Desert Potholes: Ephemeral Aquatic

Microsystems

MARJORIE A. CHAN1,*, KATRINA MOSER2, JIM M. DAVIS2,GORDON SOUTHAM3, KEBBI HUGHES3 and TIM GRAHAM4

1Department of Geology & Geophysics; 2Department of Geography, University of Utah, Salt Lake

City, UT 84112, USA; 3Department of Earth Sciences, The University of Western Ontario,

London, Ontario, Canada N6A 5B7; 4U.S. Geological Survey, 82 Dogwood Ave., Moab, UT 84532,

USA

(Received 1 December 2003; accepted 12 November 2004)

Abstract. An enigma of the Colorado Plateau high desert is the ‘‘pothole’’, which ranges from

shallow ephemeral puddles to deeply carved pools. The existence of prokaryotic to eukaryotic

organisms within these pools is largely controlled by the presence of collected rainwater. Mul-

tivariate statistical analysis of physical and chemical limnologic data variables measured from

potholes indicates spatial and temporal variations, particularly in water depth, manganese, iron,

nitrate and sulfate concentrations and salinity. Variation in water depth and salinity are likely

related to the amount of time since the last precipitation, whereas the other variables may be

related to redox potential. The spatial and temporal variations in water chemistry affect the

distribution of organisms, which must adapt to daily and seasonal extremes of fluctuating

temperature (0–60 �C), pH changes of as much as 5 units over 12 days, and desiccation. For

example, many species become dormant when potholes dry, in order to endure intense heat, UV

radiation, desiccation and freezing, only to flourish again upon rehydration. But the pothole

organisms also have a profound impact on the potholes. Through photosynthesis and respira-

tion, pothole organisms affect redox potential, and indirectly alter the water chemistry.

Laboratory examination of dried biofilm from the potholes revealed that within 2 weeks of

hydration, the surface of the desiccated, black biofilm became green from cyanobacterial growth,

which supported significant growth in heterotrophic bacterial populations. This complex biofilm

is persumably responsible for dissolving the cement between the sandstone grains, allowing the

potholes to enlarge, and for sealing the potholes, enabling them to retain water longer than the

surrounding sandstone. Despite the remarkable ability of life in potholes to persist, desert pot-

holes may be extremely sensitive to anthropogenic effects. The unique limnology and ecology of

Utah potholes holds great scientific value for understanding water–rock–biological interactions

with possible applications to life on other planetary bodies.

Key words: potholes, desert, Colorado Plateau, sandstone, Utah, microorganisms, redox, bio-

film, cyanobacteria, limnology

*Author for correspondence: E-mail: machan@mines.utah.edu

Aquatic Geochemistry (2005) 11:279–302 � Springer 2005

DOI 10.1007/s10498-004-6274-8

1. Introduction

Water-filled potholes and weathering pits have been used by humans andanimals for millennia (e.g., Bayly, 1999) and have been given a wide varietyof names such as ephemeral pools, cisterns, tanks, caldrons, water pockets,and tinajas (Netoff et al., 1995). Intermittent ponds have been reported on awide range of lithologies formed by a combination of physical, chemical, andbiological processes (e.g., Udden, 1925; Reed et al., 1963; Twidale, 1982;Birkeland et al., 1987).

Desert potholes are shallow puddles (cm-scale) to deep pools (dam-scale)that collect precipitation from their small catchments. Each is its own mini-ecosystem with complex connections and interrelationships between life–water–host rock. Life is based on the presence of water, and both likely playan important role in how potholes form, develop and sustain life in the hostrock substrate. In the desert environment, potholes are typically dry for partor all of the summer, and are only occasionally wet throughout the otherseasons of the year. This desert regime produces extremely variable envi-ronmental conditions, yet despite this extreme variability, life both exists andpersists in these pools.

Potholes have been recognized on a variety of substrates across the world(Williams, 1985). In southern Utah there is a diverse suite of desert potholesthat exhibit a number of physical forms, with different hydro-periods anddiversity of biota. Although these potholes have received modest acclaim inNational Park areas as well as in studies of Findley and Sisson (1975), Barnes(1994) and Graham (1999), much of the science of potholes has remainedunexplored. Giant weathering pits within the Middle Jurassic EntradaSandstone near Lake Powell, Utah were initially described by Netoff andShroba (1993) and Netoff et al. (1995).

The purpose of this paper is to describe the variety and extreme variationsin the environmental conditions (physical and chemical) of desert potholes inthe Moab area. As well, this paper investigates pothole biofilms, which arelikely an important component to the pothole development. This paper alsorecords organisms that have adapted to the variable conditions and considersthe link between water–rock–biology.

2. Materials and Methods

2.1. LOCALITY

Potholes occur throughout much of the Colorado Plateau that is centered onthe Four Corners area of Utah, Arizona, Colorado, and New Mexico. Forthis study, we focus on several Jurassic sandstone sites close to Moab, Utah(Figure 1) where there are a variety of different pothole shapes and sizes(Figure 2). The host rock for potholes is typically porous eolian sandstone.

280 MARJORIE A. CHAN ET AL.

Potholes are situated at three different location types: (1) at the top of theformation, where an overlying shale unit is preferentially weathered back toexpose the more resistant sandstone; (2) on exposed regional supersurfaces(Kocurek, 1996), and (3) along flat, exposed first-order bounding surfaceswhere horizontally laminated interdune deposits are common.

It is clear that the Utah potholes must be post-Jurassic (age of the hostrock) and postdate the joint and fracture patterns along which they form.Joint patterns are likely Cenozoic in age, and associated with shifting of saltwithin the Paradox Basin (e.g., Doelling, 1988) and uplift and exposure onthe Colorado Plateau in the last 6 million years or so (Hunt, 1969; Luchitta,1979; Huntoon, 1988). Pothole sites can vary from those that are relativelyisolated to those that are greatly impacted by humans and livestock.

2.2. SAMPLING AND FIELD ANALYSIS

Four different sites were selected for sampling to reflect the variety of pot-holes, and are informally named: (1) Sand Flats (SAFT); (2) Swiss Cheese

Figure 1. Locality map of selected primary study pothole sites (not inclusive) in south-

eastern Utah. Pothole Flat locality is hosted at the very top of the Jurassic Navajo

Sandstone and shows relatively simple pothole development. Sand Flat Recreation Area

locality is hosted in the middle portion of the Jurassic Navajo Sandstone, and includes

Swiss Cheese Ridge sample areas. Bartlett Wash locality is hosted in the Jurassic Curtis

Formation/Moab Tongue of the Entrada Sandstone (see Figure 3). Both the Sand Flat

Recreation Area and the Bartlett Wash localities show a range of relatively simple to

moderately complex pothole development stages (see Figure 2).

DESERT POTHOLES 281

Ridge (SWCR); (3) Bartlett Wash (BARW), and Pothole Point (PHPT). Onesite (�150 km south of Moab) fromMuley Point- Moki Dugway (MPMD) isalso included for comparison (Figures 1 and 2).

