SEARCH

SEARCH BY CITATION

Keywords:

  • Arizona (USA);
  • climate change;
  • elevation gradient;
  • first flowering date;
  • phenology;
  • plant–climate interactions;
  • semi-arid environment

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information
  • Temperatures for the southwestern USA are predicted to increase in coming decades, especially during the summer season; however, little is known about how summer precipitation patterns may change. We aimed to better understand how nonsucculent plants of a water-limited gradient encompassing xeric desert to mesic mountain-top may respond to changes in summer conditions.
  • We used a species-rich 26-yr flowering record to determine species’ relationships with precipitation and temperature in months coincident with and previous to flowering.
  • The onset of summer flowering was strongly influenced by the amount and timing of July precipitation, regardless of elevation or life form, suggesting the critical importance of soil moisture in triggering summer flowering in this region.
  • Future changes in the timing or consistency of the early monsoon will probably impact directly on the onset of flowering for many species in this region. In addition, a key implication of predicted increasing temperatures is a decrease in available soil moisture. At all elevations, many species may be expected to flower later in the summer under the decreased soil moisture conditions associated with warmer temperatures. However, impacts on summer flowering may be greater at higher elevations, because of the greater sensitivity of mesic plants to water stress.

Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Changes in plant and animal phenology have been explored across the globe in recent years as important indicators of ecosystem response under changing climate conditions (Badeck et al., 2004; Parmesan, 2007; Gordo & Sanz, 2009; Thackeray et al., 2010). The focus of these studies has been almost exclusively on the spring and autumn seasons. Studies exploring summer phenology are few in number, and have primarily focused on temperate climates (Fitter & Fitter, 2002; Menzel et al., 2006; Kirbyshire & Bigg, 2010). Most have documented an earlier onset of summer phenophases and have associated them with warming temperatures (Defila & Clot, 2001; Menzel et al., 2006; Thackeray et al., 2010).

Drivers of plant phenology are different in water-limited environments than in temperate systems. In dry environments, the importance of precipitation as a driver of ecosystem functioning and phenological activity has been widely documented (Beatley, 1974; Bowers & Dimmitt, 1994; Friedel et al., 1994; Abd E-Ghani, 1997; Ghazanfar, 1997; Peñuelas et al., 2004), regardless of the season under investigation. Precipitation is the dominant limiting factor, shaping species’ composition and productivity (Noy-Meir, 1973; MacMahon & Wagner, 1985; McClaran & Van Devender, 1995) and playing a critical role in cueing phenological events (Abd E-Ghani, 1997; Ghazanfar, 1997; Peñuelas et al., 2004; Crimmins et al., 2010). However, available soil moisture, rather than total precipitation, is what is biologically meaningful to plants (Reynolds et al., 2004). Not all incoming rain in arid systems is useful to plants; factors such as soil depth, soil texture, parent material, organic matter content and soil surface characteristics can affect the extent to which rains will infiltrate the soil or run off its surface (Loik et al., 2004). Precipitation event size, duration and frequency, and air temperatures during and following storm events, also influence soil moisture conditions (Reynolds et al., 2004). Accordingly, future changes to seasonal precipitation patterns in water-limited systems could have major repercussions for plant communities.

The ‘Sky Islands’ region of the arid southwestern USA is characterized by forested mountain ranges separated by desert or grassland plains, and the biotic communities represented along the mountains have been likened to a latitudinal gradient stretching from the southern USA to Canada (Merriam, 1894). Low elevations of this region are characterized by two distinct growth and flowering periods: spring (c. March–May) and summer monsoon (mid-July–mid-October) (Crimmins et al., 2008). At higher elevations in the mountains, plants mimic the unimodal growth and flowering pattern more characteristic of temperate communities; plant activity primarily occurs during the summer season (Crimmins et al., 2008). At all elevations in the Sky Islands region, summer plant activity follows the onset of summer rainfall (Salinas-Zavala et al., 2002; Weiss et al., 2004). In the southwestern USA, monsoon season storms follow a hot, arid foresummer period, during which there is a complete shutdown of flowering, with the exception of succulent plants. Plants then respond rapidly to the initiation of the onset of monsoon rainfall.

Sky Islands offer the unique opportunity to explore plant response to changing climate conditions across a variety of vegetation communities, as the elevation gradients represented are also effectively moisture gradients (Whittaker & Niering, 1965). In the xeric low-elevation desert communities, moisture is limited, variable both seasonally and annually, and highly unpredictable (Noy-Meir, 1973), coming in many small events (< 5 mm) and few large events (> 30 mm) (Loik et al., 2004; Reynolds et al., 2004). In addition, low elevations are characterized by high temperatures and rapid evaporation near the soil surface, resulting in ephemeral soil moisture for much of the summer (Ehleringer et al., 1991; Reynolds et al., 2000). Plants of these communities are adapted to use incoming moisture immediately (Sala & Lauenroth, 1982; Sala et al., 1982). Conversely, mesic mountain-top plant communities are characterized by more total precipitation and lower temperatures, resulting in greater soil water recharge and longer growth episodes throughout the growing season (Whittaker & Niering, 1965). Accordingly, plants of these high-elevation communities behave more like plants of temperate environments, responding more slowly to precipitation events.

Two factors account for the soil moisture gradient present in Sky Islands, characterized by greater and more consistent soil moisture at higher elevations. First, terrain-driven convection and thunderstorm development, the primary mechanisms behind summer monsoon season precipitation at the study site (Watson et al., 1994; Wallace et al., 1999), generally result in more total seasonal precipitation on mountain-tops than in the intervening valleys (Hawkins, 2003). Second, when thunderstorm events and resulting precipitation are able to move or expand from mountain-tops into lower elevation areas, subsequent soil moisture regimes still vary dramatically across the gradient, as cooler temperatures at higher elevations support longer residence of soil moisture (Ehleringer et al., 1991; Reynolds et al., 2000).

Temperatures for the southwestern USA are predicted to increase by 2–4°C in the coming decades, with the strongest trends occurring during the summer season (Karl et al., 2009). However, little is known about how summer precipitation patterns associated with the North American monsoon may change (Liang et al., 2008). The investigation of the relationships between climate characteristics and summer plant response can improve our ability to anticipate potential future changes to the Sky Islands region in its most productive season.

