Arbuscular mycorrhizal infection in Cyperaceae and Gramineae from natural,disturbed and restored savannas in La Gran Sabana,Venezuela

 The mycorrhizae of a tropical savanna growing in oligotrophic and stony soils were compared with those of a disturbed area that had been reclaimed with introduced species and of an area that was disturbed but not revegetated. All were compared with natural regeneration in a savanna that had been disturbed 12 years previously. Arbuscular mycorrhizae (AM) were common in savannas. Cyperaceae species, which were codominant with Graminaea, showed high levels of infection frequency (45%) like the Gramineae (61%). Arbuscules observed in the Cyperaceae indicated functionality. There were few plants in disturbed, nonrevegetated sites, but those present had AM. Observations of roots from soil monoliths showed that AM were present in disturbed areas, but compared with natural, succesional and revegetated savanna had a lower infection frequency (48–59% vs 75%), lower intensity (10–15% vs 25%) and a lower percentage of arbuscules (0.7–0.8% vs 2.3%). The percentage of vesicles was also lower in succesional savanna than in natural savanna (1.6% vs 4.8%). The revegetated site had the highest percentage of vesicles (6.6%). Although a high frequency of mycorrhizal infection has been reestablished in disturbed areas, the intensity and structure of the infection suggests that mycorrhizal function has not been restored to the original levels. We hypothesize that neither plants nor fungi have adapted to the new edaphic conditions. Accepted: 10 July 1995     

Direct and indirect effects of termites on savanna tree-seedling growth

Termites are considered to be ecosystem engineers because they modify their biophysical environments. We tested the effects of soil-nutrient alteration in termite-enriched soils compared with termite-free open savannas. We also tested whether non-nutrient alterations (soil disturbance) induced by termites led to changes in tree-seedling growth. Soil samples taken from termite-enriched soils and adjacent open savanna sites in KwaZulu-Natal, South Africa were analyzed for nitrogen, pH, organic carbon and water-holding capacity. Seeds from three dominant tree species, Acacia sieberiana, Celtis africana and Ziziphus mucronata, were grown in soils taken from termite-enriched soils and adjacent savannas. Overall, organic carbon and nitrogen content were higher in termite-enriched soils than in adjacent savannas. We found that these differences in nutrients did not directly affect seedling growth rates or final height. However, C. africana had increased growth rates in compacted termite-enriched soils, while A. sieberiana seedlings were taller in non-compacted soils. We conclude that the indirect effects of disturbance by termites may be as important as the direct effects of increased nutrients for growth of savanna trees.     

Restoration of Midwestern U.S. Savannas: One Size Does Not Fit All

Lowland savannas are a rare variant of Midwestern United States savanna occurring on alluvial soils, for which reference information is sparse. To evaluate the appropriateness of using upland savanna as a surrogate source of reference information for lowland savanna, we studied a pre‐Euro‐American lowland savanna using original U.S. Public Land Survey data and other historical records. Historical vegetation was reconstructed and compared among upland savannas, lowland savannas, and lowland forests; we also evaluated potential disturbance dynamics maintaining these systems. We found that all three communities were dominated by members of the genus Quercus but also had extensive representation by many other tree species, especially notable for savannas in this region. There were no clear size–density relationships for species in the genus Quercus, indicating that these historical savannas were not characterized exclusively by large, scattered oak trees but rather by trees of many oak species and nonoak species in a wide range of size classes. Both upland and lowland savannas also contained a substantial shrub component. We found no evidence that lowland savannas were maintained by flooding, although the uneven‐aged canopy structure suggested that periodic disturbance occurred. Restoration of lowland savanna in this region should include provisions for maintaining nonoak species and shrubs, with disturbance timed to maintain an uneven‐aged canopy structure. Although the appropriateness of historical data in the face of climate change may be questionable, in this region, a warmer climate may actually help promote the “oak parkland” that was present from 8,000 BP up to Euro‐American settlement.     