Water samples were collected from potholes from each of the four sites. Atotal of 39 measurements were taken representing 28 potholes (13 SWCR, 11SAFT, 11 BARW, 3 PHPT and 1 MPMD) during five field excursions (5/99,11/99, 2/00, 11/00, 4/01, 2/02). For a few potholes, measurements were takenin the same pothole over several years. For each pothole, specific conduc-tivity and temperature were measured using a YSI model 130 conductivitymeter and pH was measured using hand-held Hanna Field pH meter.Environmental variables were also recorded, including pothole size, shape(circular, ellipse, half-circles, and complex), depth (shallow, intermediate,deep), water depth and presence or absence of aquatic vascular vegetation.

Water depth is an estimate of the deepest part of the pothole and repre-sents an average of several measurements taken in the center of the pothole.Water samples for chemical analyses were collected from about 10 cm belowthe surface by reaching out approximately 50 cm from the edge of the pot-hole with pre-cleaned and pre-rinsed polyethylene bottles. These samples

Figure 2. Examples of potholes on relatively flat, exposed sandstone surfaces: (a) Shallow,

evenly spaced potholes, Pothole Point. (b) Deep weathering pits/potholes, Sand Flat

Recreation Area. (c) Simple pothole developmental pothole stage, Bartlett Wash Area. (d)

Moderately complex developmental pothole stage from the Bartlett Wash Area. (e) Joint-

aligned potholes, Sand Flat Recreation Area. (f) Small potholes with black biofilms, as a

function of water pouroffs and accumulations along the joint pattern, Sand Flat Recre-

ation Area. See Figure 1 for locations and corresponding host rock, and Figure 3 for the

stratigraphic units.

282 MARJORIE A. CHAN ET AL.

were filtered through a 0.45 lm filter in the field. Two water samples werecollected from each pothole; one for the measurement of anions and one forthe measurement of cations. Water samples were kept cool and in the darkuntil returned to the lab where concentrated nitric acid was added to thecation samples. Samples were shipped to the Department of Geology andGeophysics at the University of Minnesota and analyzed for Si, Fe, Mn, Ca,Mg, Na, K, F, Cl, NO2-N, NO3-N, Br, PO4 and SO4. The anion sampleswere run on a Dionex ion chromatography system with a GP40 gradientpump, a CD20 conductivity detector, and an AS40 autosampler. The systemuses a carbonate/bicarbonate eluent on an AS14 analytical column. Thecation samples are run on a Thermo Elemental PQ ExCell quadrupole basedinductively coupled plasma mass spectrometer (ICP-MS). Samples submittedprior to September 2000 were analyzed using a Perkin Elmer Elan 5000 ICP-MS. PO4 was removed from all analyses because all values were belowdetection limit.

A rock sample collected from Pothole Point (PHPT of Figures 1 and 2A,)was used for a laboratory model to examine the response of the black biofilmto the addition of water. Sediments ± hydration were also collected fromfive potholes near Pothole Point to determine the response of the aerobicheterotrophic population to hydration.

2.3. SCANNING ELECTRON MICROSCOPY (SEM)

The general structure of the black coating (not desert varnish) on the uppersurface of a pothole rock sample from Pothole Point was mounted on analuminum SEM stud via a carbon tack, then gold sputter coated andexamined by using scanning electron microscopy (SEM) (LEO 435VP) at 15–20 keV. An uncoated sample was also prepared for SEM–EDS (energy dis-persive spectroscopy) without gold coating to determine the elementalcomposition of the black coating and examined using a LEO 435VP/EDS(Quantum Kevex -3300) system.

2.4. BACTERIOLOGY

One cm2 samples of dried and laboratory hydrated biofilm (after 1 monthincubation at room temperature) were scraped using a 70% ethanol sterilizedspatula from the mineral surface, processed through 10-fold limiting dilu-tions in 0.85% (wt./vol.) saline solution and spread plated onto a R2A(Difco�). Plates were incubated at RT for 2 weeks before enumeration.Aerobic heterotrophs were also enumerated from five potholes before (i.e.,dehydrated material) and after field hydration in 2002. Briefly, sterile distilledwater was added to the pothole samples, which were allowed to incubate for2 weeks prior to sampling and analysis.

DESERT POTHOLES 283

2.5. CLASSIFICATION OF EUKARYOTIC ORGANISMS

An initial table of eukaryotic organisms that were identified by (co-author)Graham over a period of reconnaissance observations with supportingmicroscopic and SEM identifications since the 1980s, and compiled fromprevious government reports is provided (Table III). It would require sig-nificant man-hours and sustained funding to do more long-term monitoringto establish the detailed cycles and controls on eukaryotic organisms.However, the table shows both the variety and diversity of organisms.

2.6. STATISTICAL ANALYSIS

Principal components analysis (PCA) was used to study the relationships be-tween different environmental variables and to compare chemical conditionsbetweenpotholes fromdifferent locations andwithin a single pothole over time.Any chemical variables that were measured in amounts above detection limitwere included in the PCA, and depth, pothole shape and vegetation were in-cluded as passive variables. PCA was done using the computer program CA-NOCO, version 4.02 for Windows (ter Braak and Smilauer, 1997).

3. Results and Discussion

3.1. PHYSICAL AND CHEMICAL LIMNOLOGICAL PROPERTIES

The Sand Flats (SAFT) and Swiss Cheese Ridge (SWCR) sites are approx-imately within the lower half of the Lower Jurassic Navajo Sandstone(Figure 3). The Pothole Point (PHPT) locality is at the top of the JurassicNavajo Sandstone and the Bartlett Wash (BARW) locality occurs within theMiddle Jurassic Curtis Formation/Moab Tongue. The Muley Point- MokiDugway (MPMD) is located within the Permian Cedar Mesa Sandstone.Field mapping (Doelling, 1993, 2001) and detailed stratigraphy (e.g., Peter-son and Pipringos, 1979; Peterson, 1994) provide a good framework for hostrock studies. These eolian sandstones are well-sorted quartz arenites withhigh porosity and permeability and are weakly cemented with silica (quartzovergrowths), clays, iron (hematite), and carbonate (calcite) (e.g., Chanet al., 2000, Figure 4).

Netoff et al. (1995) proposed a classification of different pit types based ontheir cross-sectional form: cylinders, bowls, armchairs, and pans. Alternativeclassification schemes have focused on width to depth ratios to characterizepits [see summary in Netoff et al., 1995]. Some potholes, where large piles ofwind-blown sand can be moved from one side of a large, deep pothole to theexact opposite side in just 1 year, can be clearly related to wind processes(Netoff, 2000, personal communication). However, the development of otherpotholes may show a strong tie to microbial and higher life forms that are

284 MARJORIE A. CHAN ET AL.

sustained because of accumulated water. The enclosed geometries of thepotholes suggest that these ‘‘closed systems’’ were mainly formed by acombination of wind abrasion and erosion, and perhaps biogeochemicaldissolution (Kurtz and Netoff, 2001). Other factors observed in this studysuch as ice, microclimate variations, biochemical processes appear to haveadditional effects on pothole formation.

