GEOG 566

         Advanced spatial statistics and GIScience

May 26, 2017

A space for time substitution on an impacted landscape

Filed under: Tutorial 2 2017 @ 1:49 pm


Has the total daily energy flow through a functional group at a given elevation changed over the eighty-year interval between 1931 and 2011?


I am hoping to better understand the forces structuring small mammal communities, specifically rodents, across space and through time. In this regard, I am using trap effort-standardized abundance data to evaluate the response of species and functional groups to changes anthropogenic land use practices and climate warming. These data are spatially explicit, and although their spatial coordinates reflect the patterns of trapping, their elevational distribution captures patterns that are reflective of biological responses to a changing environment. Increasing elevation is inversely correlated with summer high temperatures and collinear with winter low temperatures, and positively correlated with increasing precipitation. There is also a trend of decreasing human impact due to land use with increasing elevation, and in the Toiyabe mountain range, human land use for cattle grazing and agriculture has also decreased over time. However, climate warming in the Great Basin has continued over the same time span and has shifted climatic bands upslope.

Species and functional groups respond to these change overtime in dynamic ways which may be recapitulated in their distribution along elevation/ climate gradients. This is referred to as the space for time hypothesis, and generally states that space, which captures the temporal gradient of interest, may be used to stand in for time; with the expectation that species distributions across space reflect their response to the variable of interest through time.

I am interested in the functional groups captured by diet preference, habitat affinity, and body size. Diet preference classes include herbivores, granivores, and omnivores, habitat affinity includes xeric adapted, mesic adapted, and generalist, and body size includes four size bins, 0-30g (size class 1), 40-80g (size class 2), 100-200g (size class 3), and 250-600g (size class 4). Any given species may be any combination of these three functional classes, although not all combinations are actually represented. Thus, patterns in the distribution of these functional groups also reflect tradeoffs that species face, and help to highlight species that may be at risk when functional categorizations become decoupled from environmental conditions. Thus, changes in the elevational distribution of functional groups over time provides key insights to species responses. However, total daily energy flow can also be estimated for these groups, because energy flow through a group is a function of both body size and abundance. This helps us to understand the interactions between abundance, body size, diet, and habitat affinity.

Thus, the question becomes, has the total daily energy flow through a functional group at a given elevation changed over the eighty years between the modern and historic; and based on the cross correlations between body size, diet, and habitat affinity, are their species or groups that appear especially at risk of local extirpation?

Analytical approach and brief description of steps followed to complete the analysis:

To perform this analysis, I used Cran-R to trap effort standardize abundance data for rodent species in the Toiyabe mountain range for both historical and modern data sets. The function in R is an iterative process which resamples all trap sites within the historic and modern data sets that have a trap effort greater than the site with the lowest number of trap nights. This process is automated using R, and generates an average species occurrence from 1,000 iterations. I then used excel to calculate the daily energy flow through each population based on body mass estimates and an equation which employs the principals of allometric scaling, Ei = aNiMib. Where Ei is energy flow (kJ/day), a and b are allometric parameters, specifically, b is the scaling parameter and equal to 0.75. N is the abundance estimate, in this case the trap effort standardized abundance, and M is the estimated body mass (Terry and Rowe 2015).

All trap localities were paired for the historic and modern time-periods, and traps site elevations were binned into three elevation ranges, bellow 5500ft, 5500-6500ft, and above 6500ft. Energy flow for each functional group was calculated and plotted based on the above stated elevation bins.

Brief description of results you obtained:

Total energy flow per day though the modern rodent community in the Toiyabe mountain range is lower than in the historic time period (Figure 1). This finding is consistent with findings by Terry and Rowe, 2015, that total energy through the rodent community in the Great Basin decreases sharply at the Holocene-Anthropocene boundary.

Using climate-elevation correlations we can substitute elevation for time in the Holocene to model the warming climate from the early to late Holocene in the Great Basin. Temperatures increase with decreasing elevation, and precipitation decreases with decreasing elevation. Historically, energy flow through xeric adapted species decreases with increasing elevation, while the opposite is true for mesic adapted species. Notably, energy flow through the historic mesic adapted species increases in a linear fashion with elevation, while in the modern energy flow through this group roughly plateaus at the middle elevation (Figure 2). While both modern and historical patterns are generally consistent with Holocene trends for these functional, neither captures the distinctive decline in energy flow through the xeric adapted group during the modern. This suggests that space for time comparisons may be valid for pre-human impact studies, but that they may not capture abrupt shifts in species responses during the Anthropocene.

Small bodied, omnivorous, habitat generalists demonstrate greatest energy flow at middle elevation in both the historical and modern time-periods, however, total energy flow through these groups are markedly decreased in the modern compared to the historic time periods at these elevations. While generalist demonstrate lower energy flow at the middle and high elevations for the modern compared to the historic, their low elevation values remain unchanged (Figure 2). This is consistent with omnivorous, habitat generalist through the Holocene and into the Anthropocene (Terry and Rowe 2015).

            Finally, herbivores demonstrate a nearly four-fold decrease in energy flow at the lowest elevations over the eighty-year period between the historic and modern time-periods (Figure 2). This is the only result that seems to strongly reflect the abrupt shift from the Holocene to the Anthropocene. This is a particularly surprising result as human land use was greater historically for this mountain range than it is today, and ground cover by herbaceous plants and grasses has increased over this time-period. This may suggest that increased temperatures at lower elevations have pushed herbivorous species upslope, indicating that these species may be more sensitive to climate. Conversely, energy flow thought granivores at low elevations increased from historic to the modern, also consistent with Terry and Rowe, 2015. This however, may be expected, as granivorous rodents in this region also tend to be xeric adapted, while herbivorous rodents tend to be mesic adapted.

            While inference based on this study is limited in scope due to the sample size of one (one mountain range), it does begin to make the case that space for time comparisons should not be assumed to hold as we move into a future where landscapes and climate are increasingly affected by human activity. By doing so, we may fail to capture the subtle ways in which certain ecological patterns are becoming decoupled, and thus, must be wary of drawing spurious conclusions based on old assumptions.

Figure 1. Total energy flow (kJ/day) for each elevational bin in the historic and modern time periods.

Figure 2. From left to right; Energy flow (kJ/day) through functional groups in the historic and modern time periods along the elevation gradient. From top to bottom, functional classifications: size class, diet preference, and habitat affinity.


Critique of the method – what was useful, what was not?

Given the nature of the data, this method of analysis was useful for me. I learned something about the data that was not obvious before my analysis, or comparison with the research at the greater temporal scale captured in the study by Terry and Rowe 2015.  While my inferences are limited, increasing the number of mountain ranges on which I perform this analysis will enable me to perform a number of more informative statistical analyses, and test the spatial gradient at both the elevational scale, and the latitudinal scale, as the three mountain ranges, for which similar data exists occur in a north to south array.


Terry, R.C. and R.J. Rowe. 2015. Energy flow and functional compensation in Great Basin small mammals under natural and anthropogenic environmental change. PNAS, 112(31):9656-9661.

Print Friendly, PDF & Email

No Comments

No comments yet.

RSS feed for comments on this post.

Sorry, the comment form is closed at this time.

© 2019 GEOG 566   Powered by WordPress MU    Hosted by