Nematode communities indicate diverse soil functioning across a fog gradient in the Namib Desert gravel plains

Abstract Soil nematodes are fundamentally aquatic animals, requiring water to move, feed, and reproduce. Nonetheless, they are ubiquitous in desert soils because they can enter an anhydrobiotic state that allows them to persist when water is biologically unavailable. In the hyper‐arid Namib Desert of Namibia, rain is rare, but fog routinely moves inland from the coast and supports plant and animal life. Very little is understood about how this fog may affect soil organisms. We investigated the role of fog moisture in the ecology of free‐living, soil nematodes across an 87‐km fog gradient in the gravel plains of the Namib Desert. We found that nematodes emerged from anhydrobiosis and became active during a fog event, suggesting that they can utilize fog moisture to survive. Nematode abundance did not differ significantly across the fog gradient and was similar under shrubs and in interplant spaces. Interplant soils harbor biological soil crusts that may sustain nematode communities. As fog declined along the gradient, nematode diversity increased in interplant soils. In areas where fog is rare, sporadic rainfall events can stimulate the germination and growth of desert ephemerals that may have a lasting effect on nematode diversity. In a 30‐day incubation experiment, nematode abundance increased when soils were amended with water and organic matter. However, these responses were not evident in field samples, which show no correlations among nematode abundance, location in the fog gradient, and soil organic matter content. Soil nematodes are found throughout the Namib Desert gravel plains under a variety of conditions. Although shown to be moisture‐ and organic matter‐limited and able to use moisture from the fog for activity, variation in fog frequency and soil organic matter across this unique ecosystem may be biologically irrelevant to soil nematodes in situ.


| INTRODUC TI ON
"The Gravel Flats of the Outer Namib are of almost unbelievable monotony. Over large areas, their local relief can be measured in increments of inches per acre". (Logan, 1960).
The Namib Desert is a narrow, hot desert that extends 2000 km along the western coast of southern Africa, from Angola, through Namibia, and into South Africa. Aridity in coastal Namibia is maintained by the Benguela Current, which brings cold water to the area and limits the formation of rainfall-producing fronts. While the Namib Desert is widely known for its soaring, red-orange dunes (i.e., the Namib "Sand Sea," a UNESCO World Heritage site), much of the Namib Desert is composed of gravel plains with little topographical relief. Nearly barren of vegetation, the gravel plains offer minimalism (and monotony) on a grand scale ( Figure 1a).
The Namib Desert is among a small group of global deserts that receive moisture from the fog as well as sporadic rainfall (~5-100 mm precipitation annually, Eckardt et al., 2013). The distribution of fog and rain varies considerably over time and space. Cold fogs and mists can extend far inland, but their frequency decreases with distance from the coast, while rain inputs tend to increase (Eckardt et al., 2013;Spirig et al., 2019). Fog is rare at 60 km from the coastline and beyond (Kaseke et al., 2018), but at 33 km, it may be present on up to 87 days of the year (Mitchell et al., 2020), making fog a relatively reliable source of moisture in parts of the gravel plains. The role of fog in the ecology of many Namib Desert species, including beetles, lizards, grasses, and shrubs has been well-documented (Jacobson et al., 2015;Mitchell et al., 2020;Wang et al., 2019). Many organisms have evolved morphological and behavioral adaptations to capture moisture from fog (Mitchell et al., 2020). Less is known about the possible use of fog by belowground biota, although fog-induced increases in soil water potential have been measured in gravel plains soils Wang et al., 2019).
Namib Desert gravel plains soils contain many of the invertebrate taxa found in soils globally, including microarthropods such as collembola (Collins et al., 2019) and mites (André et al., 1997), but little is known about soil nematodes. Nematode diversity and abundance have been studied in soils under Acanthosicyos horridus (!nara melon shrubs) in the Namib Desert sand dunes (Marais et al., 2020), and some nematode species have been described that were isolated from soil associated with Welwitschia mirabilis plants in the gravel plains . However, nematode distribution and abundance in the gravel plains have not been systematically investigated, and whether the activity, abundance, and diversity of these animals are connected to fog is unknown. Nematodes reside within water films on the surfaces of soil particles (Yeates, 2004).
In the absence of water, nematodes become quiescent, entering an ametabolic state known as anhydrobiosis (Demeure et al., 1979;Treonis et al., 2000). When nematodes are anhydrobiotic, they coil their bodies, and they cannot feed, move, or reproduce (Crowe et al., 1992). Prior research has shown that soil nematodes in desert ecosystems rapidly become active in response to rainfall (Whitford et al., 1981) and melting snowfall (Treonis et al., 2000).
It is unknown whether nematodes may be similarly sensitive to fog events.
Soil nematode communities rely on vegetation as either a direct source of energy (e.g., plant-parasitic taxa) or indirectly through the detrital food web (bacterivores, fungivores, etc.). In deserts, where plant productivity is limited by the extreme environment, nematode abundance usually is highest in association with perennial plants, such as shrubs (Freckman & Mankau, 1977;Pen-Mouratov et al., 2003;Treonis et al., 2019). However, biological soil crusts, found in interplant spaces in arid ecosystems, also can support nematode communities (Darby et al., 2007). The periodic germination and growth of desert ephemerals also have been found to increase soil nematode abundance (Treonis et al., 2019).
We studied soil nematode communities in the gravel plains of the Namib Desert in a series of experiments. The purpose of this work was to determine how the fog gradient may impact the abundance and diversity of soil nematodes. First, we conducted a survey of nematode abundance and diversity across the gradient in shrub and interplant soils. We predicted that nematode communities would be more abundant and diverse under shrubs and where fog precipitation is highest. Second, we investigated the degree to which nematodes are limited by moisture and energy availability in a laboratory incubation experiment in which we amended soils with water and/or organic matter. We predicted that these would be co-limiting factors for nematode abundance in gravel plains soils. Finally, we studied the activity of soil nematodes in situ during a fog event. We predicted that nematodes in gravel plains soils could utilize fog moisture to emerge from anhydrobiosis.