Figure 4. Photomicrograph of the Jurassic Navajo Sandstone eolian host rock

(well-sorted quartz arenite) in (a) plane light and (b) crossed polarized light. Note: light

areas in (a) between grains is mostly porosity (epoxy). View width of each image is �1 mm

across.

Figure 3. Stratigraphic units of the Moab area, with indications (bold lettering) of where

potholes are common. Modified from Doelling (2000).

DESERT POTHOLES 285

The potholes from each of the four main study sites have different mor-phologies, which likely indicate differences in origin and stage of develop-ment. In the case of PHPT, small shallow pools form where water collects onthe exposed gently-dipping sandstone, but oddly, many of the shallow pot-holes seem to have some optimum spacing even in the absence of any jointsor fractures in the rock (Figure 2A). The potholes at SAFT and SWCR areperched on topographic highs of sandstone ridges and are deeply carved andoften cylindrical in form (Figure 2B), whereas the potholes at BARW are acombination of deep and shallow potholes with simple (Figure 2C) to com-plex (Figure 2D) shapes. For many of the potholes at SAFT, SWCR andBARW there is clear joint control (matching regional mapped joints) on thedevelopment of potholes, where the potholes are preferentially aligned alongthe fractures, which likely focus water flow and accumulation (Figure 2E).Still other potholes, are formed in a connected chain from water pouroffs(Figure 2F).

All of the PHPT potholes are relatively small in area (�1–64 m2) and haveshallow water depths, whereas the deepest pools are at SAFT and SWCR(Table I). The amount of aquatic vegetation also varies considerably betweenpotholes. At PHPT there is insufficient water and sediment to supportaquatic plants, and at SWCR and SAFT the steep sides of the potholes maylimit aquatic plant growth. However, the potholes with intermediate depths,found both at SAFT, SWCR and BARW can contain considerable aquaticvegetation.

Summary statistics for water variables are provided in Table I. Surfacetemperatures of dry potholes can get as high as 60 �C (Graham, 1999). Thehighest water temperature we noted was 34 �C in May of 2003, and watertemperatures of 35 �C have been recorded in this same area in July (Schol-nick, 1994). Temperature varied from the top to the bottom of one pothole(99–11) by 5.3 �C (19.9 �C at the top to 14.6 �C at the bottom). Tempera-tures have also been noted to fluctuate diurnally by as much as 17.5 �C,which is similar to results reported by Scholnick (1994). We have observed icepresent between the winter months of November and February, and ofparticular interest is that at times some of the potholes freeze to the bottom.Some potholes experience repeated freeze/thaw cycles – melting occurs at theedges during the day and refreezes at night. This freeze/thaw cycling couldcontribute to weathering and pothole growth (Alexandrowicz, 1989).

Similar to temperature, pH varies seasonally (pH 7.4–10.1) and alsodiurnally (ranging from pH 7 at night to pH 10 in the day; Graham, 1999).The changes in pH are a function of cyanobacterial (see below) and otheralgal (photosynthetic) activity, in which CO2 is ‘‘fixed’’ from HCO3

)

releasing a hydroxyl ion (Thompson and Ferris, 1990; Scholnick, 1994). Inaddition, heterotrophic respiration, which is relatively more dominant atnight, produces organic acids that also reduce the pH (in combination with

286 MARJORIE A. CHAN ET AL.

Table

I.Summary

pothole

data.Themean,maxim

um,minim

um

andstandard

deviationforeach

site

andforallmeasurements

foreach

chem

ical

variable

isreported.Allpotholestotaln¼

39

pH

Tem

p

�CSpCond

(lScm

)1)

Si

(mgl)1)

Fe

(lgl)1)

Mn

(lgl)1)

Ca

(mgl)1)

Mg

(mgl)1)

Na

(mgl)1)

K (mgl)1)

F (lgl)1)

Cl

(mgl)1)

NO

2-N

(lgl)1)

NO

3-N

(lgl)1)

Br

(mgl)1)

SO

4

(mgl)1)

Depth

(cm)

MEAN

SAFT

8.8

8.4

126

0.9

114

50

20.6

1.5

2.1

1.7

110

3.3

39

412

11.7

6.4

51

SWCR

8.5

8.6

116

0.3

21

520.1

2.3

1.2

1.2

131

1.9

19

149

14.4

4.5

34

BARW

8.7

14.9

171

0.4

43

41

31.0

2.4

1.4

1.8

94

2.5

33

650

21.8

11.1

27

PHPT

8.0

12.0

74

0.5

24

10.5

1.0

0.7

0.9

91.8

33

2312

11.0

4.6

7

All

Potholes

8.6

10.8

135

0.6

52

27

23.4

2.0

1.5

1.5

134

2.5

29

527

15.1

7.1

34

MAX

SAFT

10.1

15.9

219

2.6

328

181

34.2

2.2

4.4

2.6

480

6.3

170

3284

27.0

16.4

100

SWCR

9.5

22.0

227

1.6

73

28

38.2

4.8

2.2

2.7

860

3.8

165

717

33.5

8.3

55

BARW

9.4

24.2

266

2.0

382

344

51.5

3.4

3.4

4.3

234

5.5

176

2901

59.5

29.0

52

PHPT

8.3

13.3

85

0.8

46

12.4

1.2

1.0

1.1

12

2.7

41

2810

17.0

5.2

11

All

Potholes

10.1

24.2

266

2.6

382

344

51.5

4.8

4.4

4.3

860

6.3

176

3284

59.5

29.0

100

MIN

SAFT

7.4

2.2

53

0.1

11

7.8

0.8

0.4

0.7

00.3

00

0.0

0.4

6

SWCR

8.0

3.0

35

0.1

00

7.3

0.7

0.5

0.2

00.0

00

0.0

0.0

7

BARW

7.7

6.6

70

0.0

00

20.5

1.6

0.5

0.0

12

0.4

00

0.0

1.3

1

PHPT

7.7

10.0

62

0.2

01

8.8

0.9

0.5

0.7

61.0

24

1535

0.0

4.2

3

All

Potholes

7.4

2.2

35

0.0

00

7.3

0.7

0.4

0.0

00.0

00

0.0

0.0

1

STDV

SAFT

0.7

5.1

49

0.8

119

52

6.9

0.4

1.5

0.7

129

2.0

48

973

11.4

4.5

40

SWCR

0.5

5.5

58

0.4

19

710.7

1.2

0.5

0.8

224

1.3

44

239

12.6

2.1

18

BARW

0.6

7.1

54

0.7

113

101

9.3

0.5

0.9

1.0

64

1.5

57

1123

19.0

9.2

20

PHPT

0.3

1.7

11

0.3

22

1.8

0.2

0.3

0.2

30.9

9682

9.5

0.5

4

All Potholes

0.6

6.3

62

0.7

94

62

11.2

0.9

1.0

0.9

242

1.6

47

969

14.4

6.1

28

DESERT POTHOLES 287

the CO2) to achieve low pH values recorded in the early morning. A key todiurnal cycling is the oxygen, where high O2 during the day promotes aerobicrespiration, i.e., CO2 production. Low/no O2 during the night causes biofilmsto become anaerobic, resulting in the formation of organic acids via anaer-obic respiration (Ramsing et al., 1993). Variations in other chemicals relatedto oxygen availability, are therefore, indirectly affected by cyanobacteria andalgal activity.

