xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001498097
007 cr mnu|||uuuuu
008 041209s2004 flua sbm s000|0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0000471
Wilson, Kelly V.
Modification of karst depressions by urbanization in Pinellas County, Florida
h [electronic resource] /
by Kelly V. Wilson.
[Tampa, Fla.] :
University of South Florida,
Thesis (M.S.)--University of South Florida, 2004.
Includes bibliographical references.
Text (Electronic thesis) in PDF format.
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
Title from PDF of title page.
Document formatted into pages; contains 68 pages.
ABSTRACT: This thesis analyzes some of the effects of urbanization in Pinellas County, Florida on the karst landscape. Many sinkholes have been obscured and/or modified for storm water retention by urbanization in Pinellas County, with a few sinkholes still identifiable by characteristic zoning of vegetation, soil moisture, and circular shape. Using aerial photos from 1926 and 2000, karst features were identified by circularity,vegetation, and moisture conditions. Mapping karst surface features using historic aerial photos and maps is a useful exercise that will assist our scientific understanding of karstification in Florida and the nature and extent of karst processes that have acted in the pre-urbanized past.The final product of this research is a digital spatial database and metadata of karst features discernable on the 1926 and 2000 aerial photos; a description of the karst landscape mapped for each time period; and a morphometric description (including sinkhole area, density, and topography) of the karst landscape mapped for each time period. A total of 2,703 sinkholes were identified on the 1926 aerial photos. By 2000 only 900 sinkholes were still visible, a loss of 87.31%. Most of the loss of these sinkholes was due to the rapid urbanization that happened between 1926 and 2000. A total of 499 sinkholes that had been identified in 1926 have now been modified into storm water retention ponds.
Adviser: Brinkmann, Robert.
x Environmental Science and Policy
t USF Electronic Theses and Dissertations.
Modification of Karst Depressions by Ur banization in Pinellas County, Florida by Kelly V. Wilson A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science Department of Environmental Science and Policy College of Arts and Sciences University of South Florida Major Professor: Robert Brinkmann, Ph.D. Philip van Beynen, Ph.D. H. Leonard Vacher, Ph.D. Date of Approval: October 25, 2004 Keywords: sinkhole, gis, kars t, aerial photographs, florida Copyright 2004, Kelly V. Wilson
Acknowledgements I would first like to thank my parents, Wayne and Joanne Wilson. Without their constant love, support, and never letti ng me forget that I could do anything I set my mind to, none of this would have been possible. Thanks Mom and Dad, I love you. Next I would like to thank Dr. Robert Brinkman, who not only gave me this thesis project, but also gave unfailingly cheerful encouragement and advice, and never doubted that I really could do it. Thanks to Dr. H.L. Vacher for allowing me into your lab and always being ready with a quick smile and any geologic advice that I may have needed while circling thousands of sinkholes. An enormous thank you must go to L ee Florea. I would firs t like to thank him for his technical assistance, w ithout which I would still be sitting in front of ArcGIS wondering how to identify sinkholes, and finally for being so incredibly patient with whatever I was asking of him. Thank you to Dr. Philip van Beynen for offering various viewpoints and helping to shape my final thesis. Thank you to the Department of Environmental Science and Policy for s upporting me while pursuing my academic endeavors, and a special thank you to Kare n Schrader for always making sure my paperwork was done and always having some candy for a mid-afternoon pick-me-up. I would also like to thank Pine llas County for funding this thesis project and for supplying the aerial photographs which it was based on. A special thanks to my fellow graduate students, especially Bob, Dave, and Lisa, for offering constant support, incredible friendships, and finally a good laugh when I needed it. I love you all. And finally, thank you to my husband Adam, my number one fan, you came late in the game, but kept me focused in the end when I got lazy, convincing me it was only a couple more months. I love you.
i Table of Contents List of Tables ii List of Figures iii Abstract iv Chapter One Â– Purpose and Introduction 1 Purpose 1 Introduction 1 Karst Landscape 2 Karst Formation 3 Karst Morphology 4 Florida Geology, Geomorphology, Hydrology 6 Florida Geology 6 Florida Geomorphology 7 Florida Hydrology 9 Florida Karst 10 Geomorphology of Sinkholes 11 Classification of Sinkholes 12 Sinkholes in Florida 13 Karst and Sinkholes in Pinellas County 15 Chapter Two Â– Methodology 19 Methodology 19 Data Management 20 Sinkhole Delineation 21 Morphometric Analysis 23 Chapter Three Â– Results 26 1926 Aerial Photos 26 Density, Total Features, Total Area, Percent Area 26 1926 Black and White Aeri al Photo Descriptions 28 Northeastern Pinellas County 28 Central Pinellas County 32 Southern Pinellas County 35 2000 Aerial Photos 35 Density, Total Features, Total Area, Percent Area 35
ii 2000 Color Aerial Photo Descriptions 37 Northeastern Pinellas County 41 Central Pinellas County 41 Southern Pinellas County 44 Chapter Four Â– Discussion 46 1926 and 2000 Aerial Photo Comparison 46 Percent Loss 46 Retention Areas 52 Sinkhole Elevations 52 Chapter Five Conclusions and Sugge stions for Further Research 55 Conclusions 55 Suggestions for Further Research 56 References 58
iii List of Tables Table 1 Karst Landscapes and Descriptions 5 Table 2 1926 Aerial Photo Data 28 Table 3 2000 Aerial Photo Data 37 Table 4 Percent of Sinkholes lost by 2000 52
iv List of Figures Figure 1 Age of Geologic Units in Florida 8 Figure 2 Ge ologic Map of Pinellas County 17 Figure 3 Data Management Chart 20 Figure 4 Map of Fl orida Â– Pinellas County Highlighted 27 Figure 5 1926 Sinkhole Map 29 Figure 6 1926 Areas of Exclusion 30 Figure 7 1926 Developed Areas 31 Figure 8 1926 Aerial P hoto Â– Northeastern Pinellas County 33 Figure 9 1926 Aerial Photo Â– Central Pinellas County 34 Figure 10 1926 Aerial P hoto Â– Southern Pinellas County 36 Figure 11 2000 Sinkhole Map 38 Figure 12 2000 St orm Water Retention Areas 39 Figure 13 2000 Undeveloped Areas 40 Figure 14 2000 Aerial Phot o Â– Northeastern Pinellas County 42 Figure 15 2000 Aerial Photo Â– Central Pinellas County 43 Figure 16 2000 Aerial Photo Â– Southern Pinellas County 45 Figure 17 Possible Sinks Pres ent on Both 2000 and 1926 Aerial Photos 47 Figure 18 Sinks Presen t on Both 2000 and 1926 Aerial Photos 48 Figure 19 Total Sinkholes Presen t in Both 1926 and 2000 w/ Intersecting Areas 49
v Figure 20 Intersecting Areas of Sinkholes Present in both 1926 and 2000 50 Figure 21 Sinkholes which formed after 1926 51 Figure 22 1926 Sinkholes Inters ecting 2000 Storm Water Retention Ponds 53 Figure 23 Topogra phic Map with Sinkhole Locations 54
vi Modification of Karst Depression by Ur banization in Pinellas County, Florida Kelly V. Wilson ABSTRACT This thesis analyzes so me of the effects of urba nization in Pinellas County, Florida on the karst landscape. Many sinkholes have been obscured and/or modified for storm water retention by urbanization in Pinellas County, with a few sinkholes still identifiable by characteristic z oning of vegetation, soil mois ture, and circular shape. Using aerial photos from 1926 and 2000, karst f eatures were identified by circularity, vegetation, and moisture conditi ons. Mapping karst surface features us ing historic aerial photos and maps is a useful exercise that will assist our scien tific understanding of karstification in Florida and the nature and extent of karst pro cesses that have acted in the pre-urbanized past. Th e final product of this research is a digital spatial database and metadata of karst features discernable on the 1926 and 2000 aerial photos; a description of the karst landscape mapped for each time period; and a morphometric description (including sinkhole area, densit y, and topography) of the kars t landscape mapped for each time period. A total of 2,703 sinkholes were identif ied on the 1926 aerial photos. By 2000 only 900 sinkholes were still visible, a lo ss of 87.31%. Most of the loss of these sinkholes was due to the rapi d urbanization that happene d between 1926 and 2000. A total of 499 sinkholes that had been identifi ed in 1926 have now been modified into storm water retention ponds.