| Gravel plains nematode communities across the fog gradient
Soils were collected from nine sites (A-I) beginning 18 km from the coast, near Walvis Bay, and extending inland 87 km (Figure 1e, Table 1). Soils were collected at 10-km intervals moving inland on 14 May 2020 (sites A-E) and 19 September 2020 (sites F-I). Ideally, all sites would have been sampled at approximately the same time, but this was not possible due to Covid-19 lockdowns and travel restrictions in Namibia in 2020. However, there had been no significant rainfall in the region for at least 24 months (Figure 2), and the soils were very dry. Based on the prior experience of the authors studying nematodes in desert ecosystems, it is unlikely nematode communities at sites F-I underwent significant changes between May and September that could obfuscate comparison to soils from the earlier sampling date. Sites were selected to maximize distance from the main road through the region because traffic raises dust and imparts some degree of disturbance to adjacent soils. However, only existing off-road tracks were used to access sites, which made it impossible to sample across the gradient in a straight line. F I G U R E 1 (a) Representative landscape of the Namib Desert gravel plains, (b) Arthraerua leubnitziae (pencilbush), (c) Pechuel-loeschea leubnitziae (wild sage), (d) map of Namibia with the location of field sites in the blue rectangle, (e) arrangement of field sites (gray squares) moving inward from the coast along a gradient of decreasing fog. Blue circles represent meteorological stations (from left to right, Kleinberg, Marble Koppie, and Garnet Koppie). The line extending eastward from Walvis Bay is the C14 road, the main route through the region For each sample, several quartz rocks were gently lifted from the soil, and the soil adhering to the bottom of the rocks (~20 g) was scraped into conical tubes. Samples were stored in coolers at ambient temperature prior to analysis.

