1	Vegetation Dynamics on a Remnant Prairie: Difficulties Measuring Rates of Colonization and Extinction Presented on 11 July 07 at Plant Biology and Botany Congress 2007 Chicago IL Dennis Nyberg
2	Inventory “A detailed list of things in one’s view or possession, esp. a periodic survey of all goods and materials in stock.” Implicit assumption of boundary and time. Inventories have more long-term value as the boundaries and time are more explicit.
3	Value of inventory Value is determined by the goal(s). The goal I am most interested in is estimating colonization and extinction rates of plant species at preserved sites. Other goals of inventories are possible: Quality of site. Quantity that can be harvested. Change is inferred from comparing two or more lists.
4	Transformations From a list of species to a statistic SR, species richness NSR, native species richness FQI, floristic quality index Other attributes that can be calculated by associating variable ‘states’ with species. Height Odor Color
5	Comparing Plant Inventories What are boundaries of area inventoried? The more poorly the boundaries are described the more ambiguity in comparing lists. Species names were determined how? Different authorities delimit species differently. Taxonomy progresses and changes continually.
6	Peacock/Woodworth Inventories Paintin 1926, 1927 published 1929 Betz and Cole 1966 published 1969 Apfelbaum & Rouffa 1979, 1980 published in 1981 Nyberg & Masters 1995, 2000 published in 2001
7
8	Events between inventories Between 1927 and 1966 the east edge became Greenwood Ave., a road, the prairie was sold to a developer and 4 northern acres were developed. Between 1966 and 1981 UIC got ownership of the prairie, and Interpretation Center and parking lot were built. It is probable that contractor brought in soil from another prairie.
9	Events between inventories In 1979 a 10 meter grid system was established on north/south and east/west lines. Marker stakes were placed at the intersect of the lines. Inventories of the square meter surrounding the grid marker were begun. The first part of this talk focuses on ‘meander’ inventories of entire site.
10
11	Changes when comparing lists Five Types of sources of differences Trivial Taxonomic progress Mistakes in identification Detection efficiency New species within boundary or species disappears from within boundary Only the last category is a ‘real’ event of interest to population biologist.
12	Trivial Problems Examples of “trivial” differences between lists Kalmii versus kalmii pratensis versus pratense buabaumii versus buxbaumii With electronic copying and spell checking such differences should decline in frequency.
13	Taxonomic diversity and progress Problem We learn more about populations and the knowledge results in taxonomic changes. I assert one must accept a single taxonomic authority to provide a list of legitimate names. If more than one authority is used, one is forced to 1) become a taxonomic authority oneself, or 2) generate artificially high rates of ‘extinction’ & ‘colonization’.
14	Mistakes in Identification Problem Observers have diverse levels of expertise and attention to taxonomic authorities. Mistakes occur. The traditional approach to this problem is to collect voucher specimens. Digital photographs and freezing bits of tissue should alleviate the voucher collection discrimination against rare species.
15	Mistakes in Identification Problem What information is useful is deciding identification mistakes? Abundance information (on both lists) Presence of other species in genus on list Distribution information (County level) Subtly of species differences (of course subtly varies with observer, so knowing the identifier is always important) I concluded 33 misidentifications on JWP lists.
16	Detection Efficiency Problem Distance sampling explicitly recognizes the efficiency of detection varies. What kinds of information increase the capacity to make inferences about efficiency? Name of observer(s) How much time was spent and on what dates? How clearly are boundaries specified? Future: GPS tracking of meanders
17	Colonization & Extinction Rates Deliberate introductions of species are a problem as one presumably means ‘natural’ colonization when one is estimating the colonization rate. At the James Woodworth Prairie eleven species are known/said to have been deliberately introduced. These eleven have been excluded from all calculations of rates of colonization and extinction.
18
19
20
21
22
23
24
25	Categories of colonizers New colonizers in last two intervals (OOOP and OOPP) have included 33 exotic species and 28 native (non-prairie) species, so rate for these types are high, But the last intervals had 7 new prairie species colonize, so even on this small preserve surrounded by residential and commercial property new prairie species continue to show up.
26
27
28	Fixing probable Detection Errors If we convert every POP sequence to PPP, we eliminate 6 of 48 colonizations by exotics 11 of the 46 colonizations by native –non-prairie And 6 of the 29 colonizations by prairie species Leaving 100 colonizations in 74 years, not much lower than original estimate
29	Quadrat Studies Can quadrat studies solve the ‘boundary’ problem? Any plot whose boundary is a line will always have problems of inclusion/exclusion. Difference is location of lines Movement of individual plant with respect to the line
30	Slight shifts in quadrat boundaries result in list changes
31	"In 1995 an inventory of..." In 1995 an inventory of the square meter around the grid stakes was done by Linda Masters and Gerould Wilhelm. In 2005 an inventory of a ¼ square meter touching the grid marker in a corner was done by Erin Haase. 7 exotic, 7 native and 6 prairie species detected in 2005, that were not detected in larger area in 1995. Probably not all additions were actual colonization, but it is very difficult to separate real from procedural.
32	How can we make the science of detecting ecological change better? Focus should be on place and abundance. Statistical techniques deal with abstractions that have no position/place. All members are equal and dimensionless. Life is intrinsically spatial. We now have the computer capacity to store spatial details. (next slide is example) The tail of species density distribution is long and most difficulties in measuring change pertain to species represented by few individuals.
33
34	Improving Inventories PLACE = mapping ABUNDANCE = focus on populations DETAILS
35	Improving Inventories - Place GPS locations and tracking have dramatically increased spatial specificity. Google Earth is another resource of and to store spatial information. Every inventory list should include a map of the boundaries. Attention to spatial issues raises questions about how extinction & colonization rates should be measured.
36
37	Improving Inventories - Details Details of effort should be available for every published inventory Observer(s), Taxonomic authority providing key Dates & hours Other useful information GPS tracks of meanders Weather info
38	Concluding Remarks Inventories are done for diverse goals. Inventories done by others can be useful, but a variety of problems make measuring change uncertain. There are diverse, easily recorded details that should be included with inventories that will improve the future ability to interpret change. An important new approach is to use mapping of species locations rather than lumping all info into a homogeneous sample.
39	Concluding Remarks -JWP Paintin’s 1929 inventory has been used as a management goal since 2000. Many species on Paintin’s list that are rare (or not detected) at JWP are being grown in the garden that surrounds the Interpretation Center. Hopefully seeds from that garden will colonize the prairie. Thus rates of change are no longer ‘natural’.