1 CHAPTER 1 PURPOSE AND INTRODUCTION Purpose The purpose of this thesis is to analy ze the effects of urbanization in Pinellas County, Florida, on the karst landscape and to map sinkholes modified through urbanization from 1926 to the present. I believ e that the urbanization of Pinellas County has resulted in the filling or modification of sinkholes that were pr esent in the 1926 aerial photos to be filled or modified. I think ma ny of the sinkholes which were modified have become storm water retention ponds. This modification can allow untreated water to enter aquifer systems and thus affect not only Pinellas County, but the rest of Florida as well. The results of this research will be useful to land managers in addressing karst issues in the County and to karst scien tists interested in predevelopment karst geomorphology. This introduction will review va rious aspects of Florida karst systems. Introduction Over 17,000 square miles of the United Stat es has been directly affected by land subsidence with eighty percent of this subsidence being a direct result of development and exploitation of groundwat er resources (Galloway and others, 1999). The increasing development of land and water resources thre atens to aggravate existing land subsidence problems and initiate new ones. The developmen t of karst terrains is riddled with both environmental and engineering hazards (Wilson and Beck, 1988). Extraction and drainage of ground water play a direct role in land subsidence by causing the compaction of vulnerable aquifer systems and the dewate ring of organic soils (Galloway and others,
2 1999). In Florida more than 10 million people and countless agricultura l interests rely on groundwater supply, thus increasing the po ssibility of land subsidence and sinkhole formation (Scott, 2002). The formation of sinkholes, one of the gr eatest karst hazards found in Florida, although fundamentally a natural process, can also be triggered by groundwater level declines caused by pumping, infiltration from reservoir impoundments, surface water diversions, or storm water runoff channels (Galloway and others, 1999). The economic losses due to karst hazards, such as sinkholes, are largely hidden beca use they are spread across an area the size of a state. The only way to completely avoid economic and environmental damage from human activity on ka rst terrain is to a void building on karst, but the demands of economic growth are ap plying overwhelming pressure to develop these areas (Cobb and Currens, 2001). Accord ing to a report published by the Florida Sinkhole Research Institute (FSRI)Â… Â“FloridaÂ’s rapid urban development dramatically increases the threat and the dangers of sinkhole collapse, particularly in the stateÂ’s urban areas. Unprecedented growth, pattern s of drought, and increased water well pumping increase the likelihood of extensive future sinkhole formation activity. Increased risks will accompany continued development.Â” (FSRI, 1983) Karst Landscape The term karst (Slavic kras) means literally, a bleak waterless place (Sinclair and others, 1985). Karst is defined in the Glossary of Geology as Â“a type of topography that is formed on limestone, gypsum and other rock s, primarily by dissolution, and that is
3 characterized by sinkholes, cav es, and underground drainage Â” (Bates and Jackson, p.337, 1980). Karst is terrain with distinctiv e hydrology and landforms arising from a combination of high rock solubility and we ll-developed secondary porosity. Karst is synonymous with limestone landscapes; however other soluble rocks such as dolomite, gypsum, and salt show karstification in some regions (Bloom, 1998). A true karst area has predominately underground drainage with a poorly developed surface network of streams. Karst landscapes are common in humid, temperate climates where ample free water is available to circulat e, but the greatest abundance and variety of karst landscapes are located in warm, tropical environments with lush vegetation and abundant rainfall, and are underlain by other soluble rock (Kochel and others, 1995). About 15 percent of the area of the cont erminous United States has karst-prone rocks at or near the surfa ce, and about 12 percent of th e earthÂ’s land area has exposed carbonate rocks although not all of it shows karst landforms. The most notorious karst region in the United States for constructi on and groundwater probl ems is Florida and parts of Georgia, Alabama, and Kentucky. Other areas of the world facing karst problems include Turkey, Greece, Italy, Fran ce, Spain, the Middle East, and northern Africa (Kochel and others, 1995). Karst Formation Two things are necessary to create the ka rst terrain present in Florida: carbonate rocks and slightly acidic water to dissolve them (Lane, 1986). The solution of limestone in a karst terrain is essentially the solution of CaCO3, by downward moving water which is accomplished through the CaCO3-CO2-H2O chemical reaction. This is an epigenic process that is driven by the hydrologic cycle. Limestone so lution, and therefore karst, is
4 also affected by biologically generated CO2 in decaying humus. Animals and plants can corrode limestone directly as well as crea te biochemical conditions that slow down solution or promote deposition of sediments (Bloom, 1998). The solution process can create and enlarge cavities within karst rocks. This leads to the progressive enlargement of voids beneath the surface allowing large amounts of water to be directed into an underground drai nage system, possibly disrupting the pattern of surface flow. Creation of karst depends on how much water any rock can hold and how easily the water moves through the rock sy stem; these two charac teristics are known as porosity and permeability. Open textures and higher secondary porosity facilitate the solution process and the development of karst (Kochel and others, 1995). Karst Morphology Two important physical characteristic s of karst are lithology and rock permeability. Most karst forms on limestone, which is an extremely diverse rock type. To be defined as a limestone, a rock shoul d contain more than 80 percent calcium and magnesium carbonate, but many limestones cont ain sand, silt, and clay (Bloom, 1998). Limestones can have granular textures and considerable por osity and permeability (Lane, 1986). Karst limestones in general are quite pure. Karst landscapes can develop their own distinctive geomorphology. Table-1 lis ts several landscape types and their characteristics. An estimated 25 percent of the EarthÂ’s hum an population is supp lied with most or all of their waters from karst aquifers (G alloway and others, 1999). Much of the water flows rapidly through conduits from point s ources in the bottom of sinkholes, and discharges at springs or into rivers or th e sea. The complex conduit system of karst
5 aquifers are not all interconnected as expl ained in the following paragraph (Bloom, 1998). Aquifers are subsurface zones of rocks or sediments that yield water in sufficient quantities to be economically useful for manÂ’s activities. Aquifers can be unconfined, semi-confined, or confined. Unconfined aqui fers contain water that is in direct contact with the atmosphere. The zone of sediments or rocks saturated with water is the vadose zone, and is referred to as a surficial or water-table aquifer. The water table Table 1. Karst landscapes and desc riptions (White, 1988 Â– p. 107 117). is the top of the zone of saturation. Semi -confined or confined aquifers are separated from direct contact with the atmosphere by im permeable material, such as clay beds or Doline karst Most common and widely dist ributed landscape; are spotted with sinkholes of varying size Cockpit karst Typically found in sub-tropical and tropical climates; very similar to doline karst but have low depression densities with high sinkhole area ratios Cone and Tower karst Thick massive limestone and well-developed fracture system. Isolated blocks surrounded by alluvial plains. Fluviokarst Abnormal drainage, blind valleys. swallow holes, large springs, closed depressions, and caves. Larger rivers maintain their surface courses and are often fed by underground tributaries. Pavement karst Areas of bare limestone, usually sculpted into karren of various types. Occur in alpine terrains where soils are thin or stripped by glaciation. Polje karst Covers very large areas; Cons ists of poljes which are very large closed depressions; re quires a great thickness of carbonate rocks. Labyrinth karst Dominated by intersecting solution corridors and solution canyons. Cave karst Regions where limestones and ot her soluble rocks crop out at the surface, there are caves a nd a well-developed underground drainage, and little surface expre ssion in the form of closed depressions or other karst landforms is found.