| Nematode community analyses
Nematodes were extracted over 72 h from 120 g soil using a Baermann funnel technique (60 g soil × 2 funnels per sample, combined when drawing off the solution, Baermann, 1917

| Nematode responses to soil amendment with water and organic matter
Approximately 7 kg of soil were collected from shrubs and interplant spaces at two sites (B, H) for use in a soil amendment experiment.
These two sites were selected because they represented a soil that experienced frequent fog and rare rainfall (B) and soil that rarely experienced fog but received more rainfall (H). Soils were sieved (2 mm) and mixed prior to distribution to 250-ml plastic containers that had been surface sterilized with 70% ethanol. Each of the four soil types (B-Shrub, B-Interplant, H-Shrub, and H-Interplant) was distributed to containers (300 g each), which were assigned the following four treatments: control, dung-amended, water-amended, and dung-and water-amended (n = 4 replicates per soil type for a total of 64 experimental units). Dung amendment was accomplished with the application of 5-g Hartmann's mountain zebra (Equus zebra hartmannae) F I G U R E 2 Meteorological data for the 4-year period preceding and encompassing this study (2017-2020) from meteorological stations located near sampled sites in the gravel plains, with Kleinberg located closest to the coast and Garnet Koppie the farthest. (a) Precipitation as fog, (b) precipitation as rain, and (c) soil temperature (3-cm depth). Points represent monthly totals for precipitation or the monthly average for temperature. Horizontal lines represent the averages for these data across the 4-year time period dung that was collected from the gravel plains. Soil properties were measured as described above.

| Nematode response to fog
Whether nematodes are anhydrobiotic or active can be determined by examining their morphology after extraction from soils using sucrose solutions (Demeure et al., 1979). Anhydrobiotic nematodes coil their bodies (Figure 7d). Soil samples were collected before, during, and after a fog event from shrub soils at site B on 27-28 September 2020. Fog events are unpredictable and challenging to study in situ, which necessitated limiting this experiment to a single field site. First, soils were collected from five shrubs at the site as described above for routine characterization of nematode abundance and diversity. Then, additional samples were collected over time from other, nearby shrubs to determine the anhydrobiotic status of the nematodes. Different shrubs were used to minimize disturbance at the site. These soils were not sieved, but 50 g of soil collected from each shrub were immediately transferred to 100 ml 1.25 M-sucrose solution to fix the nematodes in the activity state they were in at the exact time of sampling (anhydrobiotic or active). Five soils were sampled and sucrose-fixed at 20:00, 08:00, and 12:00. Sucrose-fixed samples were stored in coolers until returning to the laboratory at 14:00, and then they were refrigerated.
Within 24-48 h, each of the sucrose-fixed samples was subjected to a sieving procedure to isolate nematodes from the soil material (Treonis et al., 2000). Sucrose solutions were used throughout this procedure to avoid exposing nematodes to pure water, which would have induced uncoiling. Soils and sucrose were mixed in the collection jar and then poured onto stacked 425-and 38μm sieves.
Sucrose (1.25 M) was used to rinse the material retained on the top sieve. Then, more sucrose was used to rinse the material retained on the bottom sieve to remove as many soil particles as possible. The soil retained was rinsed into a beaker. Finally, to separate nematodes from silt particles, the solution in the beaker was slowly transferred with a 5-ml pipette into a 50-ml conical tube of cold 2M sucrose solution (2°C). These tubes were allowed to rest (5°C) for 5 h, during which heavier particles settled to the bottom of the tube while lighter material (including nematodes) was retained above. This top layer was poured onto a 25μm sieve, and the material retained on the sieve was transferred to a dish for microscopic analysis. The morphology of each nematode encountered was classified as either uncoiled (active or dead at the time of sampling) or coiled in a full circle or whorl (anhydrobiotic and inactive at the time of sampling).