Nutrient dynamics and plant assemblages of Macrotermes falciger mounds in a savanna ecosystem

Termites through mound construction and foraging activities contribute significantly to carbon and nutrient fluxes in nutrient-poor savannas. Despite this recognition, studies on the influence of termite mounds on carbon and nitrogen dynamics in sub-tropical savannas are limited. In this regard, we examined soil nutrient concentrations, organic carbon and nitrogen mineralization in incubation experiments in mounds of Macrotermes falciger and surrounding soils of sub-tropical savanna, northeast Zimbabwe. We also addressed whether termite mounds altered the plant community and if effects were similar across functional groups i.e. grasses, forbs or woody plants. Mound soils had significantly higher silt and clay content, pH and concentrations of calcium (Ca), magnesium (Mg), potassium (K), organic carbon (C), ammonium (NH₄⁺) and nitrate (NO₃⁻) than surrounding soils, with marginal differences in phosphorus (P) and sodium (Na) between mounds and matrix soils. Nutrient enrichment increased by a factor ranging from 1.5 for C, 4.9 for Mg up to 10.3 for Ca. Although C mineralization, nitrification and nitrification fraction were similar between mounds and matrix soils, nitrogen mineralization was elevated on mounds relative to surrounding matrix soils. As a result, termite mounds supported unique plant communities rich and abundant in woody species but less diverse in grasses and forbs than the surrounding savanna matrix in response to mound-induced shifts in soil parameters specifically increased clay content, drainage and water availability, nutrient status and base cation (mainly Ca, Mg and Na) concentration. In conclusion, by altering soil properties such as texture, moisture content and nutrient status, termite mounds can alter the structure and composition of sub-tropical savanna plant communities, and these results are consistent with findings in other savanna systems suggesting that increase in soil clay content, nutrient status and associated changes in the plant community assemblage may be a general property of mound building termites.     

The savannization of moist forests in the Sierra Nevada de Santa Marta, Colombia

In the Rio Ranchería watershed of the Sierra Nevada de Santa Marta, between 500 and 1500 m, savanna vegetation is interspersed with moist forests. The savannas are composed of native savanna grasses like Aristida adscensionis L., Arundinella sp., Panicum olyroides Kunth, and Schyzachyrium microstachyum (Desv.) Roseng., Arrill & Izag and the African Melinis minutiflora P. Beauv. There is also Curatella americana L. and Byrsonima crassifolia (L.) H.B.K., two typical tree species of the neotropical savannas. Although moist forest patches occur more often on lower slopes and narrow valley bottoms, they can also be found on mid- and upper-slopes and less often on ridges. Thus, these forest patches are not gallery forests as are found throughout the neotropics, but the result of deforestation and fractionation of a continuous forest. A comparison of soil profiles between the savannas and remnant forest patches on the same slope, showed the disappearance of the A and B horizons (approx. 50 cm) under savanna vegetation. The sharp difference between the savanna and forest soils at the Rio Ranchería does not appear to be due to a change in soil water status along a toposequence or differences in the underlying bedrock. We hypothesize that the savannas of the Rio Ranchería watershed, are the result of deforestation and land practices on infertile soils derived from granite. The savannization process was likely initiated by Amerindians by means of the frequent use of fire or clearing lands for the cultivation of maize. The introduction of cattle by Spaniards (c. 1530) and the frequent use of fire to maintain grazing fields, contributed to further degradation of the habitat. While some tropical landscapes recovered their forest cover when human pressure was removed approximately 500 years ago, areas such as the Rio Ranchería watershed have suffered permanent damage. The savannas of this region are likely to remain unless fire is suppressed and soil restoration practices implemented.     

When the same is not the same: phenotypic variation reveals different plant ecological strategies within species occurring in distinct Neotropical savanna habitats