6 consolidated rocks. Confin ement may impose pressure on the contained water and create artesian conditions (Lane, 1986) Springs are expressions of fl ow from a water-table, se mi-confined, or a confined aquifer. When downward percolation of wate r is impeded by a conf ining layer the water is forced to move laterally, downslope, a nd discharge where the permeable sand and less permeable clay intersect the land surface (Lane, 1986). These terms are important when discussing Florida karst systems. A review of Florida karst hydr ology and geomorphology follows. Florida Geology, Geomorphology, Hydrology The following section discusses th e geology, geomorphology, and hydrology in Florida. The Florida Platform is believed to have been part of the West African continental margin near Sengal and was ri fted from that margin during a Triassic breakup. Geochemical and geochronologic da ta have provided support for the proposed correlations of Florida basement terrains with the West Africa and northeastern South America (Heatherington and Mueller, 1997). Florida Geology The entire Florida Platform is cove red by a blanket of carbonate sediments ranging in age from Miocene to Holocene. Older sediments are exposed only along rivers and streams and in sinkholes that cut through younger sediments. Younger sediments were deposited under marine conditi ons (Scott, 1997). The Florida peninsula most likely emerged from a submarine e nvironment during the Neogene (Smith and Lord, 1997).
7 The carbonate formations of Florida ar e generally dolomitic limestones, with varying amounts of interbedded evaporites; th ey represent peritidal and subtidal shelf environments of deposition, reflecting sma ll and large scale sea-level fluctuation (Randazzo, 1997). Dolostones are the dominan t carbonate sediments in the northern twothirds of the peninsula while limestone predominates in the southern peninsula and in the eastern panhandle area (Scott, 1992). Figure 1 visually depicts th e geology of the state of Florida. Florida Geomorphology Marine and coastal proce sses have been the dominant factors in shaping and modifying the Florida platform as well as the exposed peninsula (Scott, 1988). The western edge of the Florida platform lies over 160 kilometers west of Tampa. The eastern edge lies only 4 or 6 kilometers off the co ast of Miami, dropping off steeply to abyssal depths of more than 3 kilometers, creati ng what is known as the Florida escarpment (Lane, 1994). The portion of the Platform that is above sea level, the peninsula, comprises the state of Florida. Sediments were eroded from the southeastern coastal plain and southern Appalachians as the siliciclas tic coastal plain advan ced southward toward the Gulf of Mexico. Sedime nts eventually filled the Gulf trough and spilled onto the carbonate platform. These sediments covere d the limestone and formed the spine of clayey sands on the peninsula. Subsequent sea-level changes altered thes e deposits and modified the elongated system of upland ridges identified on Florid aÂ’s present landscape (Schmidt, 1997). The following section describes the hydrology of the Florida peninsula.
8 Figure 1. Age of Geologic Units in Florid a. (FGS website, taken on 9/9/2003).
9 Florida Hydrology Across most of the state, highly permeable soil or rock is present at or near the land surface. Drainage density is low, but surface-water features include extensive wetlands and more than 7,700 lakes. Many of th e lakes that occur in central and western Florida occupy basins formed by sinkholes that ar e the result of dissoluti on of part of the limestone bedrock. Water levels in these la kes fluctuate directly with variations in aquifer levels. Water levels in many Florid a streams also depend on aquifer levels. The rise and fall in the river stage is generally paralleled by rise and fall in aquifer water levels Â– both of which change in resp onse to precipitation (Miller, 1997). There are five principal aquife rs or aquifer systems in Florida. Four of these crop out at the land surface or are covered by a thin layer of so il and/or weathered rock; the fifth, the intermediate aquifer system of sout hwestern Florida, is completely buried by shallower aquifers or confining units. Th e Floridan aquifer system extends in the subsurface throughout the state and is the most prolific aquifer system in the region (Miller, 1997). Miocene sediments compose the confining unit of the Floridan aquifer system, Miocene-Pliocene sediments form the intermediate aquifer system, and PliocenePleistocene sediments make up the shallow aqui fer system (Scott, 1997). Recharge to the aquifer system occurs over approximately 55 pe rcent of the state as rainwater infiltrates the overlying sediments. Recharge rates vary from less than 2.5 cpy to more than 25 cpy depending on location (Scott, 1992). The geomorphology of the state, along with the geologic framework, controls the distribution of springs throughout the state. The springs o ccur where karst features are common and the surface elevations are low e nough to allow groundwater to flow at the
10 surface (Scott & others, 2002). Florida has 300 known springs with a combined estimated discharge of about 12,600 cfs or ei ght billion gpd (Lane, 1986). Florida has 33 first-magnitude springs, more than any other state or country. Fi rst-magnitude springs discharge more than 64 million gallons of wa ter per day or more than 100 cfs (Scott, 2002). All of FloridaÂ’s major springs discha rge from the Floridan Aquifer, which is estimated to hold more than 2.2 quadrillion gall ons of fresh water. The water flowing out of these springs allows a window into the aquife r, thus assisting in determining the health of the aquifer (Scott, & others, 2002). The following section will summarize the many karst features found in Florida. Florida Karst Karst development in westcentral Florida is contro lled by lithology and water movement, dissolution by chemically aggressive water, aquifer material, and sea levels (Sinclair and others, 1985). Li mestone dissolution in Florida has been intense because of the warm climate, heavy precipitation, low relief that encourages infiltration, and multiple Pleistocene sea-level changes of 100 me ters or more (Bloom, 1998). Florida is almost entirely underlain by carbonate rocks, almost all of which are limestone or dolomite. The carbonate-evaporite sequences of Florida are extremely vulnerable to recrystallization, replacement, dissolution, and ce mentation. This diagenesis of FloridaÂ’s carbonate rocks has produced si gnificant changes in rock ma keup and has facilitated the development of many types of pore spaces incl uding moldic, vug, and interparticle pore. Areas with extensive dissolution can create caves and caverns, wh ich are the underlying essentials of karst landscapes (Randazz o, 1997; Lane, 1986; Scott, 1992).
11 Karst drainage in Florida is char acterized by sinkholes, springs, caves, disappearing streams, and underground drainage channels. Chemical weathering is the predominant erosive process that forms the ka rst terrains like the one s found in Florida. Limestones, by nature, tend to be fractured, jo inted, laminated, and have units of differing texture, all characteristics which, from the standpoint of percolating ground water, are potential zones of weakness, which can be enlarged and extended by acidic rain water (Lane, 1994; Scott, 1992). Chemical weathering of limestone is the ul timate cause of land subsidence such as sinkhole development, but localized stress may also trigger overburden collapse into preexisting cavities. Most of FloridaÂ’s more than 7,700 lakes are solution-based lakes created by groundwater solution of underlyi ng limestone and subsequent lowering of local land surface. These lakes as a result, ha ve physical characteristics of sinkholes such as relatively steep sloping sides, no surface streams into or out of them, and circular outlines (Lane, 1986). Geomorphology of Sinkholes The term doline (sinkhole) describes a particular landform produced by karst processes. Dolines are ch aracterized by circular depressions and underground water drainage networks (Tihansky, 1999). Identically shaped closed depressions can form by subsidence, volcanism, wind deflation, glaciatio n Â– or any other process that selectively displaces a mass of rock and permits the surrounding material to slump into the excavation. Some sinkholes have gently sloping sides, while others especially those known as collapse sinks, have vertical or overhanging cl iffs. Sinkholes are usua lly circular in plan