| Statistical analyses
Statistical analyses were performed with R version 4.1.0 (https:// www.r-proje ct.org) (R Development Core Team, 2021). Elevation increased with distance from the coast (Table 1), allowing this variable to serve as a proxy for the fog gradient. Correlations between elevation, soil properties, nematode abundance, taxonomic richness, and the Gini-Simpson Index were tested using the Kendall rank correlation coefficient due to non-normal distributions for many variables.
For the soil amendment experiment, ANOVA was used to investigate differences in ln(x + 1)-transformed nematode abundance in response to water and/or dung, with soil type as a fixed effect. The Shapiro-Wilk test was used to assess the normality of the data following transformation. Where treatment effects were significant, means were compared using the Tukey HSD multiple comparison procedure. The proportion of nematodes uncoiled in samples collected over a fog event was ln-transformed, and the three time points were compared by one-way ANOVA, followed by Tukey's HSD. A constrained linear canonical ordination (RDA, Rao, 1964) was performed using the R vegan package (Oksanen et al., 2018) to test the contribution of measured soil properties to the abundance of nematodes from each taxa represented, using only nematode types found in more than one sample. Variables included soil moisture, organic content, pH, electrical conductivity, and whether the sample was from a shrub or interplant space. The abundances of each nematode taxon were Hellinger-transformed prior to RDA analysis.

| Gravel plains nematode communities across the fog gradient
Meteorological data (2017)(2018)(2019)(2020) show that fog decreased from west to east, moving inland from the coast (Figure 2a). Averaged monthly totals for fog precipitation were an order of magnitude greater than rainfall at the two meteorological stations closer to the coast, and rain precipitation was slightly greater than fog at the meteorological station furthest inland (Figure 2a,b). While the averages across the time period highlight the overall variation in fog and rain moisture along the gradient, it is important to note that on most days, soils across the gradient do not experience any fog or rain.
Elevation was used as a proxy variable for the fog gradient, as elevation increased moving inland from the coast, correlating to decreasing fog frequency. Soil properties varied as elevation increased, but for most variables the pattern was different for shrub and interplant soils (Figure 3). The organic matter did not vary with elevation in shrub soils (Kendall's rank correlation test, p = .87), but it decreased with a gain in elevation in interplant soils (Kendall's rank correlation test, p < .0001, Figure 3a,e). Soil organic matter was higher in soils under shrubs than in interplant soils (0.87% ± 0.087 vs. 0.57% ± 0.096, t-test, p = .001). Soil electrical conductivity declined with increasing elevation in shrub and interplant soils (Kendall's rank correlation test, p < .001 for both, Figure 3b,f). Soil pH declined in soils under shrubs (Kendall's rank correlation test, p = .0037) and increased in interplant soils (Kendall's rank correlation test, p = .0062), as elevation increased (Figure 3c,g). Finally, soil moisture did not change with elevation in shrub soils (Kendall's rank correlation test, p = .071), but declined in interplant soils (Kendall's rank correlation test, p < .001, Figure 3d,h). Overall, shrub soils had a similar organic matter and moisture contents across the fog gradient, but pH decreased, which may reflect differences in shrub litter chemistry between the two species found at each end of the gradient. Interplant soil organic matter declined across the gradient, which may reflect F I G U R E 3 Correlations between elevation and soil properties in shrub and interplant soils collected along a fog gradient in the Namib Desert gravel plains. The elevation is a proxy variable for the fog gradient, as fog declines with increasing elevation. The shaded area represents a 5% confidence interval of Kendall's rank correlation coefficient (R, n = 54). Masl = meters above sea level a reduction in fog-supported biological crust productivity and/or an increased decomposition rate. All soils showed declining electrical conductivity as elevation increased, which may reflect the influence of more frequent rainfall on salt mobility.