Variations in abiotic characteristics such as soil water availability and fertility impose different selective pressures on plant populations. This may produce intraspecific variability in functional traits, even at a fine spatial scale. We investigated whether functional traits related to water-use efficiency, resource-retention strategy, soil nutrient acquisition, and fire tolerance differ in species that occur in two different habitats of Brazilian Cerrado: rocky savannas and savanna woodlands. Rocky savannas occur over sandstone, quartzite outcrops and have shallow nutrient-poor and low-moisture rocky soils, while savanna woodlands occur over well-drained and deep soils with frequent fire regimes. We measured nine functional traits of 40 tree species that occur in both habitats. Rocky savanna individuals exhibited a greater water-use efficiency strategy. The resource-retention strategy in rocky savanna individuals was corroborated by lower adult maximum height. However, despite the lower nutrient availability in rocky savanna soils, we only detected lower leaf phosphorus content in individuals from this habitat. Furthermore, individuals from both habitats had equally thick bark, suggesting that the fire-defense strategy is related to a stable, rather than plastic trait. Overall, our results highlight the central role of contrasting soil water availability patterns in driving phenotypic plasticity within species. We conclude that savanna species are responding to water and nutrient availabilities, via plasticity in traits related to the resource-retention strategy, and preparing for future fires, via uniformly thick bark. Wide plant distribution in contrasting habitats is possible for species that can shift ecological strategies to survive in nutrient- and water-limited habitats such as rocky savannas.     

Is the lack of leguminous savanna trees in grasslands of South Africa related to nutritional constraints?

As with many grasslands globally, the Highveld grasslands of South Africa are tree-less, despite having a climate that can support tree growth. Models predict that fire maintains these grasslands. The question arises as to why fire-tolerant savanna trees do not survive in these ecosystems? Savanna tree survival in mesic areas is restricted by demographic bottlenecks, specifically limitations to sapling-escape from fire. It was hypothesised that ancient highly leached soils from grassland areas would prevent saplings from growing fast enough to escape the fire-trap. Growth rates of savanna tree seedlings (Acacia karroo Hayne and Acacia sieberiana Burtt Davy) were measured in a common garden experiment using soils from ten sites collected along a savanna-grassland continuum. Soils from grassland sites were relatively nutrient-poor compared to those from savannas with lower pH, and associated cations. A. sieberiana growth rates responded to pH and these nutrients, whereas A. karroo growth was less strongly linked to specific nutrients. Even so, both species accumulated more biomass when grown in soils from savanna sites compared to grassland sites. An exception was a low elevation low nutrient savanna site that resulted in poor growth, yet sustains high tree biomass in situ. Differences between growth in grassland and savanna soils were small. They may contribute to, but are unlikely to explain, the treeless nature of these grasslands.     

Reproductive phenology of two co‐occurring Neotropical mountain grasslands

Aim

Climate tends to explain phenological variations in tropical ecosystems. However, water availability and nutrient content in soil strongly affect plant communities, especially those on old, climatically buffered, infertile landscapes (OCBILs), and may impact these ecosystems’ plant reproductive phenology over time. Here, we compare the reproductive phenology of sandy and stony tropical grasslands, two co‐occurring herbaceous communities of the campo rupestreOCBILs. We asked whether flowering, fruiting and dispersal are seasonal in both grasslands, and whether these phenophases differ due to variations in soil properties. We also asked whether the phenological strategies and the number of flowers and fruits differ between these two grasslands as soil conditions vary.

Location

Serra do Cipó, Minas Gerais, Brazil.

Methods

The phenology of herbaceous species of sandy and stony grasslands was monitored monthly over two consecutive years.

Results

Plants on sandy and stony grasslands flowered and fruited throughout the year. We did not find a distinct seasonal pattern at the community level of either studied grassland. However, flowering, fruiting and seed dissemination occurred in stony grasslands mainly during the rainy season, while sandy grassland species flowered in both seasons and fruited and disseminated seed mainly during the dry season, as observed in other savanna vegetation types in the Cerrado. Flower and fruit production was higher in sandy grasslands than in stony grasslands, which may be linked to higher water retention in sandy grassland soils. In both communities, species of Cyperaceae, Eriocaulaceae and Xyridaceae contributed most to overall production, whereas Poaceae and Velloziaceae, two important families in campo rupestre, barely participated in the reproductive phenology during our 2‐yr survey.

Conclusions

Despite a strong seasonal climate, there was no reproductive seasonal pattern at the community level in campo rupestre. This first investigation of Neotropical grassland phenology indicates that the differences in soil content may constrain the grassland reproductive phenology and restrict reproduction of stony grassland species to the most favourable season. Further studies of grassland phenology are necessary to disentangle the relative importance of soil, climate and other triggers, especially fire.     