12 view and less commonly elongate or oval. Sinkholes range in size from shallow soil depressions a few meters in diameter and a meter deep to major landforms several kilometers in diameter and hundreds of meters in depth (Bloom, 1998). Factors that initiate and enlarge sink holes include: gr oundwater recharge, secondary porosity, overburden thickness and sheer strength, and hydraulic gradient. The initiation and enlargement of primary sinkhol es leads to the generation of secondary sinkholes. Mixed sinkhole popul ations, those with primary and secondary sinks, can exist within the same karst area (Kochel and others, 1995). Sinkholes not only have an impact on surface land features but also have an impact on hydrological systems such as lakes, streams, and wetlands by changing wate r chemistry and rates of recharge or runoff (Tihansky, 1999). Classification of Sinkholes Five major classes of sinkholes or dolines are recognize d. The two most contrasting types are the funne l-shaped solution/subsidence do line and the steep or cliffed collapse doline. Subsidence dolines and collapse dolines ar e surface forms in nonsoluble rock, caused by solution of buried karst (Bloom, 1998). Mo st sinkholes are a combination of the solutiona l and collapse types. Solution sinkholes form as water infiltrati ng into joints and fissures enlarges the cracks by corrosion. A cone depression is produced in the perched water table as downward flow rates in the enlarged fractur es exceed that of the surrounding area. Collapse sinkholes are depressions that are in itiated by solution that occurs beneath the surface. The expansion of caverns, caused by corrosion and reduction of buoyant support
13 may lead to collapse by decreasing the suppor t of the overlying rock material (Kochel and others, 1995). The following section discusses sink hole formation in Florida. Sinkholes in Florida Sinkholes are a predominant landform in Flor ida; with west-central Florida being delineated as having the highest frequenc y of sinkhole activity. Sinkhole formation commonly damages buildings, roads, and uti lities, diminishing the usefulness of the affected land. Millions of dollars are lost each year in Florida because of structural damage caused by sinkholes (Upchurch a nd Randazzo, 1997). The Florida Sinkhole Research Institute conservatively estima ted that sinkholes cause approximately $10 million in damage each year in the stat e of Florida (Beck and Sayed, 1991). The occurrence of sinkholes has become more frequent with the increased development of ground water and land res ources (Tihansky, 1999). Dissolution of limestone in Florida appears to occur preferentially in rech arge areas, such as sinkholes and wetlands, and near the saltwater/fr eshwater coastal mixing zone. Recharge environments are the more important of these two environments for sinkhole development (Upchurch and Randazzo, 1997). Sinkholes are more common where the Floridan aquifer system is unconfined or thinly confined (Miller, 1997). The multiple stages of the development of karst in Florida have resulted in a number of different types of sinkholes pres ent in the state. The most common are collapse and solution sinkholes. Collapse si nkholes form when limestone substrate has layers of clays and sands overlying it, and a cav ern below. If the c over is cohesive and thick enough, it may be able to bridge a cavern in the absence of a limestone roof. When
14 the cover eventually collapses, a large f unnel shaped depression results. Solution sinkholes form as a result of dissolution of rock. The geologic framework is similar to that of a collapse sinkhole, with the exception that a group of nearly vertical joints is present. If water passes vertically along the joints, they will become enlarged by dissolution. The removal of ro ck by dissolution allows settling of rock and washing of overburden into the cavern. Consequently slow subsidence occurs (Tihansky, 1999). Because there is a long histor y of karstification in Florida, sinkholes vary in age and degree of development. Many of the ol der sinkholes have been partially filled by marine and wetland sediments. These older, partly to completely filled sinkholes are called alluvial sinkholes (U pchurch and Randazzo 1997). Where the water table is shallow, they form lakes and cypress do mes. Rejuvenated alluvial sinkholes are sometimes called Â“raveling sinks.Â” Alluvial sinkholes are reactiv ated by a number of different processes. Where heavy pumping is present, especially for agricultural freeze protection when the water table is low, ma ny sinkholes become reactivated. Turbulent flow causes erosion and reduction of the hydrau lic head in the limestone aquifer reducing buoyancy and support of the overburden. The er osion is initially ac companied by loss of cohesion and upward piping. Settling and cr acking begins, the failure works upward until the cover can no longer be supported, and finally rapid subsidence begins (Tihansky, 1999). Sinkholes can operate as pathways of local or regional groundwater contamination. In the past, sinkholes were c onsidered to be conveni ent, low-cost wastedisposal sites. Because kars t areas typically have internal drainage, agricultural wastes, lawn fertilizers, and other sources of nut rients are unintentionally washed into the
15 sinkholes, and eventually into the aquifers below. These excess nutrients can cause enrichment of surface waters and gr oundwater (Upchurch and Randazzo, 1997). Karst and Sinkholes in Pinellas County Surficial evidence of a karst landscape in Pinellas County is limited to sinkholes, sinkhole associated features, and springs. Th ese features provide distinct problems for planners and environmental scientists in the region. Land stability is perhaps the largest issue for citizens of Pinellas County; with contaminant transport through karst features, water quality, and hurricanes on the forefront as well (Schmidt, 1997). At least 500 homes have been damaged by sinkholes in Pinellas County, Florida between 1990 and 1994, according to the count y property appraiser (Hutchinson, 1994), with more forming every year. Sinkholes can form from a variety of processes. Those found in Pinellas County are cove r-collapse sinkholes that fo rm when Pleistocene sands above limestone collapse into cavit ies in the bedrock (Schmidt, 1997). Pinellas County is a peninsula, consis ting of a mainland and several barrier islands or keys, and is part of the Gulf Coastal Lowlands physiographic region as described by White (1970). This area consis ts of low-angle scarps and terraces formed during several Pleistocene sealevel stands as illustrated in Figure 2 (White, 1970). In Pinellas County the overburden materials are generally thicker, 9 Â– 60 meters thick, and less permeable than in other areas of this regi on. Greater cohesion of the clay in this area postpones failure, and ultimate collapse tends to occur more abruptly in the form of cover collapse sinkholes (Tihansky, 1999; Frank a nd Beck, 1991). The uppermost consolidated rock in Pinellas County is Tampa limestone which is white to light yellow, soft, moderately sandy and clayey, finely gra nular, and locally fossiliferous, with high
16 porosity. Recharge to the aquifer in areas where this limestone is present is likely concentrated at sinkholes. Th e water table in the surficial aquifer in this area generally lies within a few meters of the land surface. Water in the Floridan aquifer in Pinellas County is under artesian pressure and will rise in tightly cased wells to a level above the clay confining layer. Movement of water from the surficial to the Floridan aquifer is greatly accelerated where the clay is absent or has been breached by sinkhole collapse or subsidence (Lane, 1986). Pinellas County has limited potable water s upplies and depends on water delivered from the surrounding counties of Pasco and H illsborough (Broska and Barnette, 1999). Recent occurrence of sinkholes has been re lated to abrupt water-level declines caused by pumping. Collapse and subsiden ce is caused, not by recent solution of limestone but by downward movement of the unconsolidated surficial material that overlies and fills existing cavities in the ro ck filling voids in which water has been removed (Sinclair, 1982). The surficial aquifer system in the area is recharged almost entirely from local precipitation. In much of west-central Fl orida water moves downward from the surficial aquifer system to recharge the Upper Florid an aquifer. The karst is mantled and less apparent at the land surface, except when th e overburden collapses or subsides into a subsurface cavity. Many small sinkhole-like depressions occur in Pinellas County, but only a few are directly connected to the underl ying limestone. The ones that are directly connected to the Floridan aquifer are near the coast and the Upper Floridan aquifer contains saline water in th ese areas. (Trommer, 1987)
17 Figure 2. Geologic Map of Pi nellas County, Florida. Qbd Â– beach ridge and dune; Qh Â– Holocene sediments; Qu Â– undifferentiated sedi ments; TQsu Â– shelly sediments of PlioPleistocene age; Th Â– Hawt horn Group; Arcadia Formati on, Tampa Member. (Modified from the FDEP Geodata Library, taken on 09/09/03).
18 Because of its mild subtropical climat e, Pinellas County has become the most densely populated county in the State of Florida, w ith a population that is currently more than 900,000. The following chapter describes the met hodology used to perform the digitization of sinkholes and to determine the loss of si nkholes to urbanization in Pinellas County, Florida.
