2006-06-05

2008-06-03

The study was conducted along the margins of the lower Tua river, prior to the construction and subsequent flooding of the valley by the Foz Tua Hydroelectric Dam.

The Rio Tua is formed by the confluence of the Rio Rabaçal and Rio Tuela at Mirandela. It flows some 45 kilometres in a mainly south-west direction to Foz Tua where it enters the Rio Douro. At Mirandela it lies at 215 metres a.s.l. falling to about 80 metres at Foz Tua. In the southern part of its course it cuts through hills which ascend to over 600 m, whereas in the northern part the surrounding higher ground only reaches 300 m. This gives the northern half of the river an open character with some wide gravelly shallows. Very much in contrast, the southern half lies in a deep ravine cut through granite rocks which is where the dam has been constructed and where most of the recording sites were located.

Between the bridge at Foz Tua and the bridge at Brunheda, there was no crossing point over the river, which could only be reached by car on tracks that were sometimes very steep. Only from Vilarinho das Azenhas to Mirandela was there a road running along the valley. However a railway line runs the whole length of the river from Foz Tua all the way to Mirandela, providing access to otherwise inaccessible areas.

According to Köppen Climate Classification, the study area was included in type Csa, corresponding to a temperate climate with dry or hot summer. Mean monthly temperature ranges between 6.1ºC (January) and 23.6ºC. Mean annual precipitation is 520mm and follows the typical seasonal Mediterranean pattern, with most (68%) precipitation concentrated in the wet semester (October-March) and virtual no precipitation in the summer months. A brief description of the seven sections of the river that were sampled (11 sampling sites) is provided from downstream to upstream. For each site the altitude of the main collecting sites is given. Any sites below 170m are nowadays submerged.

Section 1 - Fiolhal. The recording site was by the railway line below Quinta da Ribeira, Altitude 130m. The ravine below the line was steep and wooded, with a variety of trees including Fraxinus angustifolia and Quercus rotundifolia. The riverbed was boulder strewn and the banks had Alnus glutinosa and Salix salvifolia. Above the line the slopes were gentler with vines and olive trees growing on terraces, and patches of Mediterranean scrub including such shrubs as Pistacia terebinthus and Lonicera implexa. Two other areas have been sampled during the day: one in the surroundings of Quinta da Ribeira and one closer to the river mouth.

Section 2 - Amieiro. The recording site was a track leading down into the ravine past the village. The track ends at altitude 150 m, beside a cable car which could be used to access the Santa Luzia railway station on the opposite bank of the river. Below the cable the narrow deep river runs at the bottom of a near vertical sided gorge. One light was always placed at a bend in the track at about 170 m. The slope on the Santa Luzia side had rather open Quercus rotundifolia woodland. The Abreiro side also had some oaks, but was occupied by a patchwork of tiny orchard and garden terraces wherever there was any space between granite cliffs and boulders. The trackside vegetation was a rich and varied selection of herbs and shrubs, while rock crevices are filled with Sedum species and Dianthus lusitanus. This site was always noticeably warmer than other sites at night.

Section 3 - São Lourenço. The recording site was beside the railway line about 500 m south of the village at altitude 160 m. Here the steep wooded ravine probably has the least disturbed vegetation of any site along the river. The vegetation was similar to that below the railway line at Fiolhal, but continues on slightly less steep slopes above the line.

Section 4 - Brunheda. The recording site was just north of the railway station at 160 m. There was some flat ground on both sides of the river at this point. Marginal trees were Salix, Alnus and Populus nigra. Trapped lights were placed on the river gravel patches, that were submerged at some times of year by then. The other bank of the river visited by day. A wide grassy area here has abundant Gratiola officinalis.