A PCA plot was constructed to explore the relationships between chemicalvariables and to examine differences between the four sites (Figure 5). Thetotal percentage of the variance explained by the first four axes was 79%,with eigenvalues of k1 ¼ 0.29, k2 ¼ 0.27, k3 ¼ 0.12 and k4 ¼ 0.11. Because thethird and fourth eigenvalues are relatively low, further discussion is limited tothe first and second axes. Based on the plot (Figure 5A) and componentloadings (Table II), several salinity-related variables, including specific con-ductivity, Ca, Na, K, Cl, and SO4 are important in determining the first axis.Mean values for all major ions (Table I), except for Ca, are below valuestypically reported for lakes and rivers (Wetzel, 2001). The dilute nature ofpotholes is the result of a combination of resistant bedrock, negligible soilcover and little atmospheric fallout (Gladney et al., 1993). Calcium values arehigher than other ions because it likely originates from calcite cement in thesandstone, which would be easily weathered.

Salinity varies temporally (see pothole 99-2 and 99-7 in Figure 5B),however, despite high evaporation rates, changes in salinity in potholes aresmall (in 99-2, specific conductivity ranges from 95.5 to 218.9 lS cm)1 and in99-7 from 124.6 to 265.5 lS cm)1). These findings are in agreement withprevious suggestions that total salt content of freshwater ephemeral poolswould be low throughout the year as a result of low initial salt concentrations(Horne, 1967; Williams, 1985; Scholnick, 1994). Several studies have sug-gested that salt weathering could play a role in pothole formation (e.g.,Netoff et al., 1995), however, our data suggest that this is unlikely. Meanvalues and standard deviations for specific conductivity and major ions aresimilar in SAFT, SWCR and BARW (Table I). PHPT potholes are slightlymore dilute than the other potholes because these potholes have shorterhydro-periods and only contain water immediately after a rainfall when ionswould be least concentrated. Rainwater measurements from nearby Can-yonlands (38E 27.30N, 109E 49.3020W) and Green River (38E59.90N,110E9.90W) have low values of all cations (1–3 orders of magnitude less thanthe values in the potholes) [National Atmospheric Deposition Program(NADP), 2004].

Based on the plot (Figure 5A) and the component loadings (Table II),Mn, Fe, NO3-N, Br, and SO4 are important in determining the second axis,although the values of Br are always low and frequently below detection. Themean values of Fe and Mn for SAFT, SWCR and BARW are typical of

288 MARJORIE A. CHAN ET AL.

Figure 5. PCA plots of water chemical variables illustrate (a) the relationships between

chemical variables measured at the potholes and (b) pothole similarities. In plot (a) each

water chemical variable is plotted as an arrow, such that the length of the projection onto

an axis is directly related to the importance of that variable in determining that axis.

Arrows of variables that are positively correlated are plotted such that the angle between

the arrows is small, and variables that are negatively correlated are plotted in opposite

directions. Variables that are not correlated plot at 90 degrees to one another. Plot b is

constructed such that the distance between sites approximates the Euclidean distance

between the sites in n-dimensional space (ter Braack, 1995). Therefore, the closer together

two sites are the more similar the sites are in terms of water chemistry. Site names are

provided on the figure (also see Figure 1). The last four numbers of each site name are the

date the sample was taken, whereas the first three numbers are the pothole number and the

first letter refers to the site (P ¼ Pothole Point; S ¼ Sand Flats; W ¼ Swiss Cheese Ridge;

B ¼ Bartlett Wash and M ¼Muley Point Moki Dugway).

DESERT POTHOLES 289

freshwater (Wetzel, 2001). The low values at PHPT may be due to the factthat they only contain water for short intervals (days) following rain events.The mean values for SO4 and NO3-N are also generally typical of freshwatersystems (Wetzel 2001). Potholes at PHPT (mean ¼ 2.3 mg L)1), B0060200(2.9 mg L)1), B1060401 (2.6 mg L)1) and S2010202 (3.3 mg L)1) have rela-tively high values of NO3-N, and are also shallow (<17 cm). The concen-tration of NO3-N in rainwater has been measured at Canyonlands and GreenRiver and values of between 0.6 and 1.99 mg L)1 were reported for winter(12/03-02/25) and spring (02/25-05/27), when the highest values of NO3-Nare recorded in our study (NADP, 2004). Therefore, variations in the con-centration of NO3-N cannot be explained by rainwater alone. The highvalues of NO3-N could be the result of greater fixation by cyanobacteriamats, which line the potholes (see below). In these shallow potholes the matsare only partially covered by sediments, whereas at the other sites greatersediment accumulation would limit light penetration, and therefore cyano-bacterial growth, including nitrogen fixation, and any subsequent N-cyclingprocesses, i.e., nitrification.

The second axis appears to be related to redox potential, although thisneeds to be tested further. In the PCA plot, Fe and Mn are positively cor-related, whereas SO4 and NO3-N are negatively correlated to axis 2(Figure 5A, Table II). Mn and Fe will increase, whereas SO4 and NO3-N willdecrease as conditions become more oxygen-deficient (e.g., Moser et al.,2002). Photosynthetic and metabolic processes can cause large seasonal anddiurnal fluctuations in oxygen in small ephemeral pools (Hamer and

Table II. Component Loadings

Loading

Component 1

Cl 0.841

K 0.823

SpCond 0.805

Na 0.770

Ca 0.766

SO4 0.579

Component 2

NO3-N )0.710Depth H2O 0.697

Fe 0.628

Br )0.595Mn 0.594

SO4 )0.576

290 MARJORIE A. CHAN ET AL.

Appleton, 1991; Scholnick, 1994). Oxygen measurements taken over a 24-hperiod at one pothole, 99-2, showed that at night and in the early morningoxygen becomes low, but during the late morning to early evening oxygenvalues are high (Figure 6). Similarly, Scholnick (1994) found that during theday potholes in the Moab area were hyperoxic (240 Torr) and during thenight were hypoxic (40 Torr). Scholnick’s (1994) study also showed thatoxygen concentrations varied seasonally, with oxygen decreasing with timeafter filling events.

The PCA plot and results from a Pearson correlation table (results notshown) indicate that Fe is significantly (p £ 0.05) positively correlated andNO3-N and SO4 are significantly (p £ 0.05) negatively correlated with waterdepth. It would be expected that potholes with greater surface areas relativeto depth would have enhanced diffusion between the atmosphere and water,and therefore oxygen fluctuations would be limited in deeper potholes(Scholnick, 1994). The deep potholes of SAFT are more protected from wind,which would reduce mixing and limit oxygen availability to the bottom of thepotholes, perhaps explaining the higher values of Mn and Fe at these sites.

Figure 5B illustrates individual potholes on the PCA plot – the closertogether two points are the more similar the water chemistry is between thetwo potholes. Potholes from the same site do not necessarily group together,although the three PHPT sites plot together in the lower left-hand quadrant.The location of these sites on the plot indicates the elevated NO3-N and low

Figure 6. Diurnal fluctuations of oxygen measured just below the water surface and at the

bottom of one of the SAFT potholes on June 1, 2004.