| Nematode communities within soil microhabitats
Soils collected at site B to investigate nematode density within microhabitats found that most nematodes were located in the

| Nematode response to soil amendment with water and organic matter
Nematodes responded to soil amendment (ANOVA, significant site*treatment effect, p < .001). In shrub soils from site B and interplant soils from sites B and H, only dung + water amendment increased nematode abundance significantly (Figure 6a
Panagrobelus was the only nematode found in these samples.

| DISCUSS ION
Nematodes are ubiquitous in the hyper-arid, gravel plains soils of the Namib Desert, which is likely tied to their use of the desiccation survival strategy, anhydrobiosis. We found that nematodes rapidly emerged from anhydrobiosis in response to soil wetting during a fog event. This study is the first to demonstrate that moisture from Our meteorological data are consistent with prior, more extensive analyses documenting that the prevalence of fog declines in the Namib Desert with increasing elevation, moving inland from the Atlantic Coast, creating a well-delineated fog gradient (Spirig et al., 2019). Prior research has shown that fog can raise the moisture content of gravel plains soils Wang et al., 2019).
However, Li et al. (2018) also failed to detect a correlation between fog and soil moisture at one gravel plains field site over an 80day period, suggesting the relationship could be more stochastic.
Unfortunately, we were not able to directly measure soil moisture in response to fog at each of our field sites due to the unpredictable nature of fog and the lack of access to deployable instrumentation.
We did measure a significant increase in soil moisture during a modest fog event at site B. Although the soil texture was similar across our field sites, it is possible that the penetrability of fog moisture into soil varies among our field sites due to topography, the presence of desert varnish, or the nature of the fog itself (e.g., droplet size, duration, etc.). More research would be valuable to establish the direct impact of fog on soil moisture across the Namib Desert gravel plains, but evidence exists to support the general likelihood that fog events can raise soil moisture.
Because fog can trigger nematode emergence from anhydrobiosis, creating a favorable environment for nematode feeding and reproduction, we predicted that more frequent fog-derived moisture inputs near the coast would promote nematode abundance and diversity. Contrary to our prediction, we found no effect of posi- In this study, microbial communities did not vary among the time points, except immediately following a significant rainfall, suggesting that fog events over the study period either failed to wet the soil or were not sufficient to impact bacterial diversity. Regardless of whether fog consistently moistens the gravel plains soils across the gradient or not, our results from soils in the field align with most of the results from the soil amendment experiment where nematodes generally did not respond numerically to amendment with water alone.
Nematodes in soils from other deserts have been shown to be non-responsive to variation in soil moisture, either through experimental amendment or from natural sources. For example, Vandegehuchte et al. (2015) found that microbivorous nematodes did not respond to 4 years of simulated rainfall in Chihuahuan Desert soils. Similarly, Simmons et al. (2009) and Treonis et al. (2002) found that moisture addition did not affect nematode density in the Antarctic Dry Valleys. Other studies have found that desert nematode abundance was not enhanced in areas of ephemeral water flows (i.e., rainfall or snowmelt channels, Treonis et al., 1999;Treonis et al., 2019). In contrast, Zawierucha et al. (2019) found that the Antarctic nematode Scottnema lindsayae increased in abundance with elevation, possibly responding to soil moisture derived from clouds, and Steinberger and Sarig (1993) documented increased F I G U R E 4 Correlations between elevation and nematode abundance and richness in shrub and interplant soils. The elevation is a proxy variable for the fog gradient, as fog declines with increasing elevation. The shaded area represents a 5% confidence interval of Kendall's rank correlation coefficient (R, n = 54). Masl = meters above sea level (a) (c) nematodes following wetting of soils in the Negev Desert. More work is needed to understand the interplay among fog, soil moisture, and nematode abundance in arid ecosystems. In the Namib Desert gravel plains, the fog gradient we studied may not induce biologically significant variation in soil moisture, either because the soil is not wetted sufficient or because water is not the primary factor limiting nematode abundance in these soils.