19 CHAPTER 2 METHODOLOGY Methodology Pinellas County, in west-central Florida, is located in one of the more active and least understood karst region of the world. Some of the more common karst features found in this area are dolines, also known as sinks and sinkholes. The topographic expression of these depressions is partially masked by Holocene sand deposits and recent urban development. Urbanization results in filled depression featur es and modifications for storm water retention. Mapping and anal ysis of pre-and post -development sinkholes have not been completed in order to asse ss the distribution of sinkholes and document changes in topography due to urba nization in Pinellas County. This is problematic in that Pinellas County, like much of the Tampa Bay re gion continues to urba nize rapidly and is likely to undertake significant construction projects that ha ve the potential to directly impact the karst landscape. Without a clea r understanding of wher e karst processes are still active, it is difficult to implement th e appropriate land use and management decisions suitable for the geology and population of the region. As Pinellas County moves toward redevelopment of lands, it will be important to understand the form er landscape that once existed in order to make appropriate land use decisions. It is clear from studies of sinkholes in undeveloped porti ons of Florida that they occur in particular clusters or Â“sinkhole regi onsÂ”. Regions of past sinkhole formation are locations where modern doline formation may also occur. Certainly there are such regions that are hidden by the urbanized nature of areas such as Pinellas County. In this
20 study, karst features will be identified by circularity, vegetation and soil moisture variations. Mapping karst surface features using histor ic aerial photos and maps will be a useful exercise that will assi st our scientific understanding of karstification in Florida and the nature and extent of karst processes that have acted in the pre-urbanized past. The methods for completing this research follow and are outlined in Figure 3. Figure 3. Data Management Chart. Data Management I have constructed a spatial database for this project. The database includes historic aerial photos from 1926 and more recent photos from 2000. The photos were acquired from the Pinellas County government. The digital topographic maps that are the framework for the GIS maps in the County were obtained from the Florida Geographic Data Library, University of Florida. The me tadata for these base maps are the foundation on which GIS layers were created for this study. DATA Sources Management Collection Analysis 1926 Air PhotoÂ’s 2000 Air PhotoÂ’s ArcGIS Digitization Identification Criteria Loss to Urbanization Morphometric
21 The 1926 aerial photos required georefer encing within ARCGIS (ESRI) due to their age and lack of a geographic project ion. The 1926 aerial photos are georeferenced first by being projected into the Albers coordinate system, using Datum D North American 1983 (NAD83). I then inserted th em into the table provided by Pinellas County to their approximate locations. Next, I had to georeference each photo to either a geomorphic detail found on the topographic maps or anthropogenic markers, such as bridges and roads. The photos are then rect ified to their actual ge ographic location. This was accomplished by stretching the photos with in ArcGIS. Finally, I constructed GIS layers of karst features id entifiable on the historic, black and white 1926 air photos. There are four layers in this map including f eatures I have discerned as sinks and possible sinks, a layer for exclusions, or areas that we re obscured or unclear in the photos and a layer for developed areas, these areas showed notable development such as clusters of roads and/or structures, with the smallest identifiable area being 0.027 km2. I then constructed GIS layers of karst features identifiable on the 2000 air photos, which are full-color high-resolution infrared images. Theses images were already georeferenced and thus require d nothing more than simple in sertion into a new ArcGIS map. This map contains four layers includ ing the features I have identified as sinks, possible sinks, storm water/retention ponds, and undeveloped areas. These areas consisted of wetlands and forests, with the smallest identif iable area being 0.32 km2. Sinkhole Delineation Specific karst features, such as sinkholes are often recognized by their signatures, or combination of image characteristics, whic h include tone, texture, shape, size, shadow, height, and spatial re lationship. Many sinkholes have b een obscured by urbanization in
22 Pinellas County, but numerous sinkholes should be definable by charac teristic zoning of vegetation and soil moisture, which will still persist even after human alteration or infilling (Lyon and McCarthy, 1995). Surface properties such as mineralogy, textures, color, and moisture differences can be identified on aerial photos (Coker, 1969) Sinkhole indicators that I will be using include: vegetation Â– changes/variation; wa ter Â– presence/absence; soil moisture; and shapes. The sinks layer on both the 1926 and 2000 aerial photo maps are features that I concluded were most likely sinkholes. The crit eria used to identify these features were circular and/or combined circ ular shapes, forming uvalas or coalescing sinkholes. Where a depression contour enclosed two of more subsets of closed depressions, the entire feature is treated as a sing le sinkhole. The possible si nks layer on both the 1926 photos and 2000 photos are areas of possi ble sinkhole activity lacking an entirely circular shape, but showed other signs such as soil moisture variations, some circular form, presence of water and indicator vegetation. Lakes that appeared to be depressions on the aerial photographs were treated as karst features a nd according to their sh ape where categorized as either sinks or possible sinks. The sinks a nd possible sinks layers were each identified in layers of different colors. On the 2000 aerial photos I also created a retention pond layer, identifying areas that I considered man-made structures. The cr iteria I used to identify these areas were straight edges on at least one side of the de pression, and the presence of water or soil moisture. The smallest identifiable retent ion pond feature identif iable on the 2000 aerial photos had an area of 260 m2.
23 The 1926 aerial photos were taken during th e dry season in FloridaMarch, April, and May so it must be noted that many of the sinks which would normally contain water, and thus be more easily identifi ed, were dry, so vegetation patterns and geomorphology were used to identify the circular patterns. It must also be noted that these criteria were difficult ones with whic h to work. The 1926 aerial photos are black and white and of very poor quality due to ag e, technology available and loss of resolution due to scanning of images (pixilation). Consequently many areas of these photos are completely blacked out and thus an exclusi on layer was created to remove these areas when carrying out density estimates and othe r analyses about sinkhol es in the County. Complete resolutions of some of the photos, both 1926 and 2000, is obscured on some frames by cloud cover and reflection from water bodies and thus were unreadable. Littlefield (1988) defined sinkholes as an area on topogr aphic maps enclosed by closed depression contours, whet her circular or irregular in shape. The minimal diameter at which a sinkhole could be recognized on the topographic maps used in his study was estimated to be 0.027 kilometers. Small sinkhol es were not identifiable due to small surface areas (Littlefield, 1988). The resolu tion of the 1926 aerial photos allowed sinks and possible sinks with a mini mal area of approximately 89 m2 to be identified. The high quality resolution of the 2000 aer ial photos would seem to al low for greater recognition of sinkholes with a minimal area smaller than those in the 1926 aer ial photos, but the smallest identifiable sinks and possi ble sinks were approximately 170 m2 in area. Morphometric Analysis I calculated morphometric characteristics of the identified features using ArcGIS. ArcGIS automatically calculat ed the area within the attr ibutes table. The sinkhole
24 density in Pinellas County is determined for 1926 and 2000 using numbers supplied by ArcGIS. US Census Bureau fi gures (2000) for the Pinellas C ounty estimate the total area to be 1,574 km2, with a land area of 725 km2 (the figure I used to determine the density). ArcGIS automatically calculated the total area in square kilome ters for each layer. I then used these numbers to calculate the dens ity for each layer in 1926 and 2000, using the following calculation: D= N/T D = density N = number of sinks T = total land area I then determined the percent of land area th at was covered by the sinks for each layer in 1926 and 2000, using the following calculation: P = K/T P = land area covered K = total square kilometers of each individual layer T = total land area To asses the effects that urbanization ha s had on the sinks of Pinellas County I then combined many of these calculations and maps to assess the modifications that have occurred. First I calculated th e loss of the features that were identified. The calculation used: L=A/B L = loss of sinkholes A = 1926 identified sinkhole area B = 2000 identified sinkhole area
25 I then created a map with a layer consisting of the total sinks from 1926 and a layer of the total sinks from 2000. I then used the intersect functi on in ArcGIS, and the results were the sinks that ar e still present in 2000. I then used the map previously described and intersected it with the 1926 total sinkholes map. This map illustrated sinkholes that have formed since the 1926 aeria l photos were taken. I also attempted to determine which sinkholes that were identifi ed in 1926 have now been converted into storm water and/or retention pond areas in the 2000 aerial photos. To do this I created a new map which consisted of a layer of the total sinks iden tified in 1926 and the retention pond layer from 2000, and preformed an intersec t function in ArcGIS. This intersected the areas from the 1926 total sinks laye r and the 2000 retention pond layer that overlapped. Finally, I took the 1926 sinkhole map and the 2000 sinkhole map and combined them with a topography map. This illustrated the sinkholes and at what elevation they occurred. The final product of this research is a di gital spatial database of karst features discernable on the 1926 and 2000 ai r photos; a description of the karst landscape mapped for each time period; and a morphometric de scription (including sinkhole area, density, and circularity) of the karst landscape mappe d for each time period. With these data I hope to be able to characterize the kars t landscape and determine a loss of karst landforms due to urbanization.