To better understand how the flowering phenology of plants of the southwestern USA Sky Islands region may respond to changes in the important summer monsoon season, we used a species-rich dataset spanning 26 yr to determine species’ relationships with precipitation and temperature for months antecedent to or coincident with summer (July–September) flowering. Our data were collected across a 1200-m elevation gradient, allowing us to explore the consistency in relationships with climatic variables across a moisture gradient encompassing xeric desert scrub to mesic pine forest communities. We investigated patterns by plant life form groups, as life forms of water-limited environments have been shown to employ different strategies to utilize moisture (Ehleringer et al., 1991; Burgess, 1995). However, the soils of the study watershed are consistently shallow (< 50 cm; Whittaker et al., 1968; Etheredge et al., 2004), suggesting competition among life forms for available moisture, rather than partitioning water use by soil depth (Lauenroth et al., 1987; Hunter, 1989; Nobel & Zhang, 1997; Reynolds et al., 2000; Schwinning et al., 2003). We expected that the climatic conditions influencing the onset of summer flowering would vary among life forms and also across the elevation and moisture gradient present. A second objective was to document evidence of changes in the onset of summer flowering over the period 1984–2009. On the basis of the relationships with climate variables identified in these analyses and the documented trends over our study period, we describe the implications of predicted climate change for species in this area.

Materials and Methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

The data examined in this study included 13 222 observations of plants in flower for 240 species made by one of us (C.D.B.). These data represent the results of 623 round-trip (16-km) hikes of the route to Mt Kimball and eight partial trips of 4.8–12.1 km made between the months of July and December over a 26-yr period (1984–2009; 4.4 ± 0.01 hikes month−1 (mean ± SE)), excluding 2004 and 2005, when the observer was unable to complete regular hikes. Days between hikes ranged from zero to 26, and the average number of days between hikes was 6.9 (± 0.2). Hikes were more frequent during months when species were expected to flower or were in flower; long periods between hikes, such as the 26-d period, occurred in months when few or no species were flowering. The route to Mt Kimball is an 8-km hiking route located on the south slope of the Santa Catalina Mountains, c. 7 km north of the city limits of Tucson, Arizona, USA, and follows the Finger Rock and Pima Canyon trails from an elevation of 945 m a.s.l. to the peak of Mt Kimball at 2213 m. The climate in southeastern Arizona is characterized by bimodally distributed precipitation, with most precipitation occurring in either the summer (July–September) monsoon season or winter (December–March) season. The complex terrain results in high temperature and precipitation gradients over short distances.

The observer recorded flowering status for angiosperm species that could be observed from the trail, with the primary focus being an area within 9.1 m (30 ft) of either side of the trail. Observations were recorded along five trail segments, c. 1 mile (1.6 km) in length (Table 1), both on the ascent and descent. Trail mile 1 refers to the lowest elevation segment; trail mile 5 refers to the highest elevation. Blooming was defined as the presence of pollen on anthers. Any species not recognized were collected for expert identification.

Table 1.   Elevation range and dominant biotic communities represented within each 1.6-km segment along the Mt Kimball route, Santa Catalina Mountains, Tucson, Arizona, USA
Trail segmentElevation (m)Elevation range (m)Dominant biotic communities (based on Brown, 1994)
1945–1079134Desert scrub, riparian scrub
21079–1372293Desert scrub, scrub grassland
31372–1671299Scrub grassland, oak woodland
41671–1939268Oak–pine woodland
51939–2213274Oak–pine woodland, pine forest

Each species was assigned to one of the following functional types: annual/biennial forbs, herbaceous perennials and woody plants. Succulent species were excluded from the analysis because these species do not follow the bimodal flowering pattern of other plants in this area and typically flower before 1 July. Functional type groupings were selected on the basis of the different water use strategies employed by plants of semi-arid environments (Ehleringer et al., 1991; Burgess, 1995).

The North American monsoon is the dominant circulation feature that governs the spatial and temporal variability of precipitation across southeastern Arizona during the summer season (Sheppard et al., 2002). A subtle shift from dry southwesterly to moist southeasterly flow occurs across the study area in late June or early July, signalling the beginning of the monsoon season (Adams & Comrie, 1997). The introduction of low-level moisture enhances convective instability and supports the development of primarily terrain-driven thunderstorms across the region (Wallace et al., 1999). This leads to highly localized and intense precipitation events that favor mountain areas. Monsoon season thunderstorm activity typically peaks in August and ends by late September (Liebmann et al., 2008).

Relationship between monsoon onset and monthly precipitation amounts

We examined the role of the timing of the monsoon onset on monthly precipitation totals by calculating local monsoon start dates using daily average dewpoint data from a nearby (< 25 km) meteorological station (Tucson International Airport; data accessed online at http://www.ncdc.noaa.gov). Several studies have documented the importance of the monsoon onset date in modulating total monsoon season precipitation, with later starts correlating well with less total precipitation (Higgins & Shi, 2000; Grantz et al., 2007; Liebmann et al., 2008). Monsoon start dates were determined for each summer season in the study period by finding the first occurrence of three consecutive days with daily average dewpoints exceeding 54°F. Ellis et al. (2004) found that this dewpoint-day criterion correlated well with the onset of consistent monsoon precipitation across southern Arizona. The total number of ‘monsoon days’ (i.e. days with average daily dewpoints > 54°F) was also calculated for June and July to characterize the consistency of low-level moisture after the monsoon onset. The combination of these metrics can help in the diagnosis of the potential daily character of precipitation events not captured in monthly precipitation totals and their subsequent impact on the soil moisture regime of the study area. Total June and July precipitations were regressed with both the monsoon onset date and the number of monsoon days in June and July.

Assessment of potential relationships between summer first flowering dates (FFDs) and monthly climate variables

Within each trail mile, summer FFDs were determined by identifying the date of the first recorded bloom between 1 July and 31 December of each year. The summer FFD was determined for each species–mile combination as the first observation of a single open flower for a species within a mile. FFD was selected for several reasons. First, it captures the commencement of flowering activity following the rather complete shutdown of plant activity during the dry foresummer season. In addition, FFD was selected over metrics such as the mean flowering date because of the highly variable nature of flowering in plants in semi-arid environments: plants in semi-arid environments are highly responsive to environmental conditions and can commence, cease and recommence flowering several times in a single season, making the estimation of the mean flowering date difficult. Finally, the conditions that sustain flowering throughout the season seem to be independent of those which initiate it (Bowers, 1987, 2005; Bowers & Dimmitt, 1994).

