Data from ILAM surveys conducted by ARCOS in Rusumo landscape for year 2016

Study Extent

The surveys were conducted in Rusumo landscape specifically the two administrative sectors (Nyamugari and Kigarama) that surrounds the Rusomo falls on Akagera river.

Sampling

The following is the methodology that was used during the survey: 1. Birds surveys a. Transect Sampling design: Three transects of 2 km long each, were designed; two covering the edges of the wetland and one covering different agro-ecosystems of the study sites in Kirehe landscape. The line transect was the best method to use since the area is open (Owiunji at al. 2005). The transect was long enough to cover the variation of habitat types and land uses (David Hill at al. 2005). b. Point counts: Points were established at an interval of 200 m along transects, as a single station from which bird counts were made with 25 m radius. Each point in this study was visited twice to maximize the sampling efforts (David Hill at al. 2005). Dominant habitat types at each point count was recorded and GPS points were taken for each point to help easier monitoring for the next years. At each point, observers waited for 2 minutes to allow birds to settle down and then record all sightings and calls of birds for a period of 10 minutes. This suits cryptic, shy and skulking species and populations of higher density and species richness in either open or closed habitats (David Hill at al. 2005). Other recorded data included habitat types and threats ranked into 4 categories. The team then moved on to the next point and repeated the same process until a transect is covered. The main target was to use recorded data to produce a bird checklist, measure species richness, and abundance, as well as the impacts of threats to birds species habitat. c. Opportunistic sampling: Opportunistic observations were also used to maximize the number of species encountered in each transect and in the landscape. All bird species seen or heard at different times of the day, outside of point counts transects was recorded and used to update the species list for the surveyed sectors (Owiunji at al. 2005). The analysis has focused on classifying bird species encountered along the transects into ecological categories, summaries of individuals in each habitat category to determine the major groups dominating the different habitats. Shannon Wiener diversity index, and evenness were calculated to compare diversity in different habitat types (A high value of H’ indicates a large number of species with similar abundances, a low value indicates domination by a few species). To assess the conservation status of the surveyed birds, their status on the IUCN Red List and Albertine Rift endemics (ARE) were determined. Ecological categories used in this analysis included: - RB: Resident in the country and breeding recorded confirmed - F: Forest visitor - FF: Forest Specialist-Species typical of forest interiors - NBR: Non-breeding Resident - VS: Visitor/Migrant/Wanderer - NF: Non-forest species - O: Occasional Visitor / - P: Palearctic migrant –a species not breads in Europe or Asia - PAM: Palearctic Migrant - W: Water bird specialist- Normally restricted to wetland or open water - M: Migrant - IAM: Interafrican Migrant - B: Breeding Note that the number of ecological classes is an initial measure of an ecosystem’s wise use while the proportion of classes and their relative abundance are affected by change in ecosystem structure. 2. Butterflies survey Established butterfly monitoring methods are designed for open habitats such as grasslands. Not all rare species occupy habitats that are easy to see across and navigate, in which cases a new approach to monitoring was necessary. The most common methods used for monitoring butterfly populations are mark-recapture and transect counts. Mark-recapture methods are the most rigorous because they allow for estimation of daily and total population sizes, recruitment, survival, and detection probabilities (Haddad et al. 2008). For this ILAM, the goal was to survey butterflies from each habitat type and calculate the species richness and abundance. The team used the same line transects set for birds surveys. maximize our time spent surveying, each site was visited twice, and the survey ‘‘point’’ (the center of the location where we were conducting our butterfly counts) was based at every 200m along that line (Erica at al. 2015). At each survey point, butterflies individuals were either captured using the insect net or a high-resolution camera to make sure that butterflies at the point were detected with certainty (Thomas at al. 2010). A magnifying lens, and a hand-book were used for direct identification of individuals capture by the insect net (Erica at al.2015). We started our survey the moment we arrived at the survey point and recorded any butterflies species flushed from the point upon approach as detected at the start of the survey period which was fixed at 10 min to not allow much movements of butterflies toward the observers and hence limit the fact of overestimating the abundance and species richness (Buckland 2006). We have only recorded butterflies that were distant from the observer in 2 m intervals for more precision and accuracy (Thomas et al. 2010). All counts were done from 9:30 to 16:30 to reduce factors that can dramatically influence detectability like weather conditions (Erica at al 2015). Data were treated using Microsoft excel to calculate the Shannon-Weiner (H’) diversity index, the Evenness (J’) as a measure of species richness in different habitat types. Biodiversity Professional (Lambshead et al. 1977) was used to plot the rarefaction curve for comparing the species richness in different habitat types and landscapes (Teodorescu and Cogălniceanu, 2002). 3 Dragonflies and damselflies survey Transects can be counted every week, two weeks, or monthly from May to September, or just during the relevant flight period for rare species (British dragonfly society, 2009). For our ILAM survey, multiple sample sites within different natural habitats (wetlands), agro-ecosystems and residential areas were sampled for dragonflies and damselflies in Kirehe landscape and Visual encounter surveys of adult odonates were carried out from the month of March to June 2016 randomly in morning 9:30 am to 12:00 and afternoon from 3pm to 5:30 pm with fine weather conditions. The counting zones were 5m out from the water’s edge on one side and 2m on either side of the transect in agroecosystem areas (British dragonfly society, 2009). The survey was conducted along a 2km long transect that could cover different types of habitats, and a scan was made in each surveying center for 10 minutes. All the dragonflies and damselflies observed were identified visually with the aid of a pair of close-focus binoculars or caught with an aerial net when necessary and identified using the field guide. Pictures of dragonflies and damselflies were taken to help in subsequent identification. Subsequent captures were identified and released from the insect nets.

Method steps

1. The steps that were followed during the surveys were different depending on the taxa at hand as described in the sampling description section above.

This dataset covers birds, odonata (dragonflies and damselflies) and butterflies.

1. Odonata
   common name: Dragonflies and Damselflies rank: order
2. Birds
   rank: class
3. Butterflies
   common name: Butterflies rank: order

Rusumo landscape is located in downstream part of Akagera basin and is characterized by lowly undulating hills separated by valleys some of which are swampy. The area is generally dominated by farmland, with few wetlands and woodland (mostly eucalyptus woodlots). The natural forests which are disappearing completely are characterized by the savanna type with a variety of trees dominated by acacia. Communities in the landscape are settled into agglomerations commonly called Umudugudu. According to the Kirehe district's development plan, the Rusumo landscape has climatic intervals of four seasons per year making it possible to make two annual harvests on the same parcel of land. Agriculture is strongly dependent on the seasonal climatic changes, primarily with regards to the rainfall.