When gravel plains soils were subject to co-amendment with moisture and organic matter for 30 days in vitro, most nematode communities were numerically responsive. No changes were found in soils amended with organic matter alone, and nematodes responded to moisture amendment alone in only one of the four soil types tested. The treatments and duration of exposure for the soil amendment experiment were different from what these nematodes are likely to experience in the field, where organic matter and moisture are more likely to be available in shorter pulses. However, the amendment experiment demonstrates the sensitivity of the community to both of these factors. The results suggest that organic matter is a key factor limiting nematode abundance in gravel plains soils and that they can only take advantage of increased availability of this resource if moisture also is available to support their activity.
However, these results are inconsistent with the lack of variation in nematode abundance across the landscape. Moisture and organic matter differed naturally among sites, but these variables were not correlated to abundance. In the field, the short duration of moisture pulses may limit the capacity for nematodes to respond to variation in organic matter.
Soil nematode abundance in deserts is thought to be closely tied to soil organic matter content, which supplies substrate to the decomposer microorganisms (bacterial and fungi) that many nematodes consume for energy (Treonis et al., 2019). Nematodes in the gravel plains of Namibia may be an exception to this trend, despite demonstrating organic matter limitations in the soil incubation experiment. Soil organic matter was higher in soils associated with shrubs, but nematode abundance was comparable to interplant soils for most samples. This contrasts with the higher nematode abundance Marais et al. (2020)  in the "lichen fields" of the Namib Desert (sites A, B, and C). Like nematodes, lichens have also been shown to activate metabolically in response to fog in the Namib Desert (Lange et al., 1994).
Biological soil crusts, consisting of fungi, lichens, cyanobacteria, and other microorganisms, occupy much of the soil surface area in arid and semi-arid ecosystems (Pointing & Belnap, 2012).
Crusts likely were present in all the interplant spaces we sampled, although only visible at sites A-C, where fog is frequent. Soil organic matter was not high in interplant soils, but the quality of the material is likely different than organic matter formed from shrub litter, and it may be highly suitable for supporting nematodes and their microbial food sources. Fungal-feeding nematodes were particularly abundant in interplant soils at sites A, B, C, and F, as well as in the hypolithic soils sampled from sites A-E. Some of these soils had higher soil moisture, perhaps due to recent fog events, and Aphelenchoides, a fungal feeder, was found by RDA to be associated with higher moisture. The interplant spaces in the fog-fed, lichen fields of the Namib Desert may represent a unique habitat for nematodes that subsist on fungi from the crusts.
Although nematode abundance did not vary across the fog gradient, our data suggest that nematode diversity was highest where the fog was least frequent, specifically in interplant soils, and again, contrary to predictions. Soils at the eastern and most inland part of the fog gradient were drier and contained less organic matter than soils nearest to the coast. They were also less saline (lower electrical conductivity), likely because of the effects of more frequent rainfall on salt mobilization. Yet, they contained more diverse nematode communities. During our study period, the gravel plains landscape appeared similar across the fog gradient, with widely spaced perennial shrubs as the only visible vegetation. This landscape changes dramatically, however, following significant rainfall, and although infrequent and unpredictable, rainfall is more frequent further inland (Eckardt et al., 2013). When significant rain does fall in Namib, such as was measured at the Garnet Koppie meteorological station in 2017 and 2018 (Figure 2), the growth of grasses and forbs in intershrub spaces dramatically transforms the landscape. Desert ephemerals such as Stipagrostis grasses germinate after at least 20 mm of rainfall (Seely & Pallett, 2008). These plants also attract grazing animals, increasing dung and urine inputs to the soil that may affect soil food webs, as has been seen in other ecosystems (Pan et al., 2022).
Furthermore, the composition and function of Namib Desert soil microbial communities have been shown to be dramatically altered by significant water inputs, both in ex situ microcosm experiments (Frossard et al., 2015) and in in situ field studies (Armstrong et al., 2016).
