A description of patches of treeless grassland in the mopane catena of Kruger National Park

@alastairpotts @adriaan_grobler @wynand_uys @troos @richardgill @tonyrebelo @jeremygilmore

In the central part of Kruger National Park, north of Letaba Rest Camp (https://en.wikipedia.org/wiki/Letaba_Rest_Camp), there occur patches of treeless grassland, in a catena generally characterised by the caesalpinioid tree, mopane (Colophospermum mopane, https://www.inaturalist.org/taxa/428749-Colophospermum-mopane and https://www.researchgate.net/figure/The-distribution-of-mopane-woodlands-in-southern-Africa-This-map-is-an-extract-from_fig1_326868228).

These are called the 'Pienaar grasslands' by local ecologists. They occupy low-lying parts of the slightly undulating landscape. The soils are naturally rich in phosphorus.

On 17 Nov. 2016, shortly after the breaking of a drought, I participated in a study of the vegetation, in search of a deeper understanding of the ecological causes of local treelessness (https://www.tandfonline.com/doi/abs/10.2989/10220119.2021.1938224).

In this Post, I record the composition of the lowest vegetation in the catena, as recorded in a series of sample plots. I make notes on the associated animals.
 
The mopane vegetation adjacent to the sampling area was pale green at the time. The new leaves had started to grow, but were still far from full-size or numbers.

It is possible that this green flush in the mopane had started only once the drought broke. However, what is more likely is that it had anticipated the rain by a week or two.
 
Between the mopane and the narrow strip of treeless grassland lies a band dominated by the multi-stemmed, soft-wooded asteraceous shrub Pechuel-loeschea leubnitziae (https://www.inaturalist.org/taxa/634052-Pechuel-loeschea-leubnitziae).

This evergreen shrub is generally about 1 m high. At the time it was in full ‘fruit’ (i.e. the usual achenes expected for a daisy). The more mature branches were intact as opposed to killed/bared/broken. However, for some reason the plants were sprouting from the base, at the time of my visit.

I do not know if this re-sprouting of P-l. leubnitziae had anticipated the rains or not. I suspect that this growth was partly to compensate for herbivory by the plains zebra (https://www.inaturalist.org/journal/milewski/67981-monogastric-herbivores-taking-weedy-dicotyledonous-pioneers-rejected-by-ruminants-with-notes-on-the-grass-bothriochloa#) and/or hook-lipped rhino. I did not have the presence of mind at the time to check the bases of the plants for the characteristic signs of clipping by the premolars of this rhino.

During my visit, I observed both the plains zebra (Equus quagga chapmani) and the blue wildebeest (Connochaetes taurinus taurinus) in this treeless grassland. I suspect that what the plains zebra was eating was mainly the fresh green leaves of Bothriochloa (https://www.inaturalist.org/observations?place_id=9074&taxon_id=72016&view=species).

A noteworthy feature is the lack of evidence of the impala (Aepyceros melampus) in this bottomland. No faeces of this species were seen.

The impala is associated with patches of treeless grassland, in which it congregates at night (https://www.inaturalist.org/journal/milewski/68005-woody-plants-surrounding-treeless-lawns-on-sodic-substrates-in-southern-kruger-national-park-part-1-floristic-composition#).

Indeed, this bovid may even be responsible for making and maintaining grazing lawns in southern Kruger National Park. However, the present patches of treeless vegetation clearly have different origins, possibly because the adjacent vegetation of mopane is hardly suitable for the impala.
 
It would have been ideal to exclude P-l. leubnitziae from all 20 plots (which would have made them all virtually devoid of woody plants). However, this was not practicable, simply because the tract of treeless grassland is not extensive enough. So, some of the plots were located in the peripheral strip of shrubby vegetation.

Despite the above caveat, it remains true that all the plots were located in vegetation only about 1 m high. Pechuel-loeschia leubnitziae may have been partly unavoidable. However, at least it never grows taller than 2 m, and has such soft wood that it is an ‘honorary’ woody plant at best.
 
Our game guard, Happy, told us that part (not all) of this low-lying part of the catena has wet ground in the rainy season. However, I did not observe any sign of even the sorts of Cyperaceae that we found in the sodic patches southwest of Skukuza. I.e. I saw nothing in the vegetation that suggested seasonal waterlogging to me.
 
Bothriochloa was generally the grass genus most apparent in these plots.

During my visit I noted that green leaves of grasses were actively sprouting, after the drought-breaking rain that had fallen, presumably a week before.

The main difference between this bottomland and the mopane lands generally is that here, at the base of the catena, this grass had been heavily trampled, to the point that the dormant tussocks were physically reduced, if not actually eaten during the drought.
 
A special feature of this sampling area was the fresh heaps of the mole-rat Cryptomys hottentotus (https://www.inaturalist.org/taxa/43749-Cryptomys-hottentotus). This rodent is widespread in Kruger National Park (http://www.thekruger.com/mammals/commonmolerat.htm). However, I saw no evidence of it on this visit to the Park, except here, and (to a lesser degree) on (similarly mopane-free) basalt near Satara.
 