Section 5 - Abreiro. The trapping site was beside a track at about 170 m, a few hundred metres south of Abreiro station (which is some two kilometres east of Abreiro village on the other side of the river and in a different municipality). The river here was wide and shallow with gravelly flats. Some rocky outcrops near the bank of the river had Bufonia macropetala, a rare plant in Portugal. There was plentiful Salix salvifolia and some Populus nigra by the river. One section of the track was bordered by a rich selection of small trees and shrubs including Prunus spinosa, Crataegus monogyna, Acer monspessulanus, Quercus rotundifolia, Cytisus, Rubus, Ruscus aculeatus and Clematis campaniflora. The embankment of the railway track was however heavily infested with the exotic Ailanthus.

Section 6 - Ribeirinha. Trapping was on the edge of the village beside the river at 185 m. The river here was wide and shallow with gravel banks. There were marginal Populus nigra and Fraxinus angustifolia.

Section 7 - Vilarinho das Azenhas. The site was immediately upstream of the road bridge about one kilometre east of the village, at altitude 195 m. The river here was fairly wide, with an artificial pool created by a weir. The river banks had abundant Alnus glutinosa and Salix salvifolia and near the banks were some Populus nigra, Ulmus minor, Crataegus monogyna and a little Celtis australis. Climbing these trees were Humulus lupulinus, Calystegia sepia and Solanum dulcamara. The narrow strip between the river and the railway opened a little after a few hundred metres into a grassy area containing some vines.

The data was collected within 7 sections of the river Tua valley (Northeast Portugal. Europe), encompassing 11 sampling sites. (Northeast Portugal. Europe). The areas in the region of Trás-os-Montes e Alto Douro, within the municipalities of Alijó, Carrazeda de Ansiães, and Vila Flor.

Sampling methods for Lepidoptera

The majority of the data was obtained from nocturnal sampling using 125 w mercury vapour bulbs suspended from a tripod over a white sheet on the ground or on a stand placed in the middle of the sheet. Three such lights were run from a single generator, using lengths of electric cable to separate the lights as far as possible, allowing sampling from different microhabitats. Egg boxes placed on the sheets provided hiding places for moths, reducing the number of moths that are active around the light at any one moment and making examination of the catch easier. Moths were identified visually, using a lens for the smaller species, but if not recognised or when known to be unidentifiable in the field, samples were collected in glass tubes or boxes and later killed in a freezer. These samples were pinned and taken home for later identification, often requiring dissection of genitalia.

Nocturnal sessions began about 30 minutes after sunset and continued until the rate at which new species were appearing rendered further sampling unprofitable. In cooler more humid localities this was earlier than in dry warm sites, where sampling sometimes continued till the first signs of returning daylight. During these hours, the three sheets were visited and examined nearly continually, with only short breaks, mainly to look at the wine ropes.

Wine ropes were also used at night. These are pieces of clothes line soaked in a solution of white sugar in red wine. Usually five were used, hung on small branches of trees or bushes soon after sunset, and then inspected periodically during the night. They were sited where they would not be directly affected by the mercury vapour lights, either at some distance away from the lights or sometimes between two lights if these were sufficiently far apart. Results from this sampling technique are notoriously unpredictable, but usually some species are attracted that have not been attracted to the lights.

Some diurnal sampling was also carried out, but the time spent on this was much less than on the night work and the methods used less efficacious. A small number of species were captured with a net during the day. Larvae were collected when found, and the food-plants noted. These were reared through to adults for identification purposes, but this was not always successful. A few species were identified from leaf mines or characteristic spinnings made by their larvae. In some cases this can be done even if the larva is no longer present.

In the data, counts are provided for each species. Using the sampling techniques given above, counting every individual of every species at the lights is not practicable with the number of species often far exceeding 100. Instead, during the following morning, estimates from memory were made for each species. Obviously this is not rigorously exact, but it does give an approximation of relative numbers. Using this method, occasional checks can be made, by attempting to count all individuals of one or two species and comparing this with an estimated number. From this it is evident that low numbers are reasonably accurate (thus an estimated 5 might actually be 4 or 6 for example) but more abundant species are consistently underestimated.