As there was no climate monitoring station at the study site, time series data from the Parameter-elevation Regressions on Independent Slopes Model (PRISM) database were used to characterize climate variability for the study site (PRISM Climate Group –Oregon State University, 2010). PRISM is a climate dataset of monthly average minimum and maximum temperatures and total precipitation values interpolated from point station data to a 4-km × 4-km grid for the continental USA (Daly et al., 2008). Time series of summer (June–September) temperature and precipitation values were extracted from four PRISM grid cells that covered the Finger Rock Canyon watershed, and were averaged into single time series of monthly average minimum and maximum temperature and monthly total precipitation for the study area to avoid the over-interpretation of this interpolated dataset.

A forward stepwise regression was performed on each species–mile combination with at least 19 observations during the period 1984–2009, with the observed FFD as the predictand and a pool of monthly climate variables as predictors. One hundred and eighteen unique species were tested. Many of the species appear on multiple trail miles; each species and trail mile combination was tested independently. In total, 334 species–mile combinations were tested. Climate variables included June, July, August and September monthly total precipitation and monthly average minimum and maximum temperatures from 1984 to 2009. Correlations between monthly climate variables were examined to reduce the risk of multicollinearity in the final regression models. Monthly maximum temperatures and total precipitation were highly correlated, especially in July (r = −0.7), because of the strong covariability and inverse relationships between these variables. Months with higher amounts of precipitation are also implicitly months with cloudier conditions and less insolation, limiting maximum temperatures (Portmann et al., 2009). Monthly precipitation was selected over maximum temperature for inclusion in the analysis because this system is limited by moisture, rather than temperature. The relationship between precipitation and monthly minimum temperatures was much weaker (= −0.3), indicating more independence between the two time series, leading to the use of only these two variables in the final predictor pool. All time series were examined for normality and outliers. Two June monthly precipitation values (1984 and 2000) were more than two standard deviations above the mean and excluded as outliers from the analysis. The pool of potential monthly climate predictors was limited to only those months coincident with and before the month of the observed FFD. Statistical tests were performed using Matlab v.R2009a (Mathworks, Inc., 2008).

We separately examined the effects of elevation (trail mile) and functional type on average summer FFD in the highest and lowest terciles of July precipitation (nine wettest and nine driest years of the record). Individual data points were average FFDs calculated over the nine wettest years or nine driest years for each species within each mile. We used the Tukey–Kramer honestly significant difference (HSD) test for multiple comparisons to determine which pairs of means were significantly different. Because data met the requirement of normality, they were not transformed. Statistical analyses were conducted using JMP v8.0 (SAS Institute, Inc., 2008).

We examined the relationship between the average FFD in each mile and the total number of monsoon days in June and July using Pearson’s correlation. Individual data points were average FFDs calculated for each year. Data were not transformed, as they met the assumption of normality.

Testing for trends in summer FFDs

We used simple linear regression to evaluate whether significant trends in summer (June–September) FFDs existed for species on a particular trail segment with at least 19 observations during the study period from 1984 to 2009. As the observations were not collected every day, the FFD was narrowed to a range of dates between the first recorded day in bloom and the previous observation in which the species was not recorded in bloom. To deal with this uncertainty in FFD, linear regressions were performed on 5000 randomly generated combinations of possible FFDs bounded by the first observed and previous ‘not-in-bloom’ date (as in Crimmins et al., 2010). A distribution of regression parameters, including slope coefficients and corresponding P values, was then used to assess the sensitivity of the trend in FFD to sampling error. Species with mean P ≤ 0.05 were considered to have significant trends in FFD sufficiently robust to overcome the uncertainty introduced through the irregular sampling intervals. Significant models were further examined to ensure that assumptions of normally distributed variables and homoscedasticity were met. Statistical tests were performed using Matlab v.R2009a (Mathworks, Inc., 2008). Finally, we used Fisher’s combined test to compare the distribution of species-specific responses with that which would be expected by chance; this test was performed using STATA v.9.2 (StataCorp, 2006).

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Monsoon onset dates vs total June–July precipitation

Monsoon onset dates during the study period varied from day 168 (17 June in 2000) to day 205 (25 July in 1987), with an average start date of day 185 (5 July). Precipitation totalled over June and July was related to both the monsoon onset date (= −0.44, = 0.03) and the number of monsoon days in June and July (= 0.80, < 0.001). Together, the onset date and number of monsoon days explained 62% of the variance in total June–July precipitation over the study period. Supporting Information Fig. S1 shows the relationship between these three variables with the bubble size representing total precipitation. Summers with both early onset and consistent low-level moisture, as represented by the number of monsoon days, also tended to be the wettest. Several years stand out as having early onset dates, but low numbers of monsoon days, indicating an early but inconsistent supply of low-level moisture, which leads to less overall precipitation. Interannual variability in the timing (i.e. monsoon start date) and consistency (i.e. number of monsoon days) of low-level atmospheric moisture leads to large interannual variability in total July precipitation. July exhibited the largest variability in monthly total precipitation of any of the summer months (Table 2).

Table 2.   Monthly total precipitation and average minimum temperature summary statistics for the route to Mt Kimball, Santa Catalina Mountains, Tucson, Arizona, USA
  JuneJulyAugustSeptember
  1. Data source: PRISM Climate Group, Oregon State University, 2010.

Precipitation (mm)Mean4.3778.987.644.8
SD4.1450.22927.1
Slope0.080.66−0.98−0.42
P value0.480.620.200.55
Temperature (°C)Mean18.720.920.117.6
SD1.50.90.81.1
Slope−0.040.020.010.04
P value0.270.500.700.16

Relationships between FFDs and monthly climatic variables

Of the 334 species–mile combinations tested, 157 showed a significant relationship with a single climate variable, and 35 showed a significant relationship with two or more climate variables (Tables 3, S1). The fit (adjusted R2) of the regression equations varied from 0.13 to 0.90. In general, model performance was consistent across trail miles and life forms (Table S1).