All the canopy cover values in this data-set refer exclusively to P-l. leubnitziae, as do the data for tallest plant in each plot. Where I fail to give data for ‘tallest plant’ specifically in the following plots, it is fair to assume that the height of P-l. leubnitziae, as the tallest species, would have been <1.5 m, and probably more like 1 m.
 
Suppressed, broken Philenoptera violacea (https://www.inaturalist.org/taxa/340211-Philenoptera-violacea) does occur in the vegetation represented by the plots, but is NOT an important component of this vegetation. In one or more plots, it is possible (although unlikely) that an individual of Vachellia tortilis (https://www.inaturalist.org/taxa/489563-Vachellia-tortilis) occurred just within the boundaries of the plot.
 
The only woody plants found in this sampling area were as follows:
 
Peschuel-loeschea leubniziae:
common as a band between the treeless grassland and the mopane, and common in several of the plots themselves. The wood is soft, and this is unimpressive as a woody plant, both in stature and in wood-density. However, it fits the term ‘shrub’, not ‘forb’.
 
Philenoptera violacea:
the main species of tree present in this bottomland, but both scarce/patchy and marginal to the treeless vegetation. One line of well-grown individuals reached the height of tall trees by the standards of Kruger National Park. However, the species did not have a significant presence in the plots, or the narrowly-defined vegetation they represent.

The fact that such a potentially tall species of tree, so well-suited to sites accumulating groundwater, was so suppressed, and made so scarce, indicates the power of whichever forces militate against trees and promote grasses in this bottomland. These forces certainly include gross damage by the African bush elephant (Loxodonta africana).

However, it is noteworthy that, despite pressure from the proboscidean, P. violacea managed to mature into tall trees in an adjacent strip, more or less corresponding with the transition from shrubby stands of P-l. leubnitziae to the mopane on the slightly higher-lying surrounding ground.
 
Vachellia tortilis:
present in the sampling area in much the same way as it is present in the relatively treeless area on basalt north of Satara: scattered and suppressed, but nevertheless ‘in its element’. I suspect that, were large herbivores excluded, this species would encroach on to this area without there necessarily being any change to the soil. Thus, I suspect that what constrains V. tortilis here is herbivory.
 
Combretum imberbe (https://www.inaturalist.org/taxa/340408-Combretum-imberbe):
technically present, but amounting to little more than a token species, being represented only by the occasional dead stump.
 
Colophospermum mopane:
marginal only.
 
Flueggea virosa:
occasional only, being less common than in the relatively treeless vegetation north of Satara.
 
Maerua parvifolia (https://www.inaturalist.org/taxa/544900-Maerua-parvifolia):
present but negligible owing to small size and scarcity.
 
The following describes each plot in turn:
 
Plot 1: we failed to locate this plot.
 
Plot 2: Canopy cover 22% (all belonging to P-l. leubnitziae). Calcrete stones nearby. The incidence of grass is similar to that in all the other plots in this sampling area that contained P-l. leubnitziae (see below). Within 25 m of the plot, on slightly higher-lying ground, is an individual tree of Philenoptera violacea, 14 m high. Essentially what this plot shows is a shrub stratum dominated by P-l. leubnitziae over a grass stratum dominated by Bothriochloa, although the latter stratum does contain various other taxa of grasses. Fresh heaps of Cryptomys hottentotus absent in this and the next several plots.
 
Plot 3: Canopy cover 20% (all belonging to P-l. leubnitziae). In or at edge of plot is at least one suppressed sapling of Vachellia tortilis, 20 cm high. Less than 25 from plot is a line of large, mature trees of Philenoptera violacea, in leaf, >15 m high. There is also a patch of Colophospermum mopane only 25 m away. A bit of calcrete shows at the surface in the plot.
 
Plot 4: This plot is located only about 30 m from the edge of the mopane vegetation. A bit of calcrete on surface. Canopy cover 25% (all belonging to P-l. leubnitziae).
 
Plot 5: In this plot there is relatively much old dry matter of Bothriochloa, remaining from the previous season. Lily geophyte absent, although present in other plots later in the series (see below).
 
Plot 6: Nearby is a suppressed, persecuted individual of Philenoptera violacea, up to 2.5 m high. There are a few fresh heaps of Cryptomys hottentotus near this plot, but none within it. A bit of fresh faeces of the African savanna buffalo nearby (but I would find it hard to believe that this species can yet get any food out of these plots, because the grass is still so short; perhaps a group merely passed through).
 
Plot 7: Largely bare except for a few old, worn tussocks of Bothriochloa.
 
Plot 8 (confused because pole was labelled as plot 12): P-l. leubnitziae relatively abundant, its canopy cover being 25%. Nearby is one individual of Flueggea virosa (https://www.inaturalist.org/taxa/340143-Flueggea-virosa), 2 m high, showing its first leaves after the rains (probably did not anticipate the rains, instead just responding promptly).
 