DESERT POTHOLES 291

Fe and Mn values in these potholes. Two BARW potholes, 2-10 (B2100202)and 1-06 (B1060401), plot in the lower right-hand quadrant indicating ele-vated SO4 values. It is not clear why these potholes have elevated values, butsome of this may be a function of proximity to mineralization along theMoab fault (Garden et al., 2001). The position of pothole 99-7 (B9970401)on the plot is characterized by high values of Fe (381.8 lg/L) and Mn(344.4 lg/L). It is clear from the plot, however, that the water chemistry of99-7 varies considerably overtime, and these elevated values probably cor-respond to a period of low oxygen.

The Muley Point-Moki Dugway pothole, plotted as the solid black box onthe far right of the diagram, is similar in terms of water chemistry to theMoab potholes. The main difference is that it contains slightly higher salinityvalues, which may be the result of its less protected location, which wouldlead to greater inputs of atmospheric dust. Finally, it is clear that the tem-poral variations in a single pothole (e.g., 99-2 and 99-7) are as great aspotholes from different sites.

3.2. MICROBIAL COMMUNITIES

The black coatings, which commonly line the bottom of early stage potholesor appear as a ‘‘high-water’’ ring around later stage potholes, are composedof desiccated microorganisms (Figure 7). The black coatings may also coatthe tops of small flat pebbles to cobbles within the potholes. These desiccatedmicroorganism films only produced a carbon signal using SEM/EDS, dem-onstrating that the black coloration is not due to the presence of secondaryminerals (e.g., manganese oxides – data not shown). The addition of water tothese desiccated biofilms promoted the growth of filamentous cyanobacteria(Figure 8), which function as the primary producers in early stage potholeenvironments.

Ortega-Calvo et al. (1995) suggest that the ability of terrestrial bacteria toretain water through the production of ‘hygroscopic extracellular polymericsubstances’, i.e., capsule, causes large volume changes during drying andwetting cycles that dislodge grains and magnify the weathering effects offreeze–thaw cycles. Also, at near neutral pH, Vandervivere et al. (1994)found that chelators, e.g., organic acids, are the most important factors inmineral dissolution. This is particularly relevant to the peculiarities of thesepotholes, where the levels of organic acids in these restricted environmentswill build up and become concentrated at the mineral interface due to des-iccation and ‘‘sealing’’ of the pothole by the biofilm. The lower pH measuredat night must be the result of anaerobic respiration and formation of organicacids because the pH is not acidic during the day, when aerobic respirationand CO2 formation would be most active. Mineral dissolution in the Moabpothole areas is facilitated because of the weak to moderate silica, clays, iron,

292 MARJORIE A. CHAN ET AL.

and carbonate cementation of the host rock, which presumably accounts forthe observed preferential weathering patterns. Silica (both as grains andcement) is susceptible to weathering as bacteria attempt to concentrateelements limited in their environment, e.g., iron (Adams et al., 1992) andphosphate (Banerjee et al., 2000), which is limited at some of these sites (seeabove). Also, Bennett et al. (1996, 2001) proposed that microbial weatheringof quartz grains could be due to their need for micronutrients, e.g., Fe, Co,Zn, Mo, Cu, or Ni, which are represented in trace amounts in the sandstone(data not shown). Microbial mats (Stal, 1995) and endolithic microbialcommunities are common in hot and cold desert environments (Friedmann,1980, 1982). Clearly, life is sustained in the potholes, and these biofilms showa very intimate relationship with the substrate and seem to be a precursor to

Figure 7. Scanning electron microscopy (SEM) photomicrographs of dried black biofilm

prior to any hydration. The surface of the desiccated biofilm possessed a bacteria-scale,

honeycomb architecture (a), which looked as if an organic blanket had been ‘‘tossed’’ over

the microorganisms (b). This coating presumably contributes to survival during desicca-

tion and to the capture of water after a rain event. Bars equal 10 lm.

DESERT POTHOLES 293

developing the more complex biotic communities that exist in the largerpotholes (Table III).

Dried sediments collected from the bottom of later-stage potholespossessed 1.5 · 106 (cfu/g) aerobic heterotrophs. After hydration, the aver-age population of aerobic heterotrophs increased almost two orders ofmagnitude to 1.2 · 108 cfu/g. This increase in heterotrophic plate countspresumably results from the input of organic carbon from the primaryproducers and perhaps from recycling of organic waste produced by thehigher organisms in these ecosystems. Regardless, these bacterial populationscan in turn support the grazing populations even within small potholeenvironments (e.g., snails at PHPT).

Although bacteria are undoubtedly important in the weathering process,other biological agents such as lichens also play a roll in mineral weathering.Lichens are visibly present on rock surfaces throughout the study area (de-tailed data beyond the scope of this paper) forming roughly circularweathering features up to 15 cm in diameter. Lichens are known to be able toweather the sediments of many rock types due to the production of signifi-cant amounts of oxalic acid (Edwards et al., 1992). Oxalic acid can mobilize

Figure 8. Scanning electron microscopy (SEM) photomicrographs of the black biofilm

after laboratory hydration, i.e., biofilm growth. Upon adding water, the organic blanket

breaks down (a and b) or is perhaps, consumed by the microorganisms, leading to the

development of a biofilm composed of filamentous cyanobacteria and their associated

heterotrophs that are usually seen in association with the filamentous organisms (c and d).

Note: both lower images are higher magnifications of the upper images. Bars equal 50 lm(a and c) and 5 lm (b and d).

294 MARJORIE A. CHAN ET AL.

potassium (Russell et al., 1998), which is limited in the rocks of our study dueto the purity of the eolian sandstones. Weathering in hot and cold deserts canalso be conducted by fungi not in symbiotic lichen association (Friedman,1982). However, our SEM observations agree with Bell et al. (1986) whosuggested that fungi do not play as important a role on the Colorado Plateauas they do in other hot and cold deserts because the rainfall is slightly greater,causing increasing competition between organisms.

3.3. HIGHER EUKARYOTES

The diversity of life found in potholes is considerable (Table III). As potholesenlarge and evolve, they contain increasingly complex ‘‘microcosms’’ of life,ranging from prokaryotic to eukaryotic organisms (including a variety ofalgae, such as diatoms and chrysopytes, and protozoans, such as testateamoebae (subphylum: Sarcodina; superclass: Rhizopoda). Pothole commu-nities differ from pool to pool (morphologies, parent rock, setting, etc.), fromone seasonal filling to the next. Interestingly, two potholes that appearsimilar in terms of physical and chemical properties can be comprised ofentirely different communities. For example, there are pools that consistentlyhave tadpoles adjacent to others that never have tadpoles.

Table III. A general list of prokaryotic and eukaryotic organisms present in potholes from the

Moab, Utah area

Microscopic

Life

cyanobacteia, aerobic heterotrophs, lichens, algae, cryptobiotic ‘soils’

Aquatic Life aquatic mites – orbatid mite (Aquanothrus sp.)