Modern pollen rain in savanna and forest ecosystems of Gabon and Cameroon,Central Atlantic Africa

Eighty modern soil surface and litter samples from southern Cameroon and Gabon, Central Atlantic Africa (5°N–4°S, 10°–15°W), were analysed for pollen content. The samples are distributed among two main vegetation types: savanna (8 samples) and forest (71 samples). The aim of this study is to provide new data on the modern pollen rain in the Guineo-Congolian phytogeographical region, mainly in forest communities (secondary and mature forests on well drained soils, and hygrophilous forests) and to interpret these data using diagrams of pollen percentages and numerical analyses. The savannas are well identified by high frequencies of non-arboreal pollen with as pollen marker the Poaceae, and the forests by high frequencies of arboreal pollen with as important families the Burseraceae, Caesalpiniaceae, Mimosaceae, Euphorbiaceae and Sapindaceae. Within the forest ecosystem, secondary and mature forests on well drained soils can be differentiated on the basis of distinct assemblages of tree pollen taxa such as Zanthoxylum, Phyllanthus, Tetrorchidium, Margaritaria discoidea in secondary forest spectra and abundance of Burseraceae, Caesalpiniaceae, Sapindaceae in mature forest ones. In addition, hygrophilous forests are well identified by the presence of high pollen contributors such as Uapaca, Nauclea, Macaranga and Raphia. This work shows that the major vegetation communities occurring today in Cameroon and Gabon can be well differentiated by their pollen assemblages.     

Composition and species diversity of pine-wiregrass savannas of the Green Swamp,North Carolina

Fire-maintained, species-rich pine-wiregrass savannas in the Green Swamp, North Carolina were sampled over their natural range of environmental conditions and fire frequencies. Species composition, species richness, diversity (Exp H′, 1/C), and aboveground production were documented and fertilization experiments conducted to assess possible mechanisms for the maintenance of high species diversity in these communities. Although savanna composition varies continuously, DECORANA ordination and TWINSPAN classification of 21 sites facilitated recognition of 3 community types: dry, mesic, and wet savannas. These savannas are remarkably species-rich with up to 42 species/0.25 m² and 84 species/625 m². Maximum richness occurred on mesic, annually burned sites. Aboveground production, reported as peak standing crop, was only 293 g · m^?2 on a frequently burned mesic savanna but was significantly higher (375 g · m^?2) on an infrequently burned mesic site. Production values from fertilized high and low fire frequency sites were equivalent. Monthly harvest samples showed that savanna biomass composition by species groups did not vary seasonally, but within groups the relative importance of species showed clear phenological progressions. The variation in species richness with fire frequency is consistent with non-equilibrium theories of species diversity, while phenological variation in production among similar species and the changing species composition across the moisture gradient suggest the importance of equilibrium processes for maintenance of savanna diversity.     

Soil disturbance favours threatened beetle species in sandy grasslands

Soil disturbance is recognised as an important restoration measure for conserving biodiversity in sandy soils. We used a soil disturbance (ploughing) experiment in a sandy grassland as well as a semi-natural disturbance (slope erosion enhanced by cattle trampling) gradient on a sandy slope to test the soil disturbance effects on the ground-living beetle community. Both experimental disturbance and semi-natural disturbance favoured sandy grassland specialists, but there was no overall effect on beetle richness and abundance. Amara lucida and Harpalus spp. were favoured by disturbance while Calathus melanocephalus was disfavoured. Experimental ploughing significantly increased the proportion of red-listed species in disturbed plots compared to non-disturbed controls. In the semi-natural disturbance gradient we found that the beetle community on the disturbed slope differed from that of the flat areas, and there were tendencies for a higher proportion of red-listed species on the slope. We conclude that increasing the area of bare sand in sandy grasslands can have positive effects on many threatened species. Soil disturbance should thus be included as a regular measure in sandy grasslands under conservation management and as a measure to restore high biodiversity in areas where bare sand is rare.     

