Donate to Research

Landscape History of Tall Timbers Research Station

Brandy J. Saffell, Ronald E. Masters, and Kevin M. Robertson

Figure 1. A map by Joseph Purcell in 1778 depicting central north Florida including the Tallahassee and Tall Timbers Research Station vicinity. Much of the land was described as "Appalatchi Old Fields" at that time. Old fields are former crop lands that have been allowed to go fallow for a long period of time. These were the same fields that Panfilo de Navaez in 1528 and Hernando de Soto in 1539 traversed and described as extensive agricultural fields, with one account describing a maize field stretching for more than 6 miles.

Figure 1. A map by Joseph Purcell in 1778 depicting central north Florida including the Tallahassee and Tall Timbers Research Station vicinity. Much of the land was described as “Appalatchi Old Fields” at that time. Old fields are former crop lands that have been allowed to go fallow for a long period of time. These were the same fields that Panfilo de Navaez in 1528 and Hernando de Soto in 1539 traversed and described as extensive agricultural fields, with one account describing a maize field stretching for more than 6 miles.

Southeastern forests, woodlands and rangelands have changed dramatically in structure and function since before the arrival of colonists in the 18th century. An often unappreciated fact is that this change happened well before the arrival of colonists in the 18th century. This trajectory of change was redirected further by fire exclusion policies primarily during the 20th century. In order to successfully perform ecological restoration, land managers must determine landscape components, arrangement and change prior to and after settlement. One source of such baseline information is the surveys conducted by the General Land Office (GLO) after the Land Ordinance of 1785. The surveyors recorded tree species, land physiognomy, soil and timber quality, and other observations about the country that they encountered along their designated route. In this study, we utilized 1819-1824 survey notes describing current Tall Timbers Research Station and periphery within 0.8 km, located north of Tallahassee, Florida, on the Georgia border. The data included were from the 1824 Florida survey and an earlier 1819 Georgia survey that erroneously extended into part of North Florida.

Methods

  1. We combined the GLO locations and cover type assessment of surveyors with the following information to create land classifications for each observation point: a digital elevation model (DEM), USDA soil maps, and existing knowledge of tree species ecology, southeastern fire ecology, community structure, and vegetation limits.
  2. We defined the cover type classes for 1824, 1931 and present (2010).
  3. We created an exploratory map with the GLO data using the Spatial Analysis tool Euclidean Allocation in ArcMap. This tool predicts the cover type for each unknown cell (or point) on the map based on the closest observation point.
  4. To visualize the transformation the landscape has experienced since the GLO survey we compared our map with the land cover of Tall Timbers in 1931 and present day. Both were digitized from aerial photography.

Preliminary Results

Table 1. Eight cover types were defined based on surveyor notes for the 1819-1824 land cover map:

1819-1824 Cover Type Description Dominant Species
Forested Wetland Regularly flooded areas Cypress (Taxodium distichum L.)
Hammock Closed canopy forests in drainage areas; roughly 30-45 m. a. s. l.; mixed pine and hardwood species Beech (Fagus grandifolia Ehrh.), magnolia (Magnolia grandiflora L.), and bay (Gordonia lasianthus L.)
Lowland Hammock Hardwood dominated stands; 45-48 m. a. s. l. Southern live oak (Quercus viginiana Mill.)
Lowland Pine Pine dominated stands; 45-48 m. a. s. l. Spruce pine (Pinus glabra Walter)
Hardwood-Pine Mixed stands; majority of hardwood species; above 48 m. a. s. l. Mixed.

 

Pine-Hardwood Mixed stands; majority of pine species; above 48 m. a. s. l. Mixed.

 

Upland Hardwood Hardwood dominated stands; above 48 m. a. s. l. Post oak (Quercus stellata Wangenh.), blackjack oak (Q. marilandica Muenchh.), and hickory (Carya spp.).
Upland Pine Pine dominated stands; above 48 m. a. s. l. Longleaf (Pinus palustris Mill.), shortleaf (P. echinata Mill.), and loblolly pine (P taeda L.).

 

Table 2. The land cover classes were adjusted to create the 1931 and present day land cover maps due to the limited information regarding species composition provided by aerial photographs. Forested wetland, lowland pine, and upland pine remained the same. Also, included were new cover types related to the impact of colonization on the landscape.

1931-Present Day Cover Type Description Species present
Forested Wetland Regularly flooded areas cypress (Taxodium distichum L.)
Hammock-Lowland Hammock Closed canopy forests in drainage areas; roughly 30-45 m. a. s. l.; mixed pine and hardwood species with hardwood dominated stands. beech (Fagus grandifolia Ehrh.), magnolia (Magnolia grandiflora L.), and bay (Gordonia lasianthus L.), southern live oak (Quercus viginiana Mill.)
Mixed Pine and Hardwood Hardwood dominated stands; 45-48 m. a. s. l.; Mixed stands; majority of hardwood species; above 48 m. a. s. l.; Mixed stands; majority of pine species; above 48 m. a. s. l.; Hardwood dominated stands; above 48 m. a. s. l. longleaf (Pinus palustris Mill.), shortleaf (P. echinata Mill.), and loblolly pine (P taeda L.); post oak (Quercus stellata Wangenh.), blackjack oak (Q. marilandica Muenchh.), and hickory (Carya spp.).
Grassland Abandoned, cleared areas or old field N/A
Cultivated Agricultural areas N/A
Developed Buildings and surrounding grounds N/A

 

Figure 2 (a) Presettlement Land Cover (1819-1824). We derived 8 cover types from survey notes. We used Spatial Analyst in ArcMap to predict the dominant landscape pattern based on distance between observation points. The bold grid lines represent the township boundaries along which surveyors traveled. The cross hatched areas cover drainage regions where hammock and hardwood cover types are largely found. (b) 1931 Land Cover and (c) Present Day Land Cover We adjusted the cover type classes from the 1819-1824 map and added new classes to illustrate the changes to the landscape.

Figure 2 (a) Presettlement Land Cover (1819-1824). We derived 8 cover types from survey notes. We used Spatial Analyst in ArcMap to predict the dominant landscape pattern based on distance between observation points. The bold grid lines represent the township boundaries along which surveyors traveled. The cross hatched areas cover drainage regions where hammock and hardwood cover types are largely found. (b) 1931 Land Cover and (c) Present Day Land Cover We adjusted the cover type classes from the 1819-1824 map and added new classes to illustrate the changes to the landscape.

Conclusions

We found that the hammock cover type was the least changed through time with species composition and tree density similar across time periods. We also found that mesic species such as water oak (Quercus nigra) and sweetgum (Liquidambar styraciflua) had moved upslope and off-site in landscape position during intervening years to present.

The methods and results of this study will be useful in determining:

  1. The dominant vegetative structures across a landscape in each time period
  2. What major environmental factors and possibly processes likely governed their distribution
  3. How successful we have been at returning Tall Timbers to some semblance of presettlement conditions
  4. How land managers can utilize environmental conditions and ecological processes (such as fire regime) to create sustainable habitat for native wildlife and vegetation on suitable sites.