Tardigrades (water bears), Rotifers, Nematodes

Plant Life Graminoids (foxtail chess – Sitanion hystrix, sand dropseed – Sporobolus

cryptandrus, Indian rice grass – Stipa hymenoides, blue grama – Bouteloua

gracilis, cattails (Typha latifolia)

Insect Life beetles (e.g., Dytiscidae, Hydrophilidae), backswimmers (Notonectidae),

water boatmen (Corixidae), mosquitoes (Culicidae), gnats (Ceratopogonidae),

midges (Chironomidae)

Other Branchiopod Crustaceans

Invertebrates fairy shrimp – Branchinecta packardi, Streptocephalus texanus,

Thamnocephalus platyurus), clam shrimp (Leptestheria compleximanus,

Eulimnadia inflecta), tadpole shrimp (Triops longicaudatus)

Ostracods – Cypriconcha sp. and Cypridopsis sp.

Water Sanils – Fossaria sp.

Vertebrates Spadefoot (Spea intermountanus and S. multiplicatus) and red spotted (Bufo

punctatus) toads, tiger salamander (Ambystoma tigrinum)

DESERT POTHOLES 295

Dodson (1987) hypothesized that predation was an important factorstructuring pothole communities. The role of predation may not be as clearas Dodson (1987) proposed (T.B. Graham, unpublished data), howeverpredation may influence community composition of potholes under certaincircumstances. Backswimmers (Belk and Cole, 1975) eat fairy shrimp(Woodward and Kiesecker, 1994) and tadpole shrimp, and tadpole shrimp inturn eat fairy shrimp. These interactions are probably important in deter-mining which species are consistently found in a pool, and whether thespecies composition remains fairly constant during a particular wet phase.Changes in species composition in a pool can have ramifications on othercomponents of the ecosystem, e.g., if fairy shrimp feed on algal species dif-ferentially, predation that reduces fairy shrimp numbers significantly mayalter algal community structure; predation on tadpole shrimp could reducethe amount of bioturbation, altering the water chemistry of the pool. Tad-poles can also alter the environment of pools by stirring up sediment,selectively eating algae, and recycling nutrients (Seale, 1980).

3.4. POTHOLE DEVELOPMENTAL STAGES

The photosynthetic microbial mats in Moab presumably function as a geo-biological agent, which modifies its environment. These contemporary mats/ecosystems also demonstrate a successful approach for the development ofimproved growth conditions within an extreme environment. A photosyn-thetic microbial mat is presumably the dominant ‘organism’ present duringthe initial stage(s) of pothole formation. The subsequent persistence of anindividual mat and development of a pothole is dependent on the efficientcoordination of resources by the diverse microbial populations that comprisethe mat.

For the purpose of this paper, we recognize two end members of devel-opmental stages:1. Early, simple pothole stages are relatively small and shallow and havemicrobial black mat films (Figure 2A, C).

2. More developed deeper, complex pothole stages, which generally sustainwater longer (throughout the year) and have a variety of life forms(Figure 2B, D, E).

Very small potholes only contain water for brief periods of time, whereasdeeper potholes can maintain water for greater lengths and sometimes evenperennially during average to wet years. Larger potholes tend to containgreater habitat diversity (they have sediment bottoms, rock walls, sometimesaquatic plants), and therefore, meet the requirements of a greater number ofspecies. Reduced light in very deep potholes, however, limits habitat avail-ability, and therefore, species diversity.

296 MARJORIE A. CHAN ET AL.

A specific corollary to our hypothesis is that microbial mats help ‘‘seal’’potholes, which helps to maintain water within them. The black surface ofthe potholes appears to glue the sand grains together (Figure 6) and likelyfunctions to help hold water, preventing desiccation of the microbial mat byslowing the percolation of water through the porous sandstone matrix. Therole of the microorganisms in sealing the potholes is in direct contrast to therole of microorganisms in catalyzing the weathering reactions responsible forpothole growth. This dichotomy in function may be reflected in the diversityof microorganisms occurring across the weathering profile, but this wouldrequire further study.

3.5. COMMUNITY LIFE IN EXTREME ENVIRONMENTS

In the initial, simple pothole stages, higher organisms are unable to competewith the microorganisms because the environment is too harsh (Friedmannand Ocampo, 1976). The cyanobacteria found in the microbial mat associ-ated with the shallow potholes are presumably the most competitive primaryproducers, i.e., the environment is sufficiently extreme to prevent algae andhigher plants from taking over as the dominant primary producers. Theextreme nature of these systems also keeps the populations of grazingorganisms low. However, over time, the continued successful colonization ofthese systems by microorganisms, i.e., the deepening in these shallow (cm-deep) potholes, reduces the environmental extremes allowing higher organ-isms (Table III) to compete with the microbial mat. It is only with continueddeepening of these potholes, causing a reduction in radiant and water stress,that a greater diversity of plants and animals are able to successfully colonizethese environments.

In order to survive in the desert environment, biological communitieslikely take one of several adaptive strategies (Graham 1999):1. Cryptobiosis (anhydrobiosis-life without water), where organisms toleratedesiccation while still maintaining and storing biologically active sub-stances (Crowe et al., 1978; Madin et al., 1978; Crowe, 1991). Organismsshut down metabolic activity (still maintaining the metabolic essence)during dormancy, and then later come to life upon rehydration.

2. Dessication resistance (‘‘Tupperware’’ strategy), where organisms sealmoisture inside the body and become dormant.

3. Escape, where an organism may reach a certain stage where it can leave(e.g., insects that develop from the larval stage and can eventually flyaway).

Cryptobiotic organisms are capable of surviving long periods withoutwater as well as other extreme environmental conditions. When dry, pot-hole sediments and the organisms in them are exposed to high (>60 �C)

DESERT POTHOLES 297

and low (<)20 �C) temperatures and increased ultraviolet radiation. Inthe dormant, cryptobiotic state, these organisms can survive these envi-ronmental stressors as they wait for rain to fall again. In fact, some ofthese organisms have been shown to be able to tolerate conditions theyhave never experienced in their evolutionary history. Rotifers have beenrevived after exposure to extreme temperatures: )273.15 �C (almost abso-lute zero, )459.67 �F), to 150 �C (300 �F) (Crowe, 1991). Some tardigradescan withstand an X-ray dose of 570,000 roentgens; 500 roentgens will killmost humans (Crowe, 1991). Brine shrimp eggs have been glued on theoutside of spacecraft and flown in the vacuum of outer space, thenretrieved and hatched (Planel et al., 1980; Gaubin et al., 1996).