Does size matter? Conservation implications of differing woody population sizes with equivalent occurrence and diversity of species for threatened savanna habitats

The Cerrado stands out from other savannas of the world for hosting the highest biodiversity and for its ongoing massive destruction. Savannas on plain relief and with deep soils (DS) in Central Brazil have been widely converted into agropastoral ecosystems. In contrast, savannas on steep relief with shallow and rocky soils (RS) have been considered as future biodiversity refuges in Central Brazil. In this study, we show that adjacent DS and RS savanna sites (each with ten 1-ha plots) differ based on their woody species population sizes, but not based on species occurrence or diversity. In addition, we discuss the implications of these results for species conservation. Our research indicates that the DS and RS savannas are complementary, but not equivalent, regarding their representation of savanna vegetation in Central Brazil. We hypothesize that if RS savannas become the only refuges for savannic vegetation, widespread biodiversity losses will occur in the short term (through loss of exclusive and habitat-specialist species) and long term (due to fragmentation, reductions in population size and loss of genetic variability). Thus, we suggest that the consideration of savannas on distinct substrates should be considered for improving conservation decision-making and initiatives and can be beneficial when expanding or creating new conservation units.     

Termitaria vs. mistletoe: Effects on soil properties and plant structure in a semi-arid savanna

Termite mounds by creating patches of increased resource availability (e.g. water and nutrients) are a major source of spatial heterogeneity in savannas. Likewise, mistletoes via input of nutrient-rich litter alter nutrient and water availability increasing environmental heterogeneity in semi-arid savanna. Despite this recognition, the influence of termitaria and mistletoe on soil properties and plant community have not been investigated together. We established eight 100 m² plots each on termitaria, under mistletoe-infected trees and in the surrounding savanna and examined the soil properties and the structure of Securinega virosa (Euphorbiaceae) and Euclea divinorum (Ebenaceae) in semi-arid savanna, southwest Zimbabwe. Soil properties significantly differed among the sampling sites (p = 0.001) with soils of increasing clay, soil moisture, pH and phosphorus, calcium and ammonium concentrations occurring on termite mounds. Soils under mistletoe-infected trees were associated with silt, organic matter, sodium, potassium, magnesium and nitrate and the surrounding savanna was associated with soils of increasing sand content. Plant structure also differed significantly between sites with greater basal area of both S. virosa and E. divinorum on termitaria relative to mistletoe-infected trees and the surrounding savanna. However, the stem density of S. virosa was greater under mistletoe-infected trees than on termitaria and in the surrounding savanna. Plant structural variables of individuals of the same species were affected by different soil properties across treatments. The major patterns showed that plant structure was influenced positively by soil moisture and nitrate and negatively by phosphorus on termitaria; positively by clay, soil moisture and ammonium and negatively by potassium under mistletoe-infected trees; and by phosphorus and calcium in the surrounding savanna. These findings show that soil properties, plant structure and their relationships differ between termitaria, mistletoe-infected trees and surrounding savanna, and these differences are suggested to increase heterogeneity in soil resources availability and vegetation structure in semi-arid savanna.     

Tiger-Moths in Savannas in Eastern Amazon: First Assessment of Diversity and Seasonal Aspects

Biodiversity knowledge on insects is urgently needed due to the ever growing demand for food and the consequent deforestation process and loss of natural habitats in many understudied tropical regions. In this paper, we describe the outcome of a biodiversity research on tiger moths performed for the first time in a poorly studied Amazonian landscape—the savanna. We sampled tiger moths monthly with UV automatic light traps for 12 consecutive months in two sampling points in an area of savanna in eastern Amazon, and we compared our results to previously available data for eastern Amazon. We found a total of 91 species of which 80 were identified to species level. The most species-rich subtribes were Phaegopterina and Euchromiina with 32 species each. Species richness and abundance did not differ among sampling sites, but in general the species richness was higher during the dry season while abundance was higher during the wet season. This seasonal diversity pattern differs from the most common patterns recorded for savannas in other parts of the world. The species composition also changed in wet and dry seasons and correlated significantly with temperature and relative humidity. Our results suggest that the alpha diversity of the Amazonian savannas in our sampling area is lower than that in nearby rain forests and similar to that in agriculturally disturbed areas surrounded by rain forests. However, the species composition differed considerably from natural and disturbed areas. These results highlight the need of basic biodiversity surveys of insects in Amazonian savannas.     

