Overview map of Petrified Forest National Park and the Southern Colorado Plateau Network upland sampling plots included in this report.

Contexts in source publication

Context 1

... park encompasses 88,437 ha in east central Arizona (Figure 1). Park elevation ranges from 1,618-1,900 m. ...

Context 2

... and yucca cover were minimal, especially in clayey fan plots. Cover by life form differed between plots within ecosites ( Figure 10). Shrub cover was greater in the sandy loam ecosite, but also more variable across plots. ...

Context 3

... ecosites were dominated by warm season grasses, including blue grama (Bouteloua gracilis), galleta grass (Hilaria jamesii) and dropseeds Sporobolus spp. (Figure 11). Dropseed composition is primarily S. airoides in the clayey fan ecosite, but also includes S. contractus, S. cryptandrus, and S. flexuous in the sandy loam upland ecosite. ...

Context 4

... spring temperatures and increased spring precipitation had a weaker, but negative effect on species richness. This model predicts high richness in 2010, which is the season we encountered several new species (Figure 15). Inset shows posterior distribution of change over time is centered above 0% indicating an increasing trend. ...

Context 5

... we looked at trend by controlling for variation in weather variables, the posterior distribution for percent change per year was slightly positive (Figure 15, inset graph). The increasing trend in species richness, seen here as well as in the clayey fan site, indicate that our covariates are not adequately explaining the change in species richness over time. ...

Context 6

... strongest effect was monsoon precipitation. In 2012 and 2018, PEFO experienced high spring temperatures following a relatively dry winter and model-predicted cover were at their lowest ( Figure 16). This is concerning, since future climate predictions indicate that warmer, drier springs will become more common at PEFO in the near future ( Andrews et al. 2019a). ...

Context 7

... winter conditions and cool spring temperatures resulted in predicted mean cover at its highest level in 2015 (a year we didn't sample in this ecosite). When we removed the effect of changes in the weather variables from the model, the distribution centered slightly above 0%, indicating a slight increasing trend over the monitoring period (Figure 16, inset graph). (2007)(2008)(2009)(2010)(2011)(2012)(2013)(2014)(2015)(2016)(2017)(2018) as a function of spring temperature and seasonal precipitation at plots sampled in the clayey fan ecosite at Petrified Forest NP. ...

Context 8

... of change in perennial grass cover in the clayey fan ecosite were very similar to that of total foliar cover ( Figure 18). Conditional on the model, spring temperature had a negative effect on grass cover, while the included seasonal precipitation variables had positive effects of roughly the same magnitude. ...

Context 9

... spring temperatures led to decreased cover of perennial grass, while more winter, spring, and monsoon precipitation increased grass cover. Despite a pattern of warming spring temperatures and decreasing precipitation overall at PEFO since 2015, model-predicted perennial grass cover was high in 2015 and 2017 due to above-average winter precipitation received at the park ( Figure 18). In 2015, the wet winter was followed by a cool and wet spring, resulting in a peak in model-predicted grass cover. ...

Context 10

... 2015, the wet winter was followed by a cool and wet spring, resulting in a peak in model-predicted grass cover. Figure 18. Model-based prediction of the change over time in mean perennial grass cover (2007)(2008)(2009)(2010)(2011)(2012)(2013)(2014)(2015)(2016)(2017)(2018) as a function of spring temperature and seasonal precipitation at plots sampled in the clayey fan ecosite at Petrified Forest NP. ...

Context 11

... we controlled for the effect of annual variation in spring temperature and monsoon precipitation by holding them constant at their mean over the monitoring period, there was evidence of a slight increasing trend in perennial grass cover in this ecosite (Figure 18, inset). ...

Context 12

... perennial grass cover showed little annual variability over the monitoring period (Figure 19). The 95% credible intervals are wide, which reflects relatively high plot-to-plot variability in perennial grass cover, particularly the very low perennial grass cover at plot S11 compared to other sandy loam plots (Figure 19). Figure 19. ...

Context 13

... perennial grass cover showed little annual variability over the monitoring period (Figure 19). The 95% credible intervals are wide, which reflects relatively high plot-to-plot variability in perennial grass cover, particularly the very low perennial grass cover at plot S11 compared to other sandy loam plots (Figure 19). Figure 19. ...

Context 14

... 95% credible intervals are wide, which reflects relatively high plot-to-plot variability in perennial grass cover, particularly the very low perennial grass cover at plot S11 compared to other sandy loam plots (Figure 19). Figure 19. Model-based prediction of the change over time in mean perennial grass cover (2007)(2008)(2009)(2010)(2011)(2012)(2013)(2014)(2015)(2016)(2017)(2018) as a function of spring temperature and seasonal precipitation at plots sampled in the sandy loam upland ecosite at Petrified Forest NP. ...

Context 15

... research has indicated that arid grassland species might respond more to a single large precipitation event than to an equal amount of precipitation derived from a series of small rainfall events, a factor we did not investigate in our covariate selection ( Sala et al. 1992). When we removed the effect of the covariates, the model showed little evidence of trend (Figure 19, inset). ...

Context 16

... temperature also had a positive effect. Similar to shrub cover in the clayey fan site, year-to-year variability was relatively small ( Figure 21). When we remove the effect of the covariates, there was no evidence of trend ( Figure 21, inset). ...

Context 17

... to shrub cover in the clayey fan site, year-to-year variability was relatively small ( Figure 21). When we remove the effect of the covariates, there was no evidence of trend ( Figure 21, inset). (2007)(2008)(2009)(2010)(2011)(2012)(2013)(2014)(2015)(2016)(2017)(2018) as a function of spring temperature and seasonal precipitation at plots sampled in the sandy loam upland ecosite at Petrified Forest NP. ...