4. Summary

Desert potholes are remarkable ecosystems, and provide us with the oppor-tunity to explore relationships of water, rock, and life in small aquatic envi-ronments. Moab potholes occur at a variety of scales from centimeters todecameters on eolian sandstones that form homogeneous, porous host rock.Despite the similarities in host rock, the limnology of the potholes shows bothspatial and temporal variation. Individual potholes experience widely fluctu-ating temperature, pH, oxygen, anion and cation variations, as well asmoisture-desiccation. Some of these extreme variations (i.e., oxygen and pH)are the result of the microorganisms living in the potholes, whereas others arethe result of the harsh climate of the desert environment. Despite the extremelyvariable environment of the potholes, all potholes, regardless of size, appear tohave biofilms supported by cyanobacteria and containing bacteria. Thecyanobacteria presumably serve as the primary producers supporting theecosystem whereas the heterotrophs are likely doing most of the mineralweathering, which helps create space for the Photoautotrophs. More complexpotholes have a greater diversity of organisms including aquatic life, plant life,insect life, invertebrates, and vertebrates. Many of the organisms have adap-tive strategies (e.g., cryptobiotic) to survive in the extreme conditions.Whetherthese systems can endure anthropogenic effects (e.g., cattle, atmosphericpollution) or other disturbances has yet to be determined.

Questions regarding pothole formation, development, and ecosystems aremultifaceted. Past studies have largely focused on single variables or smallgroups of variables, but the study presented here attempts a more compre-hensive approach that examines chemical, physical, and biological features,and interrelationships between them. Future studies will help discern theinteractions of environmental controls and the development of life in desertaquatic settings, as well as applications to other comparable systems.

298 MARJORIE A. CHAN ET AL.

The search for biosignatures and extraterrestrial life on other planets hasrelied heavily on understanding potential analogs here on Earth (e.g.,Banfield et al., 2001; Wierzchos and Ascaso, 2002). These potholes can serveas a springboard to understanding the relationships, structure, organization,and responses of microbial life in small aquatic systems. As new remotesensing techniques or mechanical life-detection devices (e.g., those used on aMars Exploratory Rover) are developed, potholes can serve as good analo-gies of meter-scale biomarkers that might be detectable on other planetarybodies.

5. Acknowledgements

This research was partially supported by a University of Utah MineralLeasing Funds grant. We gratefully acknowledge anonymous reviewers fortheir input on this manuscript, and William T. Parry for his assistance insome figure preparations.

References

Adams J. B., Palmer F. and Staley J. T. (1992) Rock weathering in deserts: mobilization and

concentration of ferric iron by microorganisms. Geomicrobiol. J. 10, 99–114.

Alexandrowicz, Z. (1989) Evolution of weathering pits on sandstone tors in the Polish Carpa-

thians. Zeitschrift fur Geomorphologie 33, 275–289.

Banerjee M., Whitton B. A. and Wynn-Williams D. D. (2000) Phosphatase activities of endo-

lithic communities in rocks of the Antarctic dry valleys. Microb. Ecol., 39, 80–91.

Banfield J. F., Moreau J. W., Chan C. S., Welch S. A. and Little, B. (2001) Mineralogical

biosignatures and the search for life on Mars. Astrobiology 1, 447–465.

Barnes F. A. (1994) Potholes: Canyon country puzzlers. Canyon Country Explorer #1. Canyon

Country Publications. Moab, Utah, pp. 68–72.

Bayly I. A. E. (1999) Review of how indigenous people managed for water in desert regions of

Australia. J. Roy. Soc. Western Aust. 82, 17–25.

Belk D. and Cole G. A. (1975) Adaption biology of desert temporary-pond inhabitants. In

Environmental Physiology of Desert Organisms (ed. N. F. Hadley), pp. 207–226, Dowden,

Hutchinson, and Ross, Stroudsburg, PA.

Bell R. A., Athey P. V. and Sommerfeld M. R. (1986) Cryptoendolithic algal communities of the

Colorado Plateau. J. Phycol. 22, 429–435.

Bennett P. C., Hiebert F. K. and Choi V. J. (1996) Microbial colonization and weathering of

silicates in a petroleum-contaminated groundwater. Chem. Geol. 132, 45–53.

Bennett P. C., Rogers J. R., Choi W. J. and Hiebert F. K. (2001) Silicates, silicate weathering,

and microbial ecology. Geomicrobiol. J. 18, 3–19.

Birkeland P. W., Burke R. M. and Shroba R. R. (1987) Holocene alpine souls in gneissic cirque

deposits, Colorado Front Range-soil chronosequences in the western United States. U.S.

Geological Survey Bulletin 1229, 140 pp.

Chan M. A., Parry W. T and Bowman J. R. (2000) Diagenetic Fe- and Mn-oxides and fault-

related fluid flow in Jurassic sandstones, southeastern Utah. Am. Assoc. Petroleum Geol. Bull.

84, 1281–1310.

DESERT POTHOLES 299

Crowe L. M. (1991) Life without water. In Yearbook of Science and the Future, Encyclopedia

Britannica (ed. D. Calhoun), pp. 114–129, Encyclopedia Britannica, Inc., Chicago.

Crowe J. H, Lambert D. T. and Crowe L. M. (1978) Ultrastructural and freeze fracture studies

on anhydrobiotic nematodes. In Dry Biological Systems (eds. J. H. Crowe and J. S. Clegg, pp.

23–51) Academic Press, New York.

Dodson S. I. (1987) Animal assemblages in temporary desert rock pools: aspects of the ecology of

Dasyhelea subletiei (Diptera: Ceratopogoidae). J. North Am. Benthol. Soc. 6, 65–71.

Doelling H. H. (1988) Geology of the Salt Valley anticline and Arches National Park, Grand

County, Utah. In Salt Deformation in the Paradox Region, Utah (eds. H. H. Doelling et al.),

pp. 1–60. Utah Geological and Mineral Survey Bulletin 122.

Doelling H. H. (1993) Interim geologic map. Moab 300 · 600 quadrangle, Grand County, Utah,

and mesa County, Colorado. Utah Geological Survey Open File Report 287.

Doelling H. H. (2000) Geology of Arches National Park, Grand County, Utah. In Geology of

Utah’s Parks and Monuments (eds. D. A. Sprinkel et al.), pp. 11–36. Utah Geological

Association Publication 28.

Doelling H. H. (2001) Geologic map of the Moab and eastern part of the San Rafael Desert

300 · 600 quadrangles, Grand and Emery Counties, Utah, and Mesa County, Colorado. Utah

Geological Survey Map M-180, 3 pl., 1:100,000.

Edwards H. G. M., Farwell D. W., Jenkins R. and Seaward M. R. D. (1992) Vibrational Raman

spectroscopic studies of calcium oxalate monohydrate and dehydrate in lichen encrustations

on Renaissance frescoes. J. Raman Spectroscopy. 25, 99–103.

Findley R. and Sisson R. F. (1975) Miracle of the Potholes. Natl. Geograp. 148, 571–579.

Friedmann E. I. (1980) Endolithic microbial life in hot and cold deserts. Orig. Life 10, 223–235.

Friedmann E. I. (1982) Endolithic microorganisms in the Antarctic cold desert. Science 215,

1045–1052.

Friedmann E. I. and Ocampo R. (1976) Endolithic blue-green algae in the dry valleys: primary

producers in the Antarctic desert ecosystems. Science 193, 1247–1249.

Gaubin Y., Prevost M. C., Cariven C., Pianezzi B., Planel H. and Soleihavoup J. P. (1996)

Enzyme activities and membrane lipids in Artemia cysts after a long duration space flight.

Adv. Space Res. 18(12), 221–227.

