Effects of grazing and fire on herbaceous species in the Bolivian Altiplano

High mountain rangelands had been under human land use for millennia in most parts of the
world. Traditional land care systems led to highly diverse and stable ecosystems. As these
traditions fade and population pressure rises, these grassland systems become overgrazed,
lose carrying capacity and biodiversity and become eroded. Semi-arid high elevation pastures
in tropical and subtropical regions are particularly endangered. This PhD project aimed at
assessing the plant inventory, the productivity and sensitivity of the herbaceous fraction of the
flora to grazing by camelids in the Festuca orthophylla tall-tussock-dominated puna called
'pajonal' in the Bolivian Altiplano at 4250 m elevation (Sajama National Park 18°08’S,
68°58’W), with annual temperatures around 8.8 °C, the minimum and maximum temperatures
were -5.9 and 31.0°C during the 2006-2008 observation period. This area receives between
250 and 350 mm of rainfall during a short (3-4 month) rainy season from December to March,
this is also the growing season of the herbaceous species. The Sajama village has a population
of about 150 families, with a long tradition in husbandry of llamas (53%), alpacas (39%) and
sheep (8%).
My project was divided in three tasks, (1) a characterization of the herbaceous inter- and intra
tussock flora, (2) a productivity assessment of the herbaceous pajonal components under
various treatments, and (3) a study of soil seed banks and erodibility. The biology and
productivity of Festuca orthophylla was investigated in a parallel PhD project (by José
Monteiro).
(1) Biomass allocation in herbaceous plants under grazing impact in the high semi-arid
Andes
Besides developmental controls, assimilatory and respiratory processes as well as allocation
of photo-assimilates to certain plant compartments are driving the rate of plant growth. These
relationships are explored in what has been termed 'functional growth analysis'. Functional
growth analysis considers (1) the relative amounts of dry matter invested in certain plant
compartments and (2) the density and functional duration (amortization) of resultant tissues as
the two major drivers of plant growth. While plant dry matter allocation is to a great part
genetically determined for a given species, environmental influences can still cause
significant shifts in investments towards needed structures. In this study we explored plant
biomass allocation in perennial herbaceous plants growing in semi-arid, high elevation sites in
the subtropical Andes. We hypothesized that under similar cold life conditions increased
herbivory pressure leads to more pronounced below-ground biomass allocation and that heavy
camelid pressure with a cold climate and water shortage favours short rotation high N foliage.
The study was carried out in two sites in the subtropical Andes which differed mainly in
grazing pressure (higher in Bolivia) but were very similar in climate conditions: One in NW
Argentina, Cumbres Calchaquíes, (26°40’ S, 65°44’ W). This high plateau at 4200-4600 m
elevation receives on average 385 mm rainfall per year. The mean air temperature during the
main growing season (November - March) is 4.5° C, and -1.6° C in winter, guanacos (wild
camelids) represented the main grazers in this area. The other area (described above) the
Sajama National Park area is under heavy grazing pressure, largely by llamas and alpacas, but
also sheep and vicuñas. For our comparison we used three main biomass compartments:
'leaves', 'stem', and 'below-ground'; where 'stems' also include reproductive organs (mainly
flowers), and 'below-ground' includes fine roots and storage organs.
The inter-tussock space in these open, dry plains ('pajonal') was dominated by rosette forming
species with a below-ground shoot apex. We found less investment in leaves at these semiarid
sites as compared to other alpine regions, and a massive below-ground storage
compartment (rhizomes, tap roots, 70% of all species), particularly at the colder Argentinean
site, with a mean leaf mass fraction of only 11%. Though grazing pressure was much greater
in Bolivia, the foliage mass fraction was larger than in Argentina. The storage organs of these
species represented more than 50% of total biomass. In both areas the combined action of
freezing conditions (down to -15.4 °C in Sajama) and grazing must have selected for these
massive below ground structures, with a greater presumed effect of frost in the much colder
site in Argentina and a greater presumed effect of grazing pressure in Bolivia.
Patty, L., Hiltbrunner, E., Körner, Ch. 2010. Biomass allocation in herbaceous plants under grazing
impact in the high semi-arid Andes. Flora 205: 695-703
(2) Herbaceous species responses to grazing and fire in the Bolivian Altiplano
The semi-arid grassland of the Bolivian Altiplano is dominated by tall tussocks of Festuca
orthophylla, the tiny, mostly perennial, herbaceous species are emerging during the rainy
season only and represent high quality fodder due to their high leaf nitrogen concentrations
and total non-structural carbohydrates (NSC). During the last decades, over-grazing (increase
in livestock) and frequent burning have increased, with Bolivian stocks alone representing c.
63% of the South American llama population. The stocking density of llamas has gone up
threefold in the last 20 years (fivefold in alpacas). This part of the project explores the
responses of these herbaceous species to grazing, fire and dung addition, in terms of diversity
and productivity as well as forage quality (leaf N and NSC concentration).
In 2006, eight plots of 25 x 25 m were established (4 fenced and 4 unfenced) and the effects
of fire (one initial burn in April 2006) and dung were tested in subplots (nested design). In
this study we identified two types of non-tussock vegetation: the inter-tussock species mainly
small rosettes and minor graminoids growing between Festuca orthophylla tussocks and the
intra-tussock vegetation, growing inside tussocks.
Even after 3 years of animal exclosure, herbaceous species regeneration was minor, and there
were only small absolute increases in productivity and diversity. The impact of grazing and
associated trampling was far greater on these minor species as compared to Festuca tussocks.
