Field data on Vegetation Structure and Effects of Human Use of the Dambos Ecosystem in Northern Mozambique

The data content of this paper is related to the original research article entitled “Vegetation Structure and Effects of Human Use of the Dambo Ecosystem in Northern Mozambique” that was published in the Global Ecology and Conservation. Woody and grass vegetation was inventoried in the dambos wetlands of the Niassa National Reserve (NNR), the largest Protected Area (PA) in Mozambique and the third largest in Africa. The six dambos assessed were selected through Google Earth, MODIS satellite images and exploratory field visits. The selected dambos were surveyed using a two-stage systematic sampling procedure in which woody vegetation was inventoried by means of transects, and the grass was inventoried using quadratic sub-plots laid down within the transects. The woody vegetation survey included the identification of all individuals to the species level, measurement of total height and diameter at breast height (DBH). The grass vegetation survey consisted of measurement of the total height and species identification within sub-plots. Woody vegetation data in this article includes also estimation of total richness, absolute and relative abundance, dominance, frequency, species volume and successional stage of each species in the vertical structure. Estimation of richness and absolute dominance is also presented for the grass vegetation.

individuals to the species level, measurement of total height and diameter at breast height (DBH). The grass vegetation survey consisted of measurement of the total height and species identification within sub-plots. Woody vegetation data in this article includes also estimation of total richness, absolute and relative abundance, dominance, frequency, species volume and successional stage of each species in the vertical structure. Estimation of richness and absolute dominance is also presented for the grass vegetation.

Data
Information of the grass species, that includes species richness and absolute ðAbÞ dominance in each dambo and the whole ecosystem of dambos assessed, is provided in the Table 1. Table 2 presents detailed information related to the species and families of woody vegetation. Woody vegetation data, includes estimation of absolute and relative abundance (Ab and ArÞ, dominance ðDa and DrÞ, frequency Specifications Table   Subject area Ecology More specific subject area Biodiversity conservation Type of data Table, Excel files and figure How data was acquired Field Survey Data format Analyzed and Raw data Experimental factors The main criteria that was used to select the dambos sampled were the following: predominance of grass vegetation (canopy cover less than 20%); Location closer to the Mbatamila main office and seasonal or permanent water dambos.

Experimental features
Data were collected using a two-stage systematic sampling procedure. In the first stage, height and diameters of tree and shrub vegetation was collected in main transects of 0.1 ha, after its identification. In a second stage, grass vegetation was counted, height and collected for later identification in six square subplots of 0.25m 2  Aristida diffusa 0 0 0 0 0 8 8 12 Aristida junciformis (continued on next page)   (Fa and FrÞ and volume and representativeness of each species in the vertical strata. Importance Value Index (IVI), was also computed in order to have a broader picture of the position of each species in the structure of the dambos [1,2]. Table 3 presented the location and the general characterization of all dambos assessed. While in the Fig. 1, represents the sampling scheme used to collect data on trees, shrub and grass and vegetation In the main transects and subplots respectively.

Experimental design, materials and methods
The six dambos sampled in this study were selected using Google Earth and MODIS satellite images. After the identification of the dambos, an exploratory field trip was made to verify whether the candidate dambos were appropriate for the establishment of the survey plots. The selection was required to offer a representative sample size and proximity to the Mbatamila Center Office of the Reserve, due to budget constraints and poor road access. A preliminary, basic characterization was conducted in each dambo, which consisted of assessing the occurrence of fire in the last two years, shifting cultivation (Sc), artisanal fishing (Af), soils and vegetation characteristics, seasonal water (Sw)  or permanent water (Pw). Table 3 presents detailed information regarding the location and characteristics of the selected dambos. The occurrence of grass vegetation was dominant in all dambos, which is a defining characteristic of these ecosystems, as described in the literature [3]. Fire also occurred in all dambos in the recent years. Data were collected using a two-stage systematic sampling procedure. In the first stage, tree and shrub vegetation information was collected in transects of 100 Â 10 m (0.1 ha), established perpendicular to the length of the dambo. The center of the plot was marked after identifying the bottom of the dambo in the middle of the transect, extending 50 m to each side. Because the distance between transects within the dambos was 300 m, the number of transects established in each dambo varied, depending on the dambo's size. In total, 58 transects were established and surveyed. In a second stage, grass vegetation was counted, height measured and collected for later identification in six square subplots of 0.25m 2 (50 Â 50 cm), established within the main transects, according to Tito et al. (2009) [4]. The distance between each pair of subplots was about 15 m. Thus, a total of 336 subplots were established in all transects (see Fig. 1).
The tree heights were measured with the support of a hypsometric bar and Vertex when necessary, whereas the diameters (dbh and D) were measured with a measuring tape. Subsequently, each stem was identified to species and family and recorded in the field, based on authoritative field guides to of trees of southern Africa [5] and the grasses of southern Africa [6]. For the species that were difficult to identify in the field, samples were collected for later identification by a botanist.
The successional stage of each species in the vertical structure was analyzed according to its position, by dividing the forest canopy in three main strata, namely: lower, middle and upper, based on the variable height (h), according to the following equation: lower (us) h j < ðh À SÞ, middle (ms) ðh ÀSÞ h j < ðh þSÞ and upper (ls) h i ! ðh þ SÞ, where h is the mean height of all trees in a given sample, S is the standard deviation of h in a given sample and h i is the total height of j-th individual tree. According to Hosokawa et al. (2008) [1], a given species is well placed in the forest when it is well represented in all forest strata, with a large proportion of trees in the lower stratum. More information regarding the data collection and analysis is provided in Mbanze et al. (2019) [7].