Table 3.   Count of significant onset of summer flowering climate models and significant summer onset trends for species recorded along the route to Mt Kimball, Santa Catalina Mountains, Tucson, Arizona, USA
Trail segmentSpecies tested within each mileSignificant relationship with one or more climate variablesSpecies showing significant trend
  1. Values in parentheses are percentages.

Mile 15031 (62)10 (18)
Mile 27044 (63)9 (12)
Mile 39752 (54)13 (13)
Mile 47039 (58)10 (14)
Mile 54726 (55)6 (12)
Total334192 (57)48 (14)

Overwhelmingly, the variable occurring most frequently in models for all life forms and across all miles was July precipitation (Table S1). July precipitation appeared in 110 (57%) of the 192 significant models and was common in models across life forms and miles. In 98% of the models, the relationship was negative, indicating that wetter Julys were associated with an earlier onset of summer flowering. In miles 4 and 5, June precipitation also played a role in many models; in mile 5, June precipitation occurred more frequently in models than July precipitation. The relationships with June precipitation were also in the negative direction, indicating earlier summer flowering with wetter June conditions.

Few models (21 models; 11%) incorporated both precipitation and temperature variables. Nearly all annual plants tested yielded a significant relationship with one or more climate variables (Fig. 1a). Contrary to our expectations, there were no clear patterns in the variables appearing in the models across the elevation gradient or among life forms. Across miles and life forms, precipitation variables were much more common than temperature variables in models of all life forms (Fig. 1); Fig. 2 shows the strong negative relationship between July precipitation and species, representing all trail miles and life forms.

image

Figure 1. Distribution of climate models by life form and trail mile for species. White, species with temperature variables appearing in models; black, species with precipitation variables appearing in models; light gray, species with both precipitation and temperature variables appearing in models; dark gray, species without a significant climate model. (a) Annual/biennial species. (b) Herbaceous perennials. (c) Woody species.

Download figure to PowerPoint

image

Figure 2. Relationship between July precipitation and first flowering date for six plant species recorded along the route to Mt Kimball, Santa Catalina Mountains, Tucson, Arizona, USA, 1984–2009. (a) Setaria macrostachya (plains bristlegrass), an herbaceous perennial in trail mile 1. (b) Acacia angustissima var. hirta (fern acacia), a woody plant in trail mile 2. (c) Muhlenbergia porteri (bush muhly), an herbaceous perennial in trail mile 2. (d) Heterosperma pinnatum (wingpetal), an annual in trail mile 3. (e) Desmodium rosei (rose ticktrefoil), an annual in trail mile 4. (f) Chenopodium graveolens var. neomexicanum (fetid goosefoot), an annual in trail mile 5.

Download figure to PowerPoint

Seasonal timing of FFDs in species with significant climate models

In the nine wettest years (i.e. highest tercile) of the precipitation record, species expressing significant relationships with one or more climate variables showed a clear trend towards earlier onset of flowering with decreasing elevation (Fig. 3a). Herbaceous perennials in miles 1 and 2 flowered significantly earlier than woody species in miles 4 and 5 (< 0.03, Tukey–Kramer HSD). In addition, herbaceous perennials flowered earlier than annual and woody plants in miles 2 and 3, and earlier than woody plants in miles 4 and 5 (Fig. 3a), although these relationships were not statistically significant.

image

Figure 3. Average summer first flowering date (FFD) by life form (mean ± 2SE) by mile for species along the route to Mt Kimball, Santa Catalina Mountains, Tucson, Arizona, USA expressing significant relationships with one or more climate variables over the period 1984–2009. (a) Average summer FFD in the nine wettest years. (b) Average summer FFD in the nine driest years. Annual/biennials, closed circles; herbaceous perennials, open circles; woody plants, closed triangles. Symbols not connected by the same letter are significantly different (Tukey–Kramer honestly significant test). No levels were significantly different in (b). Dashed lines represent global mean (day 237, 25 August).

Download figure to PowerPoint

In the nine driest years (i.e. lowest tercile) of the precipitation record, species expressing significant relationships with one or more climate variables showed little difference in average FFD across life forms or the elevation gradient. There were no significant differences between average FFDs of any mile–life form combinations (Tukey–Kramer HSD). However, species flowering at higher elevations generally tended to flower later than those at lower elevations, and herbaceous perennials tended to begin flowering sooner than annuals and woody species (Fig. 3b). Finally, flowering generally began earlier in the wetter (lowest tercile) years (Fig. 3), across life forms and along the elevation gradient.

The FFDs averaged across species with significant climate models within each mile demonstrated strong relationships with the number of monsoon days in June and July (Pearson’s r from 0.57 to 0.78; < 0.0019). In all miles, the relationship was negative, indicating that a greater number of monsoon days in June and July were associated with an earlier onset of summer flowering.

The average onset of flowering varied considerably from year to year (Fig. 4). The lowest miles showed greater variability in average FFD than higher miles. Some of the years with the most favorable soil moisture conditions, such as 1984, 1990 and 1999, resulted in very early average FFDs, especially in the lowest miles. In general, these years were also characterized by a spread between average FFD between miles 1 and 5 on the order of 20 d. By contrast, some of the driest and least favorable years, such as 1995 and 1997, resulted in the latest onset of flowering across all miles, as well as a compression in the time of average onset of flowering across the miles, on the order of 5–10 d. However, other years (e.g. 1986 and 2000) did not follow this pattern, suggesting a more complex explanation for the patterns observed.

image

Figure 4. (a) Time series of average summer first flowering date (FFD) (mean ± 2SE) for trail miles 1 (solid line) and 5 (dotted line) for species along the route to Mt Kimball, Santa Catalina Mountains, Tucson, Arizona, USA expressing significant relationships with one or more climate variables over the period 1984–2009. Trail miles 2, 3 and 4 not shown to streamline figure. (b) July precipitation over the period 1984–2009.