Garden I. R., Guscott S. C., Burley S. D., Foxford K. A., Walsh J. J. and Marshall J. (2001) An

exhumed palaeo-hydrocarbon migration fairway in a faulted carrier system, Entrada Sand-

stone of SE Utah, USA. Geofluids 1, 195–213.

Gladney E. S., GrahamT., Ferenbaugh R.W., GellM. G., Burns C.,Morgan J. D. andNickell E. J.

(1993) Chemical analysis of selected pothole water sources in southwestern National Parks Mon-

uments and Recreation areas. Los Alamos National Laboratory Report LA-12605-MS, 55 pp.

Graham T. B. (1999) Life in the fast pool. Plateau Journal. Museum of Northern Arizona,

Winter 1999–2000, 28–45.

Hamer M. L. and Appleton C. C. (1991) Physical and chemical characteristics and phyllopod

fauna of temporary pools in north-eastern Natal, Republic of South Africa. Hydrobiologia

212, 95–104.

Horne F. R. (1967) Effects of physical–chemical factors on the distribution of and occurrence of

some southwestern Wyoming phyllopods. Ecology 48, 472–477.

Hunt C. B. (1969) Geological history of the Colorado River. In The Colorado River Region and

John Wesley Powell. Geological Survey Professional Paper 669C, pp. 59–130.

Huntoon P. W. (1988) Late Cenozoic gravity tectonic deformation related to the Paradox salts in

the Canyonlands area of Utah. In Salt Deformation in the Paradox Region, Utah (eds. H. H.

Doelling et al.), pp. 81–93. Utah Geological and Mineral Survey Bulletin 122.

Kocurek G. (1996) Desert Aeolian systems. In Sedimentary Environments – Processes, Facies and

Stratigraphy (ed. H. G. Reading), Chap. 5, pp. 125–153. Blackwell Science, Oxford, England.

300 MARJORIE A. CHAN ET AL.

Kurtz H. D. and Netoff D. I. (2001) Stabilization of friable sandstone surfaces in a desiccating,

wind-abraded environment of south-central Utah by rock surface microorganisms. J. Arid

Environ. 48, pp. 89–100.

Madin K. A. C., Crowe J. H. and Loomis S. H. (1978) Metabolic transitions in a nematode

during induction of and recovery from anhydrobiosis. In Dry Biological Systems (eds. J. H.

Crowe and J. S. Clegg), pp. 155–174. Academic Press, New York.

Moser K. A., Smol J. P., MacDonald G. M. and Larsen C. P. S. (2002) 19th century eutro-

phication of a remote boreal lake: a consequence of climate warming? J. Paleolimnol. 28,

269–281.

National Atmospheric Deposition Program (2004) National Trends Network Data Access.

Retrieved July, 2004, from http://nadp.sws.uiuc.edu.

Netoff D. I. and Shroba R. R. (1993) Morphology and possible origin of giant weathering pits in

the Entrada Sandstone, southeastern Utah – preliminary findings. US Geological Survey

OFR-93–390, 450 pp.

Netoff D. I., Cooper B. J. and Shroba R. R. (1995) Giant sandstone weathering pits near Cookie

Jar Butte, southeastern Utah. In Proceeding of the Second Biennial Conference on Research in

Colorado Plateau National Parks (ed. C. Riper), pp. 25–53. Transactions and Proceeding

Series NPS/NRNAU/NRTIP-95-11. US Department of the Interior, National Park Service.

Ortega-Calvo J. J., Arino X., Hernandez-Marine and Saiz-Jimenez C. (1995) Factors affecting

the weathering and colonization of monuments by phototrophic microorganisms. Sci. Total

Environ. 167, 329–341.

Peterson F. (1994) Sand dunes, sabkhas, streams, and shallow seas: Jurassic paleogeography in

the southern part of the Western Interior Basin. In Mesozoic Systems of the Rocky Mountain

Region, U. S. A. (eds. M.V. Caputo et al.), pp. 233–272. Rocky Mountain Society for Sed-

imentary Geology, Denver, Colorado.

Peterson F. and Pipiringos G. N. (1979) Stratigraphic relations of the Navajo Sandstone to

Middle Jurassic Formations, southern Utah and northern Arizona. US Geological Survey

Professional Paper 1035-B, 43 pp.

Planel H., Gaubin Y., Kaiser R. and Pianezzi B. (1980) Effects of space environmental factors on

Artemia eggs. Morphol. Genet., Radiobiol. Toxicol. 1, 189–198.

Ramsing N. B., Kuhl M. and Jørgensen B. B. (1993) Distribution of sulfate reducing bacteria,

O2, and H2S in photosynthetic biofims determined by oligonucleotide probes and micro-

electrodes. Appl. Environ. Microbiol. 59, 3840–3849.

Reed J. C., Bryant B. and Hack J. T. (1963) Origin of some intermittent ponds on quartzite

ridges in western North Carolina. Geol. Soc. Am. 74, 1183–1188.

Russell N. C., Edwards H. G. M. and Wynn-Williams D. D. (1998) F-T-Raman spectroscopic

analysis of endolithic microbial communities from Beacon Sandstone in Victoria Land,

Antarctica. Antarct. Sci. 10, 63–74.

Scholnick D. A. (1994) Seasonal variation and diurnal fluctuations in ephemeral desert pools.

Hydrobiologia 294, 111–116.

Seale D. (1980) Influence of amphibian larvae on primary production, nutrient flux, and com-

petition in a pond ecosystem. Ecology 61, 1531–1550.

Stal L. (1995) Physiological ecology of cyanobacteria in microbial mats and other communities.

Tansley Review No. 84. New Phytol. 131, 1–32.

ter Braak C. J. F. and Smilauer P. (1997) Canoco for Windows Version 4.02. Centre for Biometry

Wageningen, Wageningen, The Netherlands.

Thompson J. B. and Ferris F. G. (1990) Cyanobacterial precipitation of gypsum, calcite, and

magnesite from natural alkaline lake water. Geology 18, 995–998.

Twidale C. R. (1982) Granite Landforms. Elsevier Publishing Co., NY 372 pp.

Udden J. A. (1925) Etched Potholes. University of Texas Bulletin 2509, 15 pp.

DESERT POTHOLES 301

Vandervivere P., Welch S. A., Ullman W. J. and Kirchman D. L. (1994) Enhanced dissolution of

silicate minerals by bacteria at near-neutral pH. Microbial Ecol. 27, 241–251.

Wetzel R. G. (2001) Limnology: Lake and River Ecosystems, 3rd edn. Academic Press, San

Diego.

Wierzchos J. and Ascaso C. (2002) Microbial fossil record of rocks from the Ross desert, Ant-

arctica: implications in the search for past life on Mars. Int. J. Astrobiol. 1, 51–59.

Williams W. D. (1985) Biotic adaptations in temporary lentic waters with special reference to

those in semi-arid and arid regions. Hydrobiologia 125, 85–110.

Woodward B. and Kiesecker J. (1994) Ecological conditions and the notnectid-fairy shrimp

interaction. Southwestern Natural. 39, 160–164.

302 MARJORIE A. CHAN ET AL.