26 CHAPTER 3 RESULTS Pinellas County, in West Central Florida (F igure 4), is located in one of the more active and least understood kars t region of the world. Surf icial evidence of a karst landscape is limited to sinkholes, si nkhole associated features, a nd springs in this county. The topographic expression of these depressi ons is partially masked by Holocene sand deposits and recent urban development. Urbani zation results in filled depression features and modifications to others for storm water retention. The following information is the resu lt of the digitizati on of the 1926 and 2000 aerial photos of Pinellas County. 1926 Aerial Photos The 1926 aerial photos acquired from Pi nellas County are the foundation of the calculations, inferences, and results within this thesis. Density, Total Features, Total Area, Percent Land Area The number of sinks identified in th e 1926 aerial photos is 1,570. They cover 19.34 km2. The density is 2.20 sinks per km2. 2.70% of Pinellas County is covered by sinks (Table 2, Figure 5). The number of po ssible sinks identified is 1,133. They cover 24.56 km2. The density is 1.59 possible sinks per km2. 3.44% of Pinellas County is covered by possible sinks (Table 2, Figure 5). Together, the identified sinks and possibl e sinks (collectively called sinkholes in this thesis) are 2,703 in number, 43.9 km2 in area, and 3.79 km2 in density. They cover 6.20% of the total land area of Pinellas County (Table 2).
27 Figure 4. Map of Florida Â– Pinellas County highlighted.
28 Areas of exclusion covered 11.26 km2 of Pinellas County, which accounted for 1.58% of the total land area of the c ounty (Table 2, Figure 6). Total # of Features Total Area (km2) Density (per/km2) Percent Land Area (%) Sinks 1570 19.34 2.20 2.70 Possible Sinks 1133 24.56 1.59 3.44 Exclusions N/A 11.26 N/A 1.58 Undeveloped N/A 609.99 N/A 85.46 Combined Sinkholes 2703 43.90 3.79 6.20 Table 2. 1926 Aerial Photo Data. Total ar ea, density, and percent land area of sinkholes, combined sinkholes, possible sinkholes, excl usion areas, and undeveloped land mapped using the 1926 air photos. As part of the analysis of the 1926 aerial photos, I examined the amount of land that was undeveloped in the county. Th e undeveloped areas of Pinellas County accounted for 609.99 km2. This accounts for 85.46% of the total land area in Pinellas County (Table 2, Figure 7). 1926 Black and White Aerial Photo Description A total of 36 black and white aerial photos were taken in 1926. The following photos and descriptions are a sample of 3 photos from th e northeastern, central, a nd southern parts of Pinellas County. The barrier islands were excl uded in this comparis on due to the lack of karstic formations and sinkhole activity. Northeastern Pinellas County The photo in Figure 8 is located in the northeastern portion of Pinellas County. To the East is Hillsborough County, and to the North is Pasco County. In 1926 this portion of Pinellas County was completely undeveloped, except for one road which is
29 Figure 5. 1926 sinkholes.
30 Figure 6. 1926 Areas of Exclusion.
31 Figure 7. 1926 Developed Areas.
32 is visible in the southwestern area of th e photograph. Karstic formations are clearly visible in this aerial photo. The majority of this photo is covere d by a highly karstified landscape. In the eastern portion of this aer ial photo many sinkhole features are visible, most of which have formed into uvalas or coalescing sinkholes. The majority of this photo appears to be wetlands, although it is di fficult to determine due to age and quality of the photo. These wetlands likely form ed as a result of the karstic activity. Central Pinellas County Figure 9 is located in cent ral Pinellas County. By 1926 development was rapidly spreading to this area, thus much of the karst landscape wa s being converted into urban areas as well as farmlands. This photo is very dark and thus some feat ures are difficult to discern, but there are several si nkholes still visible in this area. Many roads have been cut into the landscape, as well as several homes In the southeastern portion of this photo there are six sinkholes visible, with roads or possibly trails, connecting them. It is possible that they were used as a wate r supply for the human population or farming activities. In the central area of the photo it appears as though there are several karst features including sinkholes and uvalas. Five of these sinkholes are single sinks with the most northern being a coalescing sink. The individual sinks range from approximately 50 to 150 meters in diameter, with the coales cing sink being approximately 300 meters in diameter. In the south-centra l portion of this photo is a pe rfectly formed sinkhole, as well as another in the southwestern portion of the photo. Both of these sinks appear to be approximately 150 meters in diameter. In the southwestern portion of the photo there also appears to be a coalescing sinkhole appr oximately 5 meters from a road bed. This sinkhole is approximately 500 meters in diameter.
33 Figure 8. 1926 Aerial Photo Â– Nort heastern Pinellas County photo.
34 Figure 9. 1926 Aerial Photo Â– Central Pinellas County.
35 Southern Pinellas County Figure 10 is southern Pinellas County. Th is area was, and is now, known as St. Petersburg. This is the largest city in Pinellas County, and by 1926 was almost completely developed. As can be seen in the image, this area of Pinellas County was fully developed, thus many of the sinkholes th at may have previously existed most likely have been filled or modified. There are stil l several sinkholes that appear in the photo. In the northeastern corner of the photo ther e is a sinkhole which has been named Mirror Lake. It appears to be a coalescing sink and is approximately 500 meters in diameter. In the west-central portion of this photo there appears to be several sinks within close proximity of each other. They range from a pproximately 50 to 300 meters in diameter. 2000 Aerial Photos The 2000 aerial photos acquired from Pi nellas County are the basis for the following calculations and maps. Density, Total Features, Total Area, Percent Land Area The number of sinks identified in th e 2000 aerial photos is 261, covering 1.60 km2, with a density of 0.36 sinks per km2. This land area account s for 0.22% of Pinellas County (Table 3, Figure 11). The number of possible sinks identified was 639, covering 3.97 km2, with a density of 0.88 possible sinks per km2. The possible sinks cover 0.55% of Pinellas County (Table 3, Fi gure 11). As noted previous ly, sinks and possible sinks are collectively going to be called sinkholes throughout the rest of this section. A combined total of 900 sinkholes were identifi ed with a combined total area of 5.57 km2, a combined density of 1.24 sinkholes per km2. The sinkholes covered a combined 0.77% of the
36 Figure 10. 1926 Aerial Photo Southern Pinellas County.
37 total land area of Pinellas County (Table 3). The areas identified as retention ponds/storm water features are area s that had at least one strai ght edge and appeared to be manmade. There were 1,646 retention ponds identified, covering 13.25 km2 of Pinellas County with a density of 2.27 retention pond features per km2. They accounted for 1.83% of the total land area of the county (Table 3, Figure 12). I also calculated the area of undeveloped land in the County. The undevelo ped areas of Pinellas County accounted for 19.54 km2, which is 2.74% of the total land area in Pinellas County (Table 3, Figure 13). Total # of Features Total Area (km2) Density (km2) Percent Land Area Sinks 261 1.60 0.36 0.22 Possible Sinks 639 3.97 0.88 0.55 Retention Ponds 1646 13.25 2.27 1.83 Undeveloped N/A 19.54 N/A 2.74 Combined Sinkholes 900 5.57 1.24 0.77 Table 3 Â– 2000 Aerial Photo Data. Tota l area, density, and percent land area of sinkholes, combined sinkholes possible sinkholes, exclusio n areas, and undeveloped land mapped using the 2000 air photos. 2000 Color Aerial Photo Descriptions The following photos and descriptions are a sa mple of 3 of the 2000 aerial photos. They are located in the northeastern, central, and southern parts of Pinellas County. The barrier islands were excluded in this comparis on due to the lack of karstic formations and sinkhole activity.