Plot 9: Woody cover as in last plot: P-l. leubnitziae has canopy cover of 25%. Some 20 m away from the plot is one individual of suppressed, hammered Philenoptera violacea. An unexpecteding pattern emerges as we work our way through the series of plots: where P-l. leubnitziae was common, re-sprouting of the grasses was more apparent. My interpretation: the new leaves of grass have been promptly consumed by the blue wildebeest, over most of this bottomland; but, because this animal avoids any cover that might hide predators, the grass has been spared among the shrubs. In this plot the new green grass, which has been thus spared from grazing, is revealed to be considerable, less than one week after the first drought-breaking rain; we see signs of Urochloa, Panicum, Eragrostis, Bothriochloa, etc. Fresh heaps of Cryptomys hottentotus absent.
 
Plot 10: Canopy cover of P-l. leubnitziae is 5%. There are a few large individual trees of Philenoptera violacea, up to 17 m high, in the general vicinity. Incidence and activity of grasses is similar to that noted for plots 8 and 9. Fresh heaps of Cryptomys hottentotus absent.
 
Plot 11: Canopy cover of woody plants (all belonging to P-l. leubnitziae) is 22%. Just outside plot is Vachellia tortilis, 1.5 m high, and unusually bare. Again, the fresh green leaves of re-sprouting grasses here are relatively apparent, ungrazed by the blue wildebeest, possibly because the cover of P-l. leubnitziae deters these animals (predation risk). Bothriochloa is, as usual, the most apparent type of grass. Fresh heaps of Cryptomys hottentotus absent.
 
Plot 12: Similar to the last plot, with canopy cover being 25%. Fresh heaps of mole-rates absent. Tallest plant is 2 m (P-l. leubnitziae). Nearby are a few small, suppressed individuals of V. tortilis, a species that occurs in the sampling area only as scattered individuals, and never in mature form.
 
Plot 13: Similar to the last plot again, with canopy cover 22%. Bothriochloa common as the mainly dormant remains of the tussocks.
 
Plot 14: Devoid of woody plants, except for one or two individuals of P-l. leubnitziae, which are situated marginally. There are others of the same species nearby, too. A few dry traces of last season’s Bothriochloa, which has partly re-greened, but is scarcer here than in some of the other plots in this series. The re-greening of Bothriochloa seems to be ahead of that of the other, ‘sweet’ grasses, and I suspect that the blue wildebeest is taking Bothriochloa along with other grasses at the moment, despite the reputation of Bothriochloa for unpalatability.
 
Plot 15: Devoid of woody plants, except for one individual of P-l. leubnitziae, which is situated marginally. Nearby, there is one individual Colophospermum mopane, 1 m high, in fresh leaf. The ground in this plot is essentially bare, partly because much of the green grass that has already sprouted after the recent rain has been promptly eaten by the local population of the blue wildebeest, which we see nearby. As usual there are a few dry traces of last season’s Bothriochloa still apparent.
 
Plot 16: Devoid of woody plants, but nearby there are

• an old stump of Combretum imberbe, and
• one small individual of Colophospermum mopane, <1 m high.

The ground is nearly bare, except for

• a few old worn/reduced tussocks of Bothriochloa, now partly re-greened,
• the first green shoots of the other grasses, which are presumably ‘sweet’,
• sundry sprouting/germinating forbs (all still small and hardly noticeable), and
• a geophytic white lily (see below).

Fresh heaps of Cryptomys hottentotus absent from this plot. The blue wildebeest is present right nearby, obviously having grazed in this plot recently. One or two individuals of P-l. leubnitziae in general vicinity.
 
Plot 17: No woody plants in plot, and very little grass dead matter is left above ground now, after the recent severe drought. The plot is essentially bare except for the first signs of green grass starting to sprout after the recent rain. What little remains of the tussocks from the previous season belongs to Bothriochloa, and these plants have already greened up partly, over the last few days. Fresh heaps of Cryptomys hottentotus absent from this plot.
 
Plot 18: No woody plants in plot, and very little grass dead matter is left above ground now, after this severe drought. Crinum-like geophytic lily (with grass-like leaves) already in flower; this species is ‘grazed’ by the blue wildebeest, according to our game guard, Happy. It seems that this geophyte is the most vigorous herbaceous plant to spring up once it rains; it is already in large flower despite the fact that only traces of fresh green grass are yet apparent. Fresh heaps of Cryptomys hottentotus present in plot.
 
Plot 19: No woody plants in plot. Although there are traces of fresh green grass, all grasses so reduced that the ground is essentially bare. A negligible amount of old, dry grass is left from the previous season. Nonetheless, Bothriochloa is recognisable. Also apparent are e.g. Panicum and Urochloa. Within the plot are fresh heaps of Cryptomys hottentotus, which in my experience is unusual in Kruger National Park. Blue wildebeest nearby, a different group from that noted above. Nearby are a few leafy sprigs of Maerua parvifolia, which are negligible as woody cover.
 
Plot 20: Fresh heaps of Cryptomys hottentotus present, perhaps the best evidence I have ever seen of this type of rodent in Kruger National Park. Tracks of the hippopotamus and the hook-lipped rhino present in plot. A few individuals of P-l. leubnitziae present in plot. Tallest woody plant about 1 m.

Posted on July 7, 2022 08:23 AM by milewski