Although we could not study the direct impact of rain on nematodes during the time span of this research, rainfall-induced vegetation may have a persistent effect on soil nematodes even after the plants are grazed down. Higher nematode diversity in the interplant soils that were most likely to experience rain may be due to the legacy effect of resources created by the ephemeral plants that grew following significant rainfall events. In an unusually rainy year, Hachfeld (2000) found an increase in plant diversity along the fog gradient encompassing our field sites, with the richness of shrubs, grasses, and forbs increasing from 4 to 14 1000 m −2 , moving from 20 to 110 km inland. Linkages between above-and below-ground biodiversity are challenging to demonstrate but have been seen in other ecosystems (Porazinska et al., 2018). Furthermore, in the Mojave Desert, a rainfall-induced "superbloom" of desert annuals was associated with a rapid increase in nematode abundance in intershrub soils (Treonis et al., 2019). Ephemeral plants appear to have an important role in supporting nematode biodiversity in desert soils.
Other studies have shown variation in biological assemblages across the fog gradient in the Namib Desert, including collembola, ants, and bacteria. Species richness for these organisms also increased moving inland, particularly when the sampling effort is extended farther and into areas that receive rain more often (Collins et al., 2019;Marsh, 1986;Scola et al., 2018).
Specific nematode taxa were linked to shrubs throughout the gravel plains. RDA showed that Panagrolaimus was specifically associated with shrubs and soils with higher organic matter content. These nematodes were found throughout the fog gradient.
Panagrobelus was also abundant under shrubs but only the higher salinity A. leubnitziae shrubs at the sites nearest to the coast (sites A-E). This nematode was not present in shrub soils under P. leubnitziae. The two shrub species that were sampled across the fog gradient are very different, A. leubnitziae is smaller, with fleshy stems and small, scale-like leaves with relatively little plant litter. In contrast, P.
Panagrolaimus and Panagrobelus were rare in interplant soils, and these nematodes may be specialists for the relatively organic-rich environment under perennial plants in the Namib Desert, although decomposition rates are likely to be very slow. Panagrobelus seems to be better adapted to A. leubnitziae shrubs in the area of the desert with frequent fog, which may relate to its ability to respond to foginduced changes in soil moisture by emerging from anhydrobiosis.
Our study provides insights into how fog influences soil nematode diversity and abundance across the Namib Desert gravel plains.
Soil nematodes appear to be well-adapted to varying resource availability and soil functioning across the fog gradient. A combination of organic matter from multiple sources, including biological soil crusts, ephemeral vegetation, and perennial shrubs, seems to be supporting nematode communities, indicative of the diverse soil functioning within this system. Nematodes responded to a fog event that wetted the soil, allowing them to emerge from anhydrobiosis and become active. Fog appears to be an important resource that sustains biological soil crusts, microbes, and microbivorous nematodes in many gravel plains soils. In contrast, in areas that rarely experience fog, sporadic rainfall events may be generating pulses of ephemeral plant productivity and diversity that sustain higher diversity nematode communities. The gravel plains of the Namib Desert are experiencing several anthropogenic threats. Mining and tourism affect the physical integrity of the soil, with mining an intense, but highly localized activity (Wassenaar et al., 2013), while tourism impacts (i.e., mainly off-road driving) are widespread and more diffuse. Furthermore, there is concern about the effects of a changing global climate on the frequency and distribution of precipitation in the Namib Desert writing -review and editing (supporting).

ACK N OWLED G M ENTS
This work was performed in the Namib-Naukluft National Park with permission from the Ministry of Environment, Forestry and Tourism. Soils were collected from Namib Naukluft National Park under an NCRST permit issued to A. Treonis (RPIV01022019).
We thank Namibia University of Science and Technology for logistical support. This research was supported by a Fulbright Scholar

Award and a University of Richmond School of Arts & Sciences
Research Grant to A. Treonis.

CO N FLI C T O F I NTE R E S T
The authors have no conflicts of interest to declare.