38 Figure 11. 2000 Sinkholes.
39 Figure 12. 2000 Storm Water Retention Areas.
40 Figure 13. 2000 Undeveloped Areas.
41 Northeastern Pinellas County Figure 14 is a 2000 aerial photo lo cated in northeastern Pinellas County and corresponds to the 1926 photo that was previously described in Figure 8. As can be seen in this photo this portion of Pinellas County is still partially undeveloped. There are large areas in this photo that are still natural we tland areas. Due to the colo rization of these photos the karstic features, including si nkholes can be identified by ve getation. As can be seen, there are several large storm wate r retention features in this photo. All of these retention features were former sinkholes as will be s een in the next chapter. The east-central region of this photo is scattered with vari ous sinkholes identified by a greenish-gray color. The rest of the photo is covered by urbanization, includi ng housing developments and golf courses. Central Pinellas County Figure 15 is a 2000 aerial photo lo cated in Central Pinellas County and corresponds to the 1926 aerial photo that was previously describe d in Figure 9. This photo is a very good example of what has become of Pinellas County by the year 2000. This area of Pinellas County has undergone rapid urbanization as can be seen when compared with Figure 9. This area would most likely be considered a bui ld out, with no areas left for development. As a result of the urbanization, most of the sinkholes in this area have been modified by being completely filled, or turned into stor m water retention areas. Most of the water features visible in this phot o are storm water retention areas There are few sinkholes left unmodified in this area. One such sinkhole is located in the northeastern corner of this
42 Figure 14. 2000 Aerial Photo Â– Northeastern Pinellas County
43 Figure 15. 2000 Aerial Photo Â– Central Pinellas County.
44 photo, and is approximately 15 meters in diameter It is possible that this sinkhole is also used for storm water retention but has retain ed a circular shape. There is another sinkhole that is visible in the southwestern po rtion of this photo. It is also surrounded by what appears to be a park. This sink seems to be a coalescing sink with a diameter of approximately 50 meters. It is possible that ot her smaller sinks exist in this photo but it is difficult to determine from an aerial photo. Southern Pinellas County Figure 16 is located in southern Pinellas County in the area that is known as St. Petersburg and corresponds to the 1926 aerial p hoto in Figure 10. As can be seen in this photo from 2000 the area continued to urba nize remaining land areas and is now a complete build out with no land left for any more development. Any other development activities in this area would be refurbishment or redevelopment of existing structures. In this photo there are very few water structures ot her than large inlets and marinas. In the south central portion of this photo there is a large trac t of land that has been specified as non-development due to wetland regulations. In this area, which can be identified by the red coloration, two large sinkhol es are visible by the vegetati on patterns. The larger of the sinks has a diameter of a pproximately 200 meters and is in the northwestern part of this undeveloped area. The other sink, which is located in the southe rn portion of this area, is approximately 100 meters in diameter. Other than these two sinks, no others are clearly visible. This area also has a lack of storm water retention ponds due to the rapid urbanization in the early 1900Â’s when regulations for inclusion of such structures with development did not exist. Other smaller sinkho les are possible in this area but are once again difficult to discern due to small size or lack of vegetation indi cator patterns.
45 Figure 16. 2000 Aerial Photo Â– Southern Pinellas County.
46 CHAPTER 4 DISCUSSION 1926 and 2000 Aerial Photo Comparison The following is a comparison of th e 1926 aerial photo and the 2000 aerial photo digitization. In ARCGIS th e intersect function was execu ted to intersect sinks and possible sinks from 1926 that still exist in 2000. Figure 17 illustra tes the comparison of possible sinks between 1926 and 2000. There is a total of 114 possibl e sinks present in both 1926 and 2000 with a total area of 0.46 km2. Figure 18 illustrates the comparison of sinks between 1926 and 2000. This allowed only si nks that were present in both 1926 and 2000 to be revealed. There is a total of 118 sinks presen t in both 1926 and 2000 with a total area of .43 km2. Figure 19 illustrates all possibl e sinks, and sinks, as well as the intersecting areas from 1926 and 2000 aerial phot os. Figure 20 illustrates intersecting areas of sinkholes (possibl e sinks and sinks) from 1926 a nd 2000 aerial photos. This map illustrates all sinkholes present in both 1926 and 2000. There is a total of 457 total sinkholes present in both 1926 and 2000 with a total combined area of 2.66 km2. Figure 21 illustrates sinkholes which most likely formed after the 1926 aerial photos were taken and now appear in the 2000 aerial photos. It appears as though approximately 400 sinkholes have formed since 1926. This could be a misinterpretation because this number seems abnormally high. Percent Loss Table 4 illustrates the percent loss of si nkholes between 1926 and 2000. The number of sinks loss to urbanizatio n by the year 2000 is 92%. A tota l of 84% of the possible sinks were lost over the same period. The to tal loss of sinkholes be tween 1926 and 2000 is
47 Figure 17. Possible Sinks present on both 1926 and 2000 aerial photos.
48 Figure 18. Sinks present on both 1926 and 2000 aerial photos.
49 Figure 19. Total sinkholes in both 1926 and 2000 aerial photos, with intersecting areas.
50 Figure 20. Total intersecting areas of sinkholes present in both 1926 & 2000 aerial photos.
51 Figure 21. Sinkholes which have formed since 1926, and appear on the 2000 aerial photos.
52 87% (Table 4). Undeveloped ar ea lost to urbanization by the ye ar 2000 is 97% (Table 4). As can be seen, there has been a dramatic loss in sinkholes and an explosion in development. Percent sinkholes remaining in 2000 8.27% Percent possible sinkholes remaining in 2000 16.16% Undeveloped area lost by 2000 96.80 Table 4. Percent of Sinkholes Lost by 2000. Sinkholes and possible sinkholes left in 2000. This figure was determined by comparing the overlap of sinkholes mapped in 1926 and sinkholes mapped in 2000. Also shown is rural area lost by 2000. Retention Areas The areas identified as retention ponds were areas on the 2000 aerial photos that appeared to be man made and functioned as storm water retention areas. It is likely that many of these storm water retention areas were former sinkholes with a possi ble connection to the aquifer underlying Pinellas County. When the intersect function was executed for the retention pond layer from the 2000 aerial photos and the 1926 sinks and possibl e sinks layers the map in Figure 22 was created. Figure 22 illustrates the areas whic h intersected. These areas may possibly be sinkholes that have now been converted into storm water retention areas. Approximately 21.76 km2 of retention pond features overlapped with 1926 sinks and possible sinks with a total of 499 sinkholes that have now b een formed into retention pond areas. Sinkhole Elevations Figure 23 illustrates sinkhole locations w ith the contour lines of Pinellas County. This map was created with the hopes of de fining a pattern of sinkhole development. Unfortunately, even after much scrutiny, th is map did not show a pattern of sinkhole development which related to elevation.
53 Figure 22. 1926 Sinkholes intersecti ng 2000 storm water retention ponds.
54 Figure 23. Topographic Map with Sinkhole Locations
55 CHAPTER 5 CONCLUSIONS AND RECCOMENDATIONS FOR FURTHER RESEARCH The results of this study have provide d new information on the sinkholes, past and present, as well as the effect s of urbanization on the karst landscape of Pinellas County, Florida. The conclusions reached from this research provide justification for further sinkhole research in Pinellas County, whic h will assist land planners and local governments with future development and redevelopment projects. Conclusions The mapping of sinkholes on historic 1926 aeria l photos within ArcGIS proved to be a useful exercise in determining the landscape that existed prior to the urbanization of Pinellas County. A total of 2,703 sinkholes we re identified on the 1926 aerial photos. These sinkholes accounted for 43.90 square kilo meters and 6.14% of the total land area. By 2000 only 900 sinkholes remained, with approximately 400 sinkholes forming since the 1926 aerial photos were taken. The sinkhol es identified on the 2000 aerial photos accounted for only 5.57 square kilometers a nd 1% of the total land area of Pinellas County. This accounted for a total loss of 87% of sinkholes between 1926 and 2000. It is likely that this lo ss is due to the rapid urbanizati on that occurred since 1926. Many of the existing sinkholes were likely filled, while others have been modified into storm water retention areas. The 2000 aerial photos of Pinellas Count y revealed 1,646 storm water retention areas. These areas cover 13.25 square kilomete rs and account for 2% of the total land area. It is probable that many of these we re former sinkholes and thus have a direct
56 connection to the aquifers located beneath Pinellas County. When 1926 sinkholes were intersected with 2000 retent ion ponds in ArcGIS, the sinkholes which have been modified in retention areas were reveal ed. A total of 499 of the 1,646 storm water retention ponds now present in Pinellas Count y were former sinkholes. They account for 21.76 square kilometers and 3% of the total land area. Suggestion for Further Research There are many opportunities for further research on the sinkholes in Pinellas County, as well as the state of Florid a. Some suggestions follow. Further research is need ed in Pinellas County to ve rify sinkholes and exact locations and sizes. This could be accomplis hed by on-site analysis or ground truthing. The sinkholes that appear to be present in both the 1926 aerial photos should be located, measured, and verified as sinkholes. It woul d be wise to verify the 1926 sinkholes that appear to be retention ponds on the 2000 aerial photos w ith the same methods listed previously. Dye traces in the 1926 sinkholes that are now retention ponds would also provide useful information. This would allow resear chers to determine if contaminated water from the storm water retention areas is flowi ng into the aquifer and at what velocity. This would help to understand the aquifer dynamics below Pinellas County as well as determine where and when pollution will show up. Pinellas County also has a collection of aerial photos taken in 1945. if these photoÂ’s were digitized within ArcGIS and th e sinkholes identified, th ey could then be compared to the sinkholes identified on 1926 and 2000 aerial photos. This would allow
57 researchers to see a progression of urbanizati on as well as possibly identify sinkholes that developed between 1926 and 2000. It would also be useful to conduct a histor ic air photo analysis of the entire state of Florida. A comprehensive inventory of the sinkholes in the state of Florida mapped during different time periods would greatly facilitate the understa nding of the stateÂ’s complex karst systems.