Effect of woody vegetation clearing on nutrient and carbon relations of semi-arid dystrophic savanna

Throughout the savanna biome, woody vegetation is cleared to increase productivity of herbaceous pasture. While clearing can result in increased pasture production of semi-arid dystrophic savannas in the short term, it is uncertain whether production is sustained in the long term. There is insufficient knowledge of how clearing affects soil nutrient and organic carbon (SOC) stocks. Using cleared-uncleared site pairs, we evaluated techniques for time-integrated assessment of nutrient and carbon relations in Australian savanna. Short-term in situ resin incubation showed that soil at cleared sites had a higher time-integrated availability of ammonium and nitrate, indicating that nitrogen (N) may turn over faster and/or is taken up slower at cleared sites than uncleared savanna. Nitrate and ammonium availability was approximately 2-fold higher in spring than in summer, likely due to greater uptake and/or loss of nitrate during summer rains. Nitrate was a prominent N source for evergreen trees, especially before summer rain, pointing to a role of trees as permanent N sinks. Stable isotope signatures of soil and vegetation indicate that N input occurs via N₂ fixing microbiotic crusts and Acacia species. 30 years after clearing, SOC contained more C₄ grass-derived carbon than uncleared savanna, but this shift in C source was not associated with the net C gain often observed in grasslands. Interactions between altered nutrient and C relations and composition of the understorey should be assessed in context of introduced buffelgrass (Cenchrus ciliaris) which had higher macronutrient concentrations than native grasses. Heterogeneity of the studied soils highlights the need for replication at several spatial scales to infer long-term dynamics with space-for-time chronosequences. We conclude that the techniques presented here are useful for gaining knowledge of the biogeochemical processes governing savannas and the systems that result from clearing.     

Structural relationships between form factor,wood density,and biomass in African savanna woodlands

Key message

Variation in tree biomass among African savanna species of equal size is driven by a wide inter-specific variation in wood specific gravity.

Abstract

Tree form and taper is a fundamental component of tree structure and has been used for over a century in forestry to estimate timber yields and in ecological theories of scaling laws. Here, we investigate variation in form factor in the context of biomass in African savannas. Biomass is a fundamental metric of vegetation state, yet in African savannas it remains unclear whether variation in form factor F (taper) or wood specific gravity (G) is a more dominant driver of biomass differences between tree species of equal stem diameter and height. Improving our knowledge of vertical mass distribution in savanna trees provides insight into differences in life strategies, such as tradeoffs between production, disturbance avoidance, and water storage. Here, we destructively harvested 782 stems in a savanna woodland near Kruger National Park, South Africa, and measured whole tree wet mass, wood specific gravity, water content, and form factor. We found that three of four dominant species can vary in mass by over twofold, yet inter-specific variation in taper was low and taper did not vary significantly between common species (P > 0.05) (species-mean form factors ranged from F = 0.57 to 0.77, where cone F =  $0.\bar{3}$ , quadratic paraboloid F = 0.5, cylinder F = 1.0). Comparison of a general biomass allometry model to species-specific models supported the conclusion that the large difference in biomass between species of the same size was explained almost entirely (R ² = 0.97) by including species-mean G with D and H in a general allometric equation, where F was constant. Our results suggest that inter-specific variation in wood density, not form factor, is the primary driver of biomass differences between species of the same size. We also determined that a simple analytical volume-filling model accurately relates wood specific gravity of these species to their water and gas content (R ² = 0.68). These results indicate which species use a wide spectrum of water storage strategies in savanna woodlands, adhering to a trade-off between the benefits of denser wood or increased water storage.     

Patterns of codon usage bias in three dicot and four monocot plant species

Codon usage in nuclear genes of four monocot and three dicot species was analyzed to find general patterns in codon choice of plant species. Codon bias was correlated with GC content at the third codon position. GC contents were higher in monocot species than in dicot species at all codon positions. The high GC contents of monocot species might be the result of relatively strong mutational bias that occurred in the lineage of the Poaceae species. In both dicot and monocot species, the effective number of codons (ENCs) for most genes was similar to that for the expected ENCs based on the GC content at the third codon positions. G and C ending codons were detected as the "preferred" codons in monocot species, as in Drosophila. Also, many "preferred" codons are the same in dicot species. Pyrimidine (C and T) is used more frequently than purine (G and A) in four-fold degenerate codon groups.     