Download figure to PowerPoint

Trends in the onset of summer flowering

Of the 334 unique species–mile combinations tested for a change in the timing of the start of summer flowering, 48 species–mile combinations (13.5%; Table 3) exhibited a significant trend ( 0.05, Fig. 5), which is significantly greater than would be expected by chance (χ2 = 1106.05, < 0.0001; Fisher’s combination procedure). In mile 1, most of the species’ trends were in the positive direction (towards a later onset of summer flowering); in the remaining miles, the majority of species showed trends in the negative direction (towards an earlier onset of flowering). Trends ranged from c. 1.5 d earlier per year to just over 2 d later per year (Fig. 5).

image

Figure 5. Slope coefficients for species exhibiting a significant trend (< 0.05) in summer first flowering date (FFD) over the period 1984–2009 for species recorded along the route to Mt Kimball, Santa Catalina Mountains, Tucson, Arizona, USA. Starred species demonstrate a significant relationship with July precipitation.

Download figure to PowerPoint

Of the 48 species–mile combinations exhibiting significant trends in summer FFD over this period, 43 species showed significant relationships with one or more climate variables (Fig. 5). July precipitation occurred most frequently in these models (24 models) and, in all cases, the relationship was negative, indicating that earlier summer flowering was associated with wetter July conditions.

Eight species showed significant trends in multiple miles (Euphorbia florida, Artemisia ludoviciana ssp. redolens, Digitaria californica, Eragrostis lehmanniana, Aristida ternipes var. ternipes, Cyperus fendlerianus, Mirabilis albida and Muhlenbergia emersleyi; Fig. 5) and, in all cases, the direction of the trend was the same. The magnitudes of the trends were similar across miles for most species as well. Anecdotal evidence indicates that these species have not shown fluctuations in abundance in the miles in which the trends were noted over the period of record (C. D. Bertelsen, pers. obs.), so that the trends in FFD are not a result of increases or decreases in population sizes (Miller-Rushing et al., 2008).

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

The strong relationship documented in this study between FFDs of species of all life forms at all elevations and precipitation underscores the importance of critical moisture to cue plant phenology in this water-limited system (Figs 1, 2). July precipitation appears to be important for the majority of species across the gradient in this study, regardless of life form. July precipitation is also strongly related to the variability in the timing of the onset of the North American monsoon season. Variability in the timing of consistent low-level atmospheric moisture associated with the beginning of the monsoon season drives much of the variability from year to year in total July precipitation and subsequent favorable soil moisture conditions that drive first flowering events. The importance of early monsoon season (June) precipitation to many species at the highest miles may be explained by the slower response of mesic environment plants to cueing precipitation events. In addition, precipitation falling in June may be more ‘useful’ at higher elevations; the lower evaporation rates occurring at higher elevations mean that soils do not dry down as dramatically between rainfall events as at low elevations with higher temperatures.

There is a clear pattern in the onset of flowering across the elevation and moisture gradient in years with greater summer precipitation, with low-elevation species flowering earliest (Fig. 3). This pattern may be explained by the adaptations of different plant communities: desert plants found at the lowest elevations are adapted to utilize moisture very rapidly as it becomes available (Sala & Lauenroth, 1982; Sala et al., 1982). By contrast, plants of higher elevation mesic communities are not exposed to such harsh conditions and therefore are not adapted to respond as quickly to precipitation (Whittaker & Niering, 1965; Engelbrecht et al., 2005; Crimmins et al., 2008).

Different life forms employ different water exploitation strategies based on root morphology, root depth and photosynthetic pathway (Ehleringer et al., 1991; Burgess, 1995). Annual plants appear to be especially sensitive to moisture conditions; nearly all species tested showed a relationship with precipitation (Fig. 1). Other studies in water-limited environments have also demonstrated the importance of soil moisture to induce flowering (Beatley, 1974; Bowers & Dimmitt, 1994; Friedel et al., 1994; Peñuelas et al., 2004).

The influence of the elevation gradient on the timing of flowering onset was apparent across different life forms. Herbaceous perennial plants tended to flower earlier than either annual or woody species in miles 2 and 3 and earlier than woody plants in miles 4 and 5 (Fig. 3a). When soil moisture in a desert environment is constrained to the upper layers, plants of all functional types access the same moisture (Hunter, 1989; Reynolds et al., 2000, 2004). As the soils of our study area are of a similar depth, differing responses seen in this study by life form do not appear to be caused by soil depth and the partitioning of water use (Lauenroth et al., 1987; Nobel & Zhang, 1997; Briones et al., 1998; Schwinning et al., 2002, 2003). Rather, differences might be explained by preceding conditions and relative differences in water use in upper soil layers, as suggested by Reynolds et al. (2004). For many of the herbaceous perennial species in this study, especially in miles 1–3, summer represents a second period of flowering. Because these plants have already invested energy in putting on leaves and flowers earlier in the year, they only need the appropriate conditions to break dormancy and to flower again. By contrast, summer annual plants must germinate, put on leaves and put on necessary biomass before being able to invest in the production of flowers, resulting in a lag in flowering in annuals in these miles. Herbaceous perennials in miles 4 and 5 primarily flower only in the summer (Crimmins et al., 2008). In these miles, the onset of flowering between herbaceous perennials and annual species is quite similar, suggesting that all of these plants initiate growth and flowering at the same time following adequate summer rainfall.

The slower response in woody species in all miles (Fig. 3a) may be caused by differences in root morphology. Shrubs and trees have coarse roots which cannot efficiently or rapidly use soil water from small rains; these plants require larger events to respond phenologically. By contrast, herbaceous plants have many fine, dense roots and can more effectively use rain from small events (Sala & Lauenroth, 1985).

The bulk of the interannual variability in average onset dates is directly related to July precipitation, which is largely modulated by the arrival of consistent low-level moisture associated with the North American monsoon system. The differences in preceding conditions and in water use by species occurring across the elevation gradient may also account for the year-to-year variability observed in the average onset of flowering across the miles (Fig. 4). In years with early monsoon onset and ample precipitation, low-elevation, desert-adapted plants quickly utilize falling moisture and begin to flower early in the season. At higher elevations in these wetter years, plants also begin to utilize the moisture; however, the response is slower, as these plants respond to incoming moisture in a manner more similar to plants of temperate environments. This results in a spread in average onset of flowering of up to a few weeks across the elevation gradient. By contrast, the onset of flowering appears to be compressed across the elevation gradient in drier years. This may reflect a stronger soil moisture gradient across elevations in drier years, resulting from two potential factors. First, even in dry years, mountain-tops still probably receive more precipitation than lower elevation areas earlier in the season because of the terrain-driven nature of monsoon thunderstorm activity. The early season precipitation at the highest elevations could serve to ‘jump start’ the flowering process in the high miles, whereas low elevations remain dry and dormant. When precipitation does eventually fall at the lower elevations, these plants respond quickly, and average FFDs are compressed in time across the elevation gradient. Alternatively, consistent precipitation falling throughout the watershed yields more favorable soil moisture conditions at higher elevations as a function of lower temperatures, enabling plant activity at higher elevations earlier in the season.

