Sabrina E. Russo, Stephen Portnoy, and Carol K. Augspurger. 2006. Incorporating animal behavior into seed dispersal models: implications for seed shadows. Ecology 87:3160–3174.

Appendix E. Summary of parameter estimates, negative log-likelihood score and Akaike’s Information Criterion (AIC) for dispersal kernels fit to the simulated seed dispersal curve for each source tree.

Kernel	Parameter 1	Parameter 2	-L	AIC
Source tree 3
Exponential	0.01	NA	26985	53972
Gaussian	106.51	200.05	31981	63966
Lognormal	3.06	1.87	24321	48646
2Dt	0.26	12.37	23980	47964
Cauchy	5.83	4.45	27175	54354
Weibull	0.54	56.50	24785	49574
Source tree 4
Exponential	0.01	NA	1196	2394
Gaussian	112.31	201.08	1405	2814
Lognormal	3.13	1.86	1080	2164
2Dt	0.26	14.23	1061	2126
Cauchy	5.71	4.51	1206	2416
Weibull	0.54	60.10	1101	2206
Source tree 5
Exponential	0.01	NA	12226	24454
Gaussian	141.51	233.20	14113	28230
Lognormal	3.26	2.09	11134	22272
2Dt	0.21	7.87	11096	22196
Cauchy	5.03	7.07	12931	25866
Weibull	0.52	75.35	11236	22476
Source tree 6
Exponential	0.01	NA	2204	4410
Gaussian	110.85	221.55	2632	5268
Lognormal	3.01	1.99	1974	3952
2Dt	0.22	5.72	1967	3938
Cauchy	4.96	8.82	2288	4580
Weibull	0.52	55.97	2003	4010
Source tree 7
Exponential	0.01	NA	14525	29052
Gaussian	142.65	242.86	16843	33690
Lognormal	3.48	1.92	13520	27044
2Dt	0.22	14.06	13549	27102
Cauchy	15.86	35.19	15569	31142
Weibull	0.56	85.07	13634	27272
Source tree 8
Exponential	0.01	NA	1780	3562
Gaussian	185.64	267.02	2004	4012
Lognormal	3.58	2.29	1667	3338
2Dt	0.16	3.59	1690	3384
Cauchy	62.86	96.37	1945	3894
Weibull	0.53	109.16	1662	3328
Source tree 11
Exponential	0.01	NA	72930	145862
Gaussian	149.96	244.48	83945	167894
Lognormal	3.55	1.90	68065	136134
2Dt	0.23	20.96	68360	136724
Cauchy	11.29	19.52	77394	154792
Weibull	0.57	90.47	68650	137304
Source tree 12
Exponential	0.01	NA	2208	4418
Gaussian	150.88	220.28	2501	5006
Lognormal	3.62	2.06	2112	4228
2Dt	0.18	7.38	2155	4314
Cauchy	49.31	76.82	2417	4838
Weibull	0.59	100.42	2104	4212
Source tree 13
Exponential	0.01	NA	13679	27360
Gaussian	175.49	258.86	15471	30946
Lognormal	3.66	2.19	13002	26008
2Dt	0.16	4.04	13311	26626
Cauchy	67.99	81.90	14769	29542
Weibull	0.56	110.75	12926	25856
Source tree 16
Exponential	0.01	NA	13474	26950
Gaussian	180.41	263.23	15208	30420
Lognormal	3.82	2.05	12949	25902
2Dt	0.17	7.93	13294	26592
Cauchy	53.71	74.69	14538	29080
Weibull	0.59	121.23	12880	25764
Source tree 17
Exponential	0.01	NA	11543	23088
Gaussian	174.05	232.49	12871	25746
Lognormal	4.05	1.83	11379	22762
2Dt	0.20	28.51	11744	23492
Cauchy	90.44	80.60	12395	24794
Weibull	0.68	135.86	11283	22570
Source tree 19
Exponential	0.005	NA	14724	29450
Gaussian	215.805	290.679	16380	32764
Lognormal	4.207	1.821	14378	28760
2Dt	0.23	90.33	14782	29568
Cauchy	64.569	90.097	15888	31780
Weibull	0.656	160.703	14324	28652
All seeds
Exponential	0.01	NA	191929	383860
Gaussian	150.43	238.13	219991	439986
Lognormal	3.56	1.96	180254	360512
2Dt	0.21	14.14	181958	363920
Cauchy	14.58	31.77	206888	413780
Weibull	0.57	92.75	181105	362214

   Notes: Summary of parameter estimates, negative log-likelihood score (-L), and Akaike’s Information Criterion (AIC) for dispersal kernels fit to the simulated seed dispersal curve for each source tree with > 200 seeds and for all seeds from all 19 source trees. Source tree numbers correspond to those in Appendix A. Dispersal kernels for all seeds are plotted in Fig. 5a. Boldface indicates the kernel with the lowest AIC; however, the 8-parameter mixture distribution always had a lower AIC than best-fitting single-distribution dispersal kernel (Appendix F). Parameters one and two (respectively) for each kernel are: exponential, ; Gaussian, and ; Lognormal, and ; 2Dt, p and s; Cauchy, and s; Weibull, p and s (Appendix C).