.0787 -0.2275 -0.0730 0.0789 1.0413 955
scale(gtemp)-ore -0.0815 0.1449 -0.3621 -0.0824 0.2125 1.1765 676
scale(gtemp)-wil -0.6422 0.2150 -1.0809 -0.6382 -0.2533 1.1419 764
scale(gtemp)-yon -0.2835 0.1853 -0.6444 -0.2836 0.0795 1.0057 2380
scale(hum)-cyl 0.8939 0.1867 0.5364 0.8937 1.2527 2.3278 36
scale(hum)-mon 0.3837 0.0872 0.2163 0.3826 0.5539 1.0928 448
scale(hum)-ore 0.6918 0.1119 0.4833 0.6892 0.9208 1.1341 451
scale(hum)-wil 0.7927 0.1924 0.4338 0.7833 1.2045 1.0391 628
scale(hum)-yon 0.5904 0.1708 0.2621 0.5882 0.9273 1.0600 790
scale(sal.precip_24h)-cyl -0.1740 0.1028 -0.3773 -0.1746 0.0224 1.0605 790
scale(sal.precip_24h)-mon -0.1094 0.0951 -0.2939 -0.1093 0.0786 1.0084 461
scale(sal.precip_24h)-ore -0.1339 0.1494 -0.4285 -0.1322 0.1565 1.2893 325
scale(sal.precip_24h)-wil -0.0917 0.1563 -0.3989 -0.0927 0.2197 1.0172 1113
scale(sal.precip_24h)-yon -0.1212 0.1508 -0.4236 -0.1208 0.1813 1.0069 1555
scale(sal.precip_3day)-cyl -0.0265 0.1353 -0.2893 -0.0265 0.2445 1.0133 833
scale(sal.precip_3day)-mon 0.0564 0.0948 -0.1257 0.0568 0.2458 1.0789 371
scale(sal.precip_3day)-ore 0.2050 0.1141 -0.0157 0.2021 0.4345 0.9999 602
scale(sal.precip_3day)-wil -0.0614 0.1971 -0.4725 -0.0603 0.3164 1.0411 765
scale(sal.precip_3day)-yon -0.1807 0.2022 -0.6036 -0.1737 0.1937 1.0419 1530
scale(sal.avg_temp)-cyl 0.0659 0.1866 -0.2935 0.0645 0.4419 1.3567 252
scale(sal.avg_temp)-mon 0.0636 0.0969 -0.1313 0.0637 0.2515 1.0253 573
scale(sal.avg_temp)-ore 0.0005 0.1237 -0.2367 -0.0002 0.2359 1.0161 573
scale(sal.avg_temp)-wil 0.2319 0.2112 -0.1573 0.2257 0.6895 1.0861 505
scale(sal.avg_temp)-yon 0.1089 0.1952 -0.2668 0.1078 0.5121 1.0321 984
scale(sal.dry_days)-cyl -0.4136 0.1581 -0.7265 -0.4122 -0.1074 1.0429 1259
scale(sal.dry_days)-mon -0.2696 0.0785 -0.4281 -0.2678 -0.1219 1.0497 779
scale(sal.dry_days)-ore 0.0403 0.1071 -0.1734 0.0420 0.2459 1.0490 1428
scale(sal.dry_days)-wil -0.1746 0.1864 -0.5429 -0.1738 0.1841 1.0629 935
scale(sal.dry_days)-yon -0.0361 0.2101 -0.4581 -0.0351 0.3786 1.0104 1722
----------------------------------------
Spatial Covariance
----------------------------------------
Mean SD 2.5% 50% 97.5% Rhat ESS
phi-1 0.0024 0.0032 2e-04 0.0006 0.0111 16.5100 7
phi-2 0.0019 0.0020 1e-04 0.0012 0.0069 7.8297 9
phi-3 0.0046 0.0050 5e-04 0.0030 0.0176 3.0529 62
Jeffrey Doser
Feb 11, 2025, 4:55:56 PMFeb 11
to spOccupancy and spAbundance users
Hi Zac,
The spatial N-mixture models can be pretty tricky to get to converge, especially the multi-species ones that, as you mention, are quite complex and data hungry. The multi-species ones in particular may require some fine-tuning of the priors and in particular the initial values to get things to converge. 64 sites is a fairly modest number of locations for fitting spatial models, so it may be difficult to get them to converge and/or there may be large amounts of uncertainty in the estimates. What I would suggest doing is first trying to get some single-species spatial N-mixture models to fit. Those will run faster than the multi-species model, and will give you a sense of whether it's feasible given your data set. You very well may have to do fine-tuning of the priors and/or initial values in order to get those to converge, but it's probably easier to test around with it using a single-species model relative to a multi-species model. This vignette (even though it's in the context of occupancy models in spOccupancy) has some suggestions on things you can try to play around with. You do have a good amount of replication and it doesn't sound like you have any extremely rare species, so both of those bode well for trying to fit a spatial model.
Jeff
Reply all
Reply to author
Forward
0 new messages