Topsoil translocation for Brazilian savanna restoration: propagation of herbs,shrubs, and trees

Topsoil translocation has been used for vegetation restoration throughout the world, but it has been poorly tested within savannas. This study describes Brazilian savanna (cerrado) regeneration for the first 3 years following topsoil translocation. The topsoil was stripped from 2.5 ha of savanna and spread on 1 ha of an abandoned laterite quarry in the Federal District, Brazil. We assessed vegetation structure and species composition in 18 circular plots (3.14/m²) after 5 and 15 months and in 30 circular plots after 37 months. In the last floristic survey, the coverage of herbs was estimated using the step‐point method. To verify the source of regeneration, a total of 181 shrubs and trees were excavated over the first 2 surveys. After 3 years, 24, 40, and 21 species of herbs, shrubs, and trees, respectively, had been recorded by the surveys. Of the 33 families found, Fabaceae, Poaceae, and Asteraceae were the most representative. At 5 and 15 months, 91 and 83% of the individuals (shrubs and trees combined) were derived from resprouting, respectively. Shrub and tree stem density reached 3.2/m² at 5 months, but declined to 0.5/m² at 37 months. By the final survey, native and exotic grasses completely covered the ground. Topsoil translocation was effective for the propagation of native herbs, shrubs, and trees, despite the need to control invasive grasses. The large number of shrub and tree resprouts from roots suggests that the bud bank is an important component of the topsoil for savanna restoration.     

Response of annual plants and communities to tilling in a semi-arid temperate savanna

Abstract. Despite the obvious importance of savannas, the response of semi-arid temperate savannas to disturbance has hardly been studied. The objective of this study was to determine the response of annual plants and vegetation to severe disturbance (tilling) in the grassland phase of a semi-arid savanna characterized by perennial bunchgrasses. Most species of annual plants responded positively to tilling; however, 25% of the species did not respond or responded negatively. Response of plants and vegetation was more strongly influenced by precipitation than disturbance, even for the severe disturbance used in this study. Tilling-induced differences were negligible during a relatively dry year. Response of annuals to disturbance in semi-arid temperate savannas may not be consistent with response in other systems.     

Vegetation–environment relationships and classification of the seasonal savannas in Venezuela

The seasonal savannas dominated by the C₄ grasses of the genus Trachypogon, are widespread in northern South America. In Venezuela, they extend from the central lowland Llanos to intermediate elevations in the Coastal and Andean Mountains and to the Guiana Plateau (Gran Sabana) in the south. This study aims to classify these savannas and to understand the environmental factors that determine their composition and structure. Thirty-seven sites were sampled, plant cover and density were measured in 376 one square meter quadrats and the importance value index (IVI) was calculated. Climate data were obtained from climatological stations and soil properties were analyzed. The results were classified by clustering and TWINSPAN. Floristic and environmental data were ordered with the canonical correspondence analysis (CCA).Trachypogon savannas are heterogeneous, both floristically and environmentally. An altitudinal gradient of 2000 m separates the intermediate elevation Coastal Mountains (Type I) and the Guiana savannas (Type II) from the lowland Llanos. In the lowlands, sparsely covered and treeless communities on extremely oligotrophic and sandy soils (Type IV) differ structurally and floristically from woody savannas (Type III). The floristics and structure of the savannas respond directly or indirectly to elevation and water and nutrient availability. These responses were evidenced by: (a) floristic composition, as Neotropical lowland savanna species merged with those of the Andean sub-páramo in Type I savannas or with those from the Guianan flora in Type II savannas. (b) Species richness, which ranged from an average of 13.8 per site in sandy Type IV savannas to 38.9 per site in Type I intermediate elevation savannas. (c) Community physiognomy, which changes from an homogeneous herbaceous matrix with dispersed trees in Type III savannas to a shrubby Type I savannas to “treeless” Types II and IV savannas. (d) Plant cover which varied from 93.8% in the fertile Type I savannas to around 80% in the sandy oligotrophic savannas of Types II and IV. Each savanna type is related to one of the major geomorphological regions of Venezuela.