58 References Bates, Robert L., and Jacks on, Julia A., eds, 1980, Glossary of Geology, Second Edition. American Geological Institute: Falls Church Virginia, 337pp. Beck, B.F. and Sayed, S., 1991, The Sinkhole Ha zard in Pinellas County: A Geologic Summary for Planning Purposes, Pinella s County, Florida Report 90-91-1, 140pp. Bloom, A.L. 1998, Chapter 8 Â– Karst and Speleology. In : Bloom, A.L., Geomorphology: A Systematic Analysis of Late Cenozoic Landforms Prentice Hall: New Jersey, p. 147-165. Broska, J.C. and Barnette, H.L., 1999, Hydrogeology and Analysis of Aquifer Characteristics in West-Central Pinellas C ounty, Florida, U.S. Geological Survey Open File Report 99-185. 23pp. Cobb, J. and Currens, J. 2001. Karst: The Stealth Hazard. Geotimes: v. 46, p. 14-17. Coker, A.E., 1969, Application of Remote Se nsing to Occurrence of Collapse Sinkholes in Alafia and Peace River Basins, Florid a. National Aeronautics and Space Administration, Earth Resources Aircraft Program, Status Review Volume III, Hydrology, Oceanography and Sensor Studies. 15pp. Denizman, C., 1998, Evolution of the karst in the lower Suwannee River basin, Florida. University of Florida, Dissertation. 214pp. Frank, E.F. and Beck, B.F., 1991, An Analysis of the Cause of Subsidence Damage in the Dunedin, Florida area 1990/1991, Pine llas County, Florida Report 90-91-02, 59pp. Florida Geological Survey website: http://www.dep.state.fl.us/g eology/gisdatamaps/index.htm Taken on September, 9, 2003. Florida Department of Environmental Protection GeoData Di rectory website: http://www.dep.state.fl.us/gis/datadir.htm Taken on June 15, 2004. Florida Sinkhole Research Institute, 1983, Update: FSRI Ne wsletter, v. 1, no. 1., p. 1-2. Galloway D., Jones, D.R., Ingebritsen, S.E ., eds, 1999, Introduction, in Land Subsidence in the United States US Geologi cal Survey Circular 1182, p. 1-158. Heatherington, A.L. and Mueller, P.A., 1997, Ge ochemistry and Origin of the Florida Crustal Basement Terranes. In: Jones, D. S. and Randazzo, A.F., eds., The Geology of Florida. University Press of Florida: Gainesville, p. 27-37
59 Hutchinson, C.B., 1994. Incidents and Causes of Land Subsidence in the Karst of Florida, U.S. Geological Survey Open-File Report 94-532, pgs 12-13. Kochel, C.R., Miller, J.R., Ritter, D.F., 1995, Chapter 12 Â– Karst-Processes and Landforms. In: Kochel, C.R., Miller, J.R., Ritter, D.F., Process Geomorphology. Wm C. Brown Publishers: Dubuque, IA, p. 400 Â– 427. Lane, E., 1994, FloridaÂ’s Geol ogical History and Geologi cal Resources, Florida Geological Survey, Special Publication no 35, 64pp. Lane, E., 1986, Karst in Florida, Florida Ge ological Survey, Special Publication no 29, 100pp. Littlefield, J.R., 1988, Evaluation of Data for Sinkhole-Development Risk Models: West-Central Florida. University of South Florida, Thesis. 75 pp. Lyon, J.G., McCarthy, J., eds., 1995, Wetland and Environmental Applications of GIS. Lewis Publishers: Boca Raton, FL, p. 253 Â– 256. Miller, J.A., 1997, Hydroge ology of Florida. In: Jones, D.S. and Randazzo, A.F., eds., The Geology of Florida. University Pre ss of Florida: Ga inesville, p. 69-88. Randazzo, A.F., 1997, The Sedimentary Platform of Florida: Mesozoic to Cenozoic. In: Jones, D.S. and Randazzo, A.F., eds., The Ge ology of Florida. University Press of Florida: Gainesville, p. 39-56. Schmidt, W., 1997, Geomorphology and Physiography of Florida. In: Jones, D.S. and Randazzo, A.F., eds., The Geology of Florida. Un iversity Press of Florida: Gainesville, p. 1-12. Scott, T.M. 2002, FloridaÂ’s Springs in Jeopardy. Geotimes: v. 47, p. 16-20. Scott, T.M., Means, G.H., Means, R.C., Meegan, Rebecca P., 2002, First Magnitude Springs of Florida, Florida Geological Survey, Open File Report no. 85, 138 pp. Scott, T.M., 1997, Miocene to Holocene History of Florida. In: Jones, D.S. and Randazzo, A.F., eds., The Geology of Florida. Un iversity Press of Florida: Gainesville, p. 57-67. Scott, T.M., 1992, A Geological Overview of Florida, Florida Geological Survey, Open File Report No. 50, 78pp. Scott, T.M., 1988, The Lithostratigraphy of th e Hawthorn Group (Miocene) of Florida, Florida Geological Survey, Bulletin No 59. 138pp.
60 Sinclair, W.C., 1982, Sinkhole Development Resulting from Ground-Water Withdrawal in the Tampa Area, Florida, USGS Water Resources Investigations Report, 81-50, 19pp. Sinclair, W.C., Stewart, J.W., Knutilla, R. L., Gilboy, A.E., Miller, R.L., 1985, Types, Features, and Occurrence of Sinkholes in the Karst of We st-Central Florida, U.S. Geological Survey, Water-Resources Investigations Report 85-4126. 81pp. Smith, D.L. and Lord, K.M., 1997, Tectonic Evolution and Geophysics of the Florida Basement. In: Jones, D.S. and Randazzo, A.F., eds., The Geology of Florida. University Press of Florida: Gainesville, p. 13-26. Tihansky, A.B., 1999, Sinkholes, West-Central Flor ida, in Land Subsidence in the United States (Galloway D., Jones, D.R., Ingebr itsen, S.E., eds), U.S. Geological Survey Circular 1182, 177pp. Trommer, J.T., 1987, Potential for Pollution of the Upper Floridan Aquifer from Five Sinkholes and an Internally Drained Basin in West-Central Florid a, U.S. Geological Survey, Water-Resources Investigations Report 87-4013, 103pp. Upchurch, S.B. and Randazzo, A.F., 1997, Environmental Geology of Florida. In: Jones, D.S. and Randazzo, A.F., eds., The Ge ology of Florida. University Press of Florida: Gainesville, p. 217-249. White, W.A., 1970, The Geomorphology of the Flor ida Peninsula: Florida Bureau of Geology Bulletin 51, 164 pp. White, W.B., 1988, Geomorphology and Hydrology of Karst Terrains. Oxford University Press: New York, p. 103-118 Wilson, W.L., and Beck, B.F., 1988, Evalua ting Sinkhole Hazards in Mantled Karst Terrane, Proceedings of Geotechnical Asp ects of Karst Terranes GT Div/ASCE, Nashville, TN, May 10-11