Figures and data in Assessing plant phenological changes based on drivers of spring phenology

Figures
Tables
Additional files

3 figures, 2 tables and 2 additional files

Figures

Figure 1

Download asset Open asset

Observed responses and phenological lags (least square means ± standard errors) in leafing and flowering (left and right set of bars in each pair, respectively) by research approach (observational and experimental), species origin (native and exotic), climatic region (boreal and temperate), and growth form (tree, shrub, herb, and grass) extracted from reported plant phenological changes in spring.

Phenological lags are calculated from the differences between observed responses and those expected from species-specific changes in spring temperatures. Different letters indicate means that differ significantly (P<0.05).

Figure 2

Download asset Open asset

A diagram showing the relationships among forcing change ( $F_{C}$ , difference in spring forcing (degree days) between baseline $\sum_{1}^{i} T_{C i}$ and warmer $\sum_{1}^{i} T_{W i}$ climates at budburst with the baseline climate O_Ci), expected response (N_E, difference between baseline and warmer climates in reaching species forcing threshold, i.e., $\sum_{1}^{i} T_{C i}$ at O_Ci), budburst temperature ( $F_{C}$ /N_E, average temperature or rate of forcing accumulation at budburst with the warmer climate), and phenological lag (N_C, difference between expected N_E and observed N_O responses or between expected E_Wi and observed O_Wi phenology with the warmer climate).

Figure 3

Download asset Open asset

Distribution of phenological studies used in this synthesis.

Numbers correspond to study information in Supplementary file 1A. For studies reporting average phenological responses for multiple locations, geographic centers are shown.

Tables

Table 1

Mean values of climatic, phenological, and biological variables by budburst (leafing or flowering), research approach (observational or experimental), species origin (native or exotic), climatic region (boreal or temperate), and growth form (tree, shrub, herb, or grass).

Budburst temperature = average temperature at budburst under the warmer climate, forcing change = warming-induced changes in spring forcing prior to budburst, spring warming = average spring temperature change, and phenological lag = difference between observed response and that expected from forcing change and budburst temperature.

Budburst	Spring phenology (Julian)	Budburst temperature (°C)	Altitude(m)	Latitude (°N)	MAT (°C)	Forcing change (degree days)	Spring warming (°C)	Phenological lag (day)	Sample size(N)
Leafing	111	11.1	687	42.8	7.8	85	1.3	1.8	447
Observational	106	11.0	487	42.6	8.4	79	1.1	1.2	359
Experimental	130	11.2	1500	43.8	5.6	106	1.8	4.1	88
Native	113	11.0	744	43.3	7.4	85	1.3	1.9	379
Exotic	100	11.7	366	40.0	10.2	78	1.2	1.1	68
Boreal	128	10.2	1147	45.7	2.0	65	1.3	2.4	154
Temperate	102	11.5	444	41.3	10.9	94	1.3	1.5	293
Tree	105	11.7	364	41.9	10.1	101	1.4	1.9	246
Shrub	109	10.9	433	43.5	7.8	80	1.4	1.1	107
Herb	131	10.0	1955	43.3	1.8	51	0.8	1.2	54
Grass	123	8.9	1637	45.9	2.1	37	0.7	3.6	40
Flowering	138	13.7	534	45.5	9.0	126	1.3	3.5	1080
Observational	136	13.4	454	46.0	9.2	120	1.2	3.3	956
Experimental	154	16.7	1153	41.8	7.3	174	1.8	4.7	124
Native	142	13.8	617	45.9	8.5	124	1.3	3.7	850
Exotic	125	13.5	229	44.3	10.9	135	1.4	2.4	230
Boreal	167	14.6	1998	42.9	3.3	109	1.4	1.3	175
Temperate	133	13.6	251	46.0	10.1	130	1.3	3.9	905
Tree	114	12.7	289	43.5	10.5	119	1.5	1.4	183
Shrub	125	12.3	446	44.8	8.5	103	1.3	2.9	148
Herb	146	14.2	608	46.1	8.7	129	1.4	3.8	659
Grass	150	15.3	636	46.5	8.9	160	1.4	5.8	90

Note: effects of forcing change and budburst temperature illustrated in Partitioning observed changes.
a) the difference in forcing change between boreal and temperate regions for leafing is 29 (94 – 65), equivalent to 2.8 day expected responses under the average budburst temperature of boreal region (10.2). Similarly, the difference in flowering is equivalent to 1.4 day expected responses, i.e., 21/14.6.
b) the difference in forcing change between grasses and trees + shrubs for leafing is 53 ((101+80)/2 – 37), equivalent to 6.0 day expected responses under the average budburst temperature of grasses (8.9).
c) the difference in forcing change between experimental and observational studies for leafing is 27 (106 –79), equivalent to 2.5 day expected responses under the average budburst temperature of observational studies (11.0).
d) different budburst temperatures explain 2.6 day difference in flowering responses between experimental and observational studies, i.e., 174/13.4 – 174/16.7, and 1.8 day difference in flowering responses between woody (trees and shrubs) and non-woody (herbs and grasses) plants, i.e., (129+160)/(12.7+12.3) – (129+160)/(14.2+15.3).

Table 2

Final stepwise regression coefficients and variable influence on observed phenological responses using Akaike information criterion (AIC).

Budburst	Intercept	Budburst temperature	Altitude	Latitude	MAT	Forcing change	Spring warming	Phenological lag	R² sample (N)
Leafing	2.6439	–0.6075	0.0003	0.0732	0.1480	0.0736	0.9463	–0.9738	0.9453
AIC (%)	449 (27.7)	340 (20.9)	2 (0.1)	15 (0.9)	11 (0.7)	68 (4.2)	7 (0.4)	732 (45.1)	(447)
Flowering	4.5210	–0.5917	0.0003	0.0869	0.1466	0.0521	0.7620	–0.9893	0.9653
AIC (%)	785 (18.1)	247 (5.7)	7 (0.2)	20 (0.5)	32 (0.7)	826 (19.1)	5 (0.1)	2409 (55.6)	(1080)

Note: Budburst temperature (°C) = average temperature at budburst (leafing or flowering) with the warmer climate.
MAT (°C) = long-term mean annual temperature.
Forcing change = warming-induced changes in spring forcing prior to budburst.
Spring warming (°C) = average spring temperature change (Julian day 1 to 182).
Phenological lag (days) = difference between observed response and that expected from forcing change and budburst temperature.
Spring phenology (date of budburst with the baseline climate; Julian days) and long-term mean precipitation (MAP, mm) were excluded from final models due to lack of significance.
Forcing change, budburst temperature, and phenological lag are calculated from phenological and temperature data for individual species, spring phenology and observed responses are provided in the original studies, and spring warming and long-term MAT and MAP are provided in the original studies, calculated from baseline temperatures, or extracted from Google Maps, FreeMapTools, or WorldClim Database (Fick and Hijmans, 2017).