The majority of species exhibiting a significant trend in the onset of summer flowering were in the direction of earlier flowering. Many of the species exhibiting these trends also demonstrated significant relationships with July precipitation, which showed a nonsignificant increasing trend over the study period (Table 2). However, environmental conditions not evaluated in this study probably better explain the trends observed in these species’ FFDs. In Mediterranean studies, Peñuelas et al. (2004) observed similar advancement in the onset of flowering under increasing moisture conditions and, similarly, dry conditions have been documented to delay flowering (Ogaya & Peñuelas, 2003; Llorens & Peñuelas, 2005). By contrast, the majority of species showing significant trends in summer FFDs in the lowest mile of this study were in the direction of later onset. Every one of the species showing a trend towards later flowering in the first mile occurred in areas with shallow, well-drained soils that received most moisture from subsurface and surface drainage, suggesting a different mechanism influencing flowering in this part of the watershed.

The patterns documented in the present study, emphasizing the importance of summer rainfall in the initiation of summer flowering, are in contrast with those documented for this region for the spring season (Crimmins et al., 2010), where previous autumn temperature and precipitation conditions were found to be most influential to the onset of flowering at low elevations, and spring temperature conditions drove the onset of spring flowering at high elevations. Previous work has also investigated the climatic conditions influencing the total number of species in flower seasonally at the study site (Crimmins et al., 2008), and documented that the summer season alpha diversity of species in flower was most influenced by the amount of summer season precipitation at low elevations and by summer temperatures at high elevations. The present study enhances previous work by demonstrating that, although seasonal temperature has a direct effect on how many species may flower over the entire summer season at high elevations in this region, the onset of flowering is mainly driven by the initiation of the monsoon rains.

A large percentage (43%) of the species in this study showed no significant relationships with the climate variables tested (Fig. S1); this number is higher than most values reported for similar studies (Fitter et al., 1995; Sparks & Carey, 1995; Sparks et al., 2000; Miller-Rushing et al., 2007; Miller-Rushing & Primack, 2008). The irregular and sometimes large (several days) sampling interval could account for this, limiting our ability to identify relationships with species with flowering onset that varies by only a few days from year to year. The lack of site-specific environmental variables (precipitation, temperature) probably also plays a role for several of the species; the spatially and temporally coarse data used limit our ability to capture site-specific variability. Finally, variables not tested here (day length, soil moisture at various depths) may be better predictors for the onset of flowering in these species.

Summer temperatures for the southwestern USA are predicted to increase markedly in the coming decades, although little is known about how summer precipitation patterns may change (Liang et al., 2008; Karl et al., 2009). It is clear from our results that any future changes in the timing of the onset or consistency of early monsoon season precipitation will have a direct impact on the timing of first flowering events for many different species endemic to the region. Although it is unclear how the North American monsoon system may change into the future, high confidence in the continuation of warming summer temperatures provides an insight into potential changes in summer flowering patterns across the region. A key implication of increasing temperatures is a decrease in available soil moisture. Recent experimental studies in this region have portended large-scale die-offs of woody plants in response to the increased water stress associated with higher temperatures (Adams et al., 2009). A possible impact on the phenology of many species at all elevations could be later flowering in the summer under decreased effective precipitation in July and soil moisture conditions associated with warmer temperatures. However, because plants at higher elevations in this region are more sensitive to moisture availability, we may expect species of these communities to be impacted more strongly by a relative decrease in available soil moisture, with greater delays in flowering or a cessation of flowering among some species. By contrast, plants of lower elevation communities, already adapted to utilize limited moisture rapidly, may not be impacted as dramatically by a temperature-driven decrease in available soil moisture. A temperature-induced decrease in effective precipitation and soil moisture could be offset by a shift to higher intensity precipitation events in July, but great uncertainty lies in the expected changes to precipitation with the North American monsoon system in a warming world (Liang et al., 2008). Regardless, even small changes in the timing of flowering could have major implications for community organization and composition and ecosystem functioning (Stenseth & Mysterud, 2002; Hegland et al., 2009).