In 2007, a dry year, the aboveground biomass of inter-tussock species in unfenced plots was
20 mg m-2 and 500 mg m-2 in fenced plots; intra-tussock species in unfenced plots arrived at
30 mg m-2, compared to 200 mg m-2 in fenced plots. In 2008, a more normal year, the biomass
was higher, reaching 110 mg m-2 and 300 mg m-2 in unfenced and fenced inter-tussock
vegetation, and 190 mg m-2 versus 500 mg m-2 in unfenced versus fenced intra-tussock
plants. Dung addition without fence had no significant effect on herbaceous species, while the
use of fire in unfenced areas suppressed the herbaceous vegetation. While fencing led to a
clear biomass increase, other treatments only showed a positive response in the presence of
the fence. Annual weather characteristics play an important role in plant species richness and
composition. As the emergence of inter-tussock and intra-tussock species coincides with the
camelids birth season; they can provide protein, sugar and starch, rapidly metabolized and
likely to improve fat and milk production. Leaves of herbaceous species showed higher mean
N (33 mg g-1 d.w.), and NSC (66 mg g-1 d.w.) concentration than Festuca orthophylla leaves
(N 10 mg g-1 d.w.), and NSC 34 mg g-1 d.w. (data for n = 4 plots).
These results indicate that the current land management is highly deleterious for the most
valuable fodder component of this ecosystem. Overstocking by llamas and intentionally set
fires cause a massive decline in diversity and productivity of perennial herbs in this semi-arid
grassland. Taking into account that future climate change may induce higher incidence of
droughts in this region (IPCC 2007), we strongly recommend the periodic reduction of llamas
livestock and banning intentional burning to facilitate regeneration of the non-tussock flora
and maintain or enhance the carrying capacity of the land. Because Festuca tussocks do not
cover the available canopy and rooting space, a large fraction of inter-tussock land remains
unproductive and species poor, a niche that could be occupied by the non-tussock, ephemeral
flora.
(3) Soil seed bank and wind erosion in the semi-arid Bolivian Altiplano
Soil seed banks have been studied because of their importance in ecosystem restoration,
establishment and dynamics of vegetation succession. Several studies in semi-arid areas and
deserts have shown great variability in the size and composition of seed banks, and fire and
grazing can modify both, the size and composition of seed banks. In the Andes, the Altiplano
soil seed bank is poorly known, but there is some evidence that the dry shrub puna offers
favourable microhabitats for the establishment of a seed bank.
Grassland burning is still a common land management practice. After burning patches of bare
soil are exposed during long periods to high wind velocity and drought. The preservation of a
seed bank during and after wind erosion may help stabilizing the land during subsequent
moist periods. In this study, I assessed the soil seed bank in the dry grassland (‘pajonal’)
under the influence of tussock burning and grazing. Wind driven sediment deposition in those
bare soil patches and seed translocation was explored for contrasting land cover by tussocks.
The soil seed bank of adjacent shrubland was also studied for comparison. The study was
conducted during the dry winter months, associated with strong and steady westerly and
north-westerly winds over the entire region. The seed bank was assessed in the 8 test plots
and simultaneously in four plots of 156 m2 each in the shrubland next to the 'pajonal', to
compare the seed quantities between open controls 'pajonal' versus grazed shrubland.
For the seed bank assessment, soil samples were collected with cores of 5 cm diameter at a
5cm depth, and eolian sedimentation was assessed with buried sediments traps. A vertical
profile of local wind velocity was registered in open (bare) and tussock covered land.
The pajonal seed bank consisted of 3% Festuca orthophylla, 1% Calycera, 1% Gamochaeta
and 90 % of all seeds came from the 'bofedal' terrain, brought in by llamas and deposited in
the soil through faeces. In the shrubland around 87% of the seed bank is formed by 'bofedal'
seeds and 13% by Festuca orthophylla. Although, the pajonal showed a higher seed density
than the shrubland, the difference was not statistically significant due to the large variation.
However, total seed biomass per square meter was significantly higher in the pajonal
(P=0.01) than in the shrubland, meaning that the heaviest seeds stay in the grassland. In
unfenced areas of the grassland, seed density was 860 seeds m-2 and in fenced plots 1120
seeds m-2. The use of fire in unfenced areas drastically reduced the seed density to 150 seeds
m-2.
The reduction of tussock density as result of overgrazing enhanced the sediment deposition in
traps by an equivalent of 59 t ha-1 during four extreme wind events (68 km h-1). This quantity
would strongly increase on a longer time base, given that the number of extreme wind events
is c. 10 to 20 day per year in the Altiplano.
It is concluded that the seed bank in the pajonal is extremely depauperate in local, adapted
taxa, and the reduction of Festuca orthophylla tussock cover through overgrazing (trampling)
and burning practices is increasing the eolian erosion and diminishes the herbaceous species
diversity, recruitment and productivity in this area through negative effects on the seed bank.
Given the poor seed bank and the high erodibility of the surface, a regeneration of inter
tussock vegetation will profit from soil protection measures (e.g. mulch by lama dung), higher
tussock density (better regeneration by periodic fencing) and amendments by sowing native
species seed that could be produced in seed farms.
Overall this project revealed that the non-tussock flora is affected most heavily by land
use and that a management for higher carrying capacity would also benefit biodiversity and
fodder quality, while at the same time reducing eolian erosion. The management implications
had been discussed with the local stakeholders and were well received. This Swiss funded
project contributed a case study to the Global Mountain Biodiversity Assessment (GMBA) of
DIVERSITAS (Spehn et al., 2006).