Thelocactus. Climatic niches and sites on limestone walls

What the distribution model describes

Mosco compared sites of nine Thelocactus species with eight climatic variables, slope exposure and a terrain ruggedness index. Maxent models use relationships between known sites and environmental conditions to identify areas with similar sets of characteristics. The climate layers had a resolution of 30 arcseconds, approximately one kilometre. Different slopes, rocks and soils may lie within such a cell.[Mosco 2019, pp. 3–5]

Each model was repeated 20 times. Performance was assessed partly through AUC, a measure of separation between sites and environmental background, and comparison with 100 random models. A high AUC value describes the result of this comparison. The predicted area remains dependent on the number and distribution of sites and the set of variables used.[Mosco 2019, pp. 4–6, Table 2]

Nine species and different dominant plants

Isothermality and precipitation of the wettest quarter had the greatest relative importance in the T. bicolor model. Isothermality is the ratio of mean diurnal temperature range to annual temperature range, multiplied by 100. A high value means that daily fluctuations make up a large part of the annual temperature range. Precipitation of the driest quarter dominated for T. conothelos, and precipitation of the wettest quarter for T. rinconensis. These are relationships detected in site data.[Mosco 2019, pp. 4–7, Table 1]

Levins' niche breadth is a dimensionless index of the distribution of predicted habitat suitability. In this set, T. bicolor occupied the broadest niche and T. tulensis the narrowest. Values describe the regional model used. They do not directly measure an individual plant's physiological tolerance of drought or temperature. Permutation importance indicates how the model result changes when a variable is shuffled, so it too remains a model measure.[Mosco 2019, pp. 4–7]

Niche similarity and co-occurrence

Niche overlap was assessed using Schoener's D, which ranges from 0 for no overlap to 1 for identical distributions. For T. hastifer and T. leucacanthus, it was 0,626. The author identified five places where they co-occurred. This was the only pair in the set without a significant difference in the niche equivalency test. The result nevertheless depends on the input data, including only 11 sites for T. hastifer.[Mosco 2019, pp. 7–11, Table 3]

For many pairs, niches were more similar than would be expected from random use of the available environmental background. Such niche conservatism means that some similar climatic relationships are retained within the genus. At the same time, most pairs were distinguished by the equivalency test. Individual species may share a broad climate type while occupying different parts of its range.[Mosco 2019, pp. 8–12, Figures 2–3]

Thelocactus tepelmemensis on canyon walls

Thelocactus tepelmemensis, described by Davis and coauthors in 2018, grows in a narrow limestone canyon in northern Oaxaca. Documented plants occupied fissures and ledges on walls facing northeast or southeast, at 1420–1460 m elevation. The substrate consisted of limestone fragments. The authors observed that plants were sheltered from afternoon sun. This information describes site conditions without measuring their effects on tissue temperature.[Davis et al. 2018, pp. 115–121, Figures 1–3]

Measurements of 20 plants, herbarium material and seed microscopy identified the species' characters. Stems had 11–13 vertical ribs with tubercles. Red-purple flowers reached 13–23 mm in length. In the seed, the border of the hilum region, the scar left by connection to the fruit, was expanded, and the micropyle lay within this border. In the T. leucacanthus compared, the micropyle was outside it. Seed structure here supplements differences visible in stems and flowers.[Davis et al. 2018, pp. 117–122, Figures 3–4, Table 1]

Along two accessible sections, 35 plants were counted over about 50 m and 15 over about 30 m. These numbers refer to wall sections accessible to observation. They do not determine the species' total abundance in the canyon. T. tepelmemensis was not among the nine species modelled by Mosco, so the indices from that analysis do not describe its niche.[Davis et al. 2018, Davis et al., pp. 120–121; Mosco, pp. 3, 6][Mosco 2019, Davis et al., pp. 120–121; Mosco, pp. 3, 6]

The regional scale and the root scale

A new T. hexaedrophorus site documented in 2021 near Mesillas in Aguascalientes contained about 15 plants on a north-facing slope at 2085 m elevation. A herbarium specimen from this place confirmed the first occurrence of the genus and species recorded in the state. Such a field point establishes actual plant occurrence at a place whose occupation is not demonstrated by a model result alone.[Sandoval-Ortega et al. 2022, pp. 48–50]

The regional rainfall pattern determines seasonal water supply. A rock fissure and its exposure determine conditions in a plant's immediate surroundings. Mosco's models and observations from the canyon describe these different scales. Combining them helps explain habitat. Converting a precipitation index into a pot-watering frequency would require data on substrate, temperature and the activity of the cultivated plant, which these studies do not provide.[Mosco 2019, Mosco, pp. 3–5, 11–12; Davis et al., pp. 120–121][Davis et al. 2018, Mosco, pp. 3–5, 11–12; Davis et al., pp. 120–121]