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

We are very thankful to the staff of the University of Arizona Herbarium and the late J. Reeder and C. Reeder, Gramineae specialists, for assistance with plant identification. W. van Leeuwen and K. Landau provided support for the entry of the most recent 5 yr of data. We also appreciate the three anonymous reviewers who provided thoughtful comments that improved the manuscript.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information
  • Abd E-Ghani MM. 1997. Phenology of ten common plant species in western Saudi Arabia. Journal of Arid Environments 35: 673683.
  • Adams DK, Comrie AC. 1997. The North American monsoon. Bulletin of the American Meteorological Society 78: 21972213.
  • Adams HD, Guardiola-Claramonte M, Barron-Gafford GA, Villegas JC, Breshears DD, Zou CB, Troch PA, Huxman TE. 2009. Temperature sensitivity of drought-induced tree mortality portends increased regional die-off under global-change-type drought. Proceedings of the National Academy of Sciences, USA 106: 70637066.
  • Badeck FW, Bondeau F, Bottcher K, Doktor D, Lucht W, Schaber J, Sitch S. 2004. Responses of spring phenology to climate change. New Phytologist 162: 295309.
  • Beatley JC. 1974. Phenological events and their environmental triggers in Mojave desert ecosystems. Ecology 55: 856863.
  • Bowers JE. 2005. El Niño and displays of spring-flowering annuals in the Mojave and Sonoran deserts. Journal of the Torrey Botanical Society 132: 3849.
  • Bowers JE, Dimmitt MA. 1994. Flowering phenology of six woody plants in the northern Sonoran Desert. Bulletin of the Torrey Botanical Club 121: 215229.
  • Bowers MA. 1987. Precipitation and the relative abundances of desert winter annuals: a 6-year study in the northern Mojave Desert. Journal of Arid Environments 12: 141149.
  • Briones O, Montaña C, Ezcurra E. 1998. Competition intensity as a function of resource availability in a semiarid ecosystem. Oecologia 116: 365372.
  • Brown DE. 1994. Biotic communities: southwestern United States and northwestern Mexico. Salt Lake City, USA: University of Utah Press.
  • Burgess TL. 1995. Desert grassland, mixed shrub savanna, shrub steppe, or semidesert scrub? The dilemma of coexisting growth forms. In: McClaranMP, Van DevenderTR, eds. The desert grassland. Tucson, AZ, USA: University of Arizona Press, 3167.
  • Crimmins TM, Crimmins MA, Bertelsen CD. 2010. Complex responses to climate drivers in onset of spring flowering across a semi-arid elevation gradient. Journal of Ecology 98: 10421051.
  • Crimmins TM, Crimmins MA, Bertelsen CD, Balmat J. 2008. Relationships between alpha diversity of plant species in bloom and climatic variables across an elevation gradient. International Journal of Biometeorology 52: 353366.
  • Daly C, Halbleib M, Smith JI, Gibson WP, Doggett MK, Taylor GH, Curtis J, Pasteris PP. 2008. Physiographically sensitive mapping of climatological temperature and precipitation across the conterminous United States. International Journal of Climatology 28: 20312064.
  • Defila C, Clot B. 2001. Phytophenological trends in Switzerland. International Journal of Biometeorology 45: 203207.
  • Ehleringer JR, Phillips SL, Schuster WSF, Sandquist DR. 1991. Differential utilization of summer rains by desert plants. Oecologia 88: 430434.
  • Ellis AW, Saffell EM, Hawkins TW. 2004. A method for defining monsoon onset and demise in the southwestern USA. International Journal of Climatology 24: 247265.
  • Engelbrecht BMJ, Kursar TA, Tyree MT. 2005. Drought effects on seedling survival in a tropical moist forest. Trees 19: 312321.
  • Etheredge D, Gutzler DS, Pazzaglia FJ. 2004. Geomorphic response to seasonal variations in rainfall in the Southwest United States. Geological Society of America Bulletin 116: 606618.
  • Fitter AH, Fitter RSR. 2002. Rapid changes in flowering time in British plants. Science 296: 16891691.
  • Fitter AH, Fitter RSR, Harris ITB, Williamson MH. 1995. Relationships between first flowering date and temperature in the flora of a locality in central England. Functional Ecology 9: 5560.
  • Friedel MH, Nelson DJ, Sparrow AD, Kinloch JE, Maconochie JR. 1994. Flowering and fruiting of arid zone species of Acacia in central Australia. Journal of Arid Environments 27: 221239.
  • Ghazanfar SA. 1997. The phenology of desert plants: a 3 year study in a gravel desert wadi in northern Oman. Journal of Arid Environments 35: 407417.
  • Gordo O, Sanz JJ. 2009. Long-term temporal changes of plant phenology in the western Mediterranean. Global Change Biology 15: 19301948.
  • Grantz K, Rajagopalan B, Clark M, Zagona E. 2007. Seasonal shifts in the North American monsoon. Journal of Climate 20: 19231935.
  • Hawkins TW. 2003. Geostatistical analysis of Arizona summertime precipitation. Journal of the Arizona-Nevada Academy of Science 36: 917.
  • Hegland SJ, Nielsen A, Lázaro A, Bjerknes A-L, Totland Ø. 2009. How does climate warming affect plant–pollinator interactions? Ecology Letters 12: 184195.
  • Higgins RW, Shi W. 2000. Dominant factors responsible for interannual variability of the summer monsoon in the southwestern United States. Journal of Climate 13: 759776.
  • Hunter RB. 1989. Competition between adult and seedling shrubs of Ambrosia dumosa in the Mojave Desert Great Basin. Nature 49: 7984.
  • Karl TR, Melillo JM, Peterson TC. 2009. Global climate change impacts in the United States. Cambridge, UK: Cambridge University Press.
  • Kirbyshire AL, Bigg GR. 2010. Is the onset of the English summer advancing? Climatic Change 100: 419431.
  • Lauenroth WK, Sala OE, Milchunas DG, Lathrop RW. 1987. Root dynamics of Bouteloua gracilis during short-term recovery from drought. Functional Ecology 1: 117124.
  • Liang X-Z, Zhu J, Kunkel KE, Ting M, Wang JXL. 2008. Do CGCMs simulate the North American monsoon precipitation seasonal–interannual variability? Journal of Climate 21: 44244448.
  • Liebmann B, Bladé I, Bond NA, Gochis D, Allured D, Bates GT. 2008. Characteristics of North American summertime rainfall with emphasis on the monsoon. Journal of Climate 21: 12771294.
  • Llorens L, Peñuelas J. 2005. Experimental evidence of future drier and warmer conditions affecting flowering of two co-occurring Mediterranean shrubs. International Journal of Plant Science 166: 235245.
  • Loik ME, Breshears DD, Lauenroth WK, Belnap J. 2004. A multi-scale perspective of water pulses in dryland ecosystems: climatology and ecohydrology of the western USA. Oecologia 141: 269281.
  • MacMahon JA, Wagner FH. 1985. The Mojave, Sonoran, and Chihuahuan deserts of North America. In: EvanariM, Noy-MeirI, GoodallDW, eds. Hot deserts and arid shrublands. Ecosystems of the world. Amsterdam, the Netherlands: Elsevier, 105202.
  • Mathworks, Inc. 2008. Matlab v. R2007b. Natick, MA, USA: Mathworks, Inc.
  • McClaran MP, Van Devender TH. 1995. The desert grassland. Tucson, AZ, USA: University of Arizona Press.
  • Menzel A, Sparks TH, Estrella N, Koch E, Aasa A, Ahas R, Alm-Kubler K, Bissolli P, Braslavska O, Briede A et al. 2006. European phenological response to climate change matches the warming pattern. Global Change Biology 12: 19691976.
  • Merriam CH. 1894. Laws of temperature control of the geographic distribution of terrestrial animals and plants. National Geographic 6: 229238.
  • Miller-Rushing AJ, Inouye DW, Primack RB. 2008. How well do first flowering dates measure plant responses to climate change? The effects of population size and sampling frequency. Journal of Ecology 96: 12891296.
  • Miller-Rushing AJ, Katsuki T, Primack RB, Ishii Y, Lee SD, Higuchi H. 2007. Impact of global warming on a group of related species and their hybrids: cherry tree (Rosaceae) flowering at Mt. Takao, Japan. American Journal of Botany 94: 14701478.
  • Miller-Rushing AJ, Primack RB. 2008. Global warming and flowering times in Thoreau’s Concord: a community perspective. Ecology 89: 332341.
  • Nobel PS, Zhang HH. 1997. Photosynthetic responses of three codominant species from the northwestern Sonoran Desert – a C3 deciduous sub-shrub, a C4 deciduous bunchgrass, and a CAM evergreen leaf succulent. Australian Journal of Plant Physiology 24: 787796.
  • Noy-Meir I. 1973. Desert ecosystems: environment and producers. Annual Review of Ecology and Systematics 4: 2551.
  • Ogaya R, Peñuelas J. 2003. Experimental drought in a holm oak forest: different photosynthetic response of the two dominant species, Quercus ilex and Phillyrea latifolia. Environmental and Experimental Botany 50: 137148.
  • Oregon State University. 2010. PRISM Climate Group [WWW document]. URL http://www.prism.oregonstate.edu/ (accessed 1 July 2010).
  • Parmesan C. 2007. Influences of species, latitudes and methodologies on estimates of phenological response to global warming. Global Change Biology 13: 18601872.
  • Peñuelas J, Filella I, Zhang X, Llorens L, Ogaya R, Lloret F, Comas P, Estiarte M, Terradas J. 2004. Complex spatiotemporal phenological shifts as a response to rainfall changes. New Phytologist 161: 837846.
  • Portmann RW, Solomon S, Hegerl GC. 2009. Spatial and seasonal patterns in climate change, temperatures, and precipitation across the United States. Proceedings of the National Academy of Sciences, USA 106: 73247329.
  • Reynolds JF, Kemp PR, Ogle K. 2004. Modifying the ‘pulse-reserve’ paradigm for deserts of North America: precipitation pulses, soil water, and plant responses. Oecologia 141: 194210.
  • Reynolds JF, Kemp PR, Tenhunen JD. 2000. Effects of long-term rainfall variability on evapotranspiration and soil water distribution in the Chihuahuan Desert: a modeling analysis. Plant Ecology 150: 145159.
  • Sala OE, Lauenroth WK. 1982. Small rainfall events: an ecological role in semiarid regions. Oecologia 53: 301304.
  • Sala OE, Lauenroth WK. 1985. Root profiles and the ecological effect of light rainshowers in arid and semiarid regions. American Midland Naturalist 114: 406408.
  • Sala OE, Lauenroth WK, Parton WJ. 1982. Plant recovery following prolonged drought in a shortgrass steppe. Agricultural Meteorology 27: 4958.
  • Salinas-Zavala CA, Douglas AV, Diaz HF. 2002. Interannual variability of NDVI in northwest Mexico. Associated climatic mechanisms and ecological implications. Remote Sensing of Environment 82: 417430.
  • SAS Institute, Inc. 2008. JMP statistical software. Cary, NC, USA: SAS Institute, Inc.
  • Schwinning S, David K, Richardson L, Ehleringer JR. 2002. Deuterium enriched irrigation indicates different forms of rain use in shrub/grass species of the Colorado Plateau. Oecologia 130: 345355.
  • Schwinning S, Starr BI, Ehleringer JR. 2003. Dominant cold desert plants do not partition warm season precipitation by event size. Oecologica 136: 252260.
  • Sheppard PR, Comrie AC, Packin GD, Angersbach K, Hughes MK. 2002. The climate of the U.S. Southwest. Climate Research 21: 219238.
  • Sparks TH, Carey PD. 1995. The responses of species to climate over two centuries: an analysis of the Marsham phonological record, 1736–1947. Journal of Ecology 83: 321329.
  • Sparks TH, Jeffree EP, Jeffree CE. 2000. An examination of the relationship between flowering times and temperature at the national scale using long-term phonological records from the UK. International Journal of Biometeorology 44: 8287.
  • StataCorp. 2006. Stata v.9.2. College Station, TX, USA: StataCorp.
  • Stenseth NC, Mysterud A. 2002. Climate, changing phenology, and other life history traits: nonlinearity and match–mismatch to the environment. Proceedings of the National Academy of Sciences, USA 99: 1337913381.
  • Thackeray SJ, Sparks TH, Frederiksen M, Burthe S, Bacon PJ, Bell JR, Botham MS, Brereton TM, Bright PW, Carvalho L et al. 2010. Trophic level asynchrony in rates of phenological change for marine, freshwater and terrestrial environments. Global Change Biology 16: 33043313.
  • Wallace CE, Maddox RA, Howard KW. 1999. Summertime convective storm environments in Central Arizona: local observations. Weather and Forecasting 14: 9941006.
  • Watson AI, López RE, Holle RL. 1994. Diurnal cloud-to-ground lightning patterns in Arizona during the Southwest Monsoon. Monthly Weather Review 122: 17161725.
  • Weiss JL, Gutzler DS, Coonrod JEA, Dahm CN. 2004. Seasonal and inter-annual relationships between vegetation and climate in central New Mexico, USA. Journal of Arid Environments 57: 507534.
  • Whittaker RH, Buol SW, Niering WA, Havens YH. 1968. A soil and vegetation pattern in the Santa Catalina Mountains, Arizona. Soil Science 105: 440450.
  • Whittaker RH, Niering WA. 1965. Vegetation of the Santa Catalina Mountains, Arizona: a gradient analysis of the south slope. Ecology 46: 429452.

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Fig. S1 Monsoon onset, persistence and precipitation totals along the route to Mt Kimball, Santa Catalina Mountains, Tucson, Arizona, USA for the period 1984–2009.

Table S1 Significant model coefficients of summer first flowering date for species recorded along the route to Mt Kimball, Santa Catalina Mountains, Tucson, Arizona, USA, 1984–2009

Please note: Wiley-Blackwell are not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing material) should be directed to the New Phytologist Central Office.

FilenameFormatSizeDescription
NPH_3705_sm_FigS1-TableS1.pdf174KSupporting info item