Saba senegalensis is a vine that grows up and over the canopy of trees in search of sunlight and out of reach of herbivores. The fruit, called “madd”, ripens from January to June and flowers in the late winter and early Spring (see Figures 14 & 15). Fruit ripens over two seasons. Ripe fruit and flowers may both be present at the same time.
Collected fruit is dried and converted into a powder form. The powder is mixed with sugar and produces a sweet, delicious drink that is usually consumed as a dessert.
There is some association between the presence of the madd vine and sufficient moisture. In the drier areas, madd is associated with riparian areas in the Kolda area but in the book Vegetation et Flore, Parc Transfrontalier Niokolo Badiar (1997), it states that madd also grows in savanna. This poses problems when developing an efficient inventory strategy. We did observe madd vines on our trip from Kolda to Tambacounda in the dry areas of the country. All the madd sites in savanna were in association with either an obvious source of intermittent water or a termite mound.
Out of the 126,000 records in the SIEF-2 database, there are only 13 records of the madd vine (see Appendix A Tables 5 & 6). This shows that the existing sample and plot design is not well suited for recording this species. There have been special studies for this vine in the past and if the need for this information is great, a special study is the only realistic way to obtain sufficient information to help identify the size of the population.
To design an efficient inventory, it is important to identify the objectives. In this case, we could estimate the amount of fruit that is currently going to an external market, amount of fruit harvested from the vine or the amount of fruit on the vine. These numbers are different and require different approaches. In discussions with Mr. Brook Johnson and Mme Bineta Coly Guèye of the Wula Nafaa project, they are interested in an estimate of the total amount of madd fruit that is grown in an area. With this information, they can estimate the maximum amount that is available for market analysis. Unfortunately, it is very difficult to estimate the total amount of fruit growing on the vines. Below we will discuss the amount of fruit that can be potentially harvested.
The current protocol measures the madd crown width at the longest and shortest diameters within a gallery canopy and takes the average to compute crown area in two dimensions. Stem diameter measurements are also taken, but it is impossible to distinguish between individual lianas once they reach sunlight and intertwine.
The crown measurement seems a reasonable approach to quantify the surface area of the plant when a madd patch is identified. It would be interesting to consider the height of the canopy as well since the volume of the canopy determines its outer surface available to leaves of the vines, and thus to their productivity. Unanswered are the relationship between the crown cover and the measured diameters; the productive capacity of the canopy; and a survey approach to estimate the amount of madd patches in the landscape. If WN wants to find the rapport between habitat characteristics and the amount of fruit produced, we recommend further exploring the volume of the canopy that supports the vines as a potentially highly-correlated variable; perhaps even more highly correlated when used with the diameter measurements.
Assuming that riparian zones or intermittently wet areas can be identified reasonably well on the photography
Areas of intermittent water can often be seen on the aerial photos associated with large trees (see Figure 16). The madd vine itself cannot be identified on the photos but since the presence of large trees are necessary but not sufficient for the madd vine, areas with large trees can act as a means of stratifying the landscape into two strata:
Stratum one: Areas of relatively high probability of locating the madd vine.
Stratum two: Areas of low probability of finding the madd vine.
The areas of relatively high probability would include those portions of the valley bottoms where trees are present and not the total valley area. The same procedures for surveying for baobab trees could be used here. In fact, using the same photos selected for baobab would save time and money both in the photo interpretation and especially in the field data collection portion of the survey. Of course, the delineation of the baobab parks and the madd fruit areas must be kept separate by means of a different coding system.
There is no compelling reason to stratify the landscape into agricultural and non-agricultural areas. Baobab trees are found in the agricultural areas hence photos from across all lands will be selected. The amount of time to scan these photos for potential madd habitat in depressions containing trees will be very short and the survey will be more complete with these photos included within the madd sample frame.
Once at the site, measure the length , width, and height of the madd vine patch or patches. This will give you a surface estimate of the area covered by the leaves. Continue to measure diameters of individuals as before, in the hopes that eventually a relationship between diameter, crown, and fruit production can be found, even if it is with the total basal area of all the vines in the patch.
The most difficult part of this inventory is to estimate the fruit production. The operational challenges to count the fruit on the vine are large, and destructive sampling is not desirable and may even be illegal under Senegalese law. We have three recommendations:
Initially, go to villages where madd patches are nearby and solicit estimates of the annual production from the villagers. Measure the size of the patch using the above protocol, average the estimates from the villagers, and produce a yield of fruit per unit surface area. This figure will be applied to the hectares of madd patches found in the aerial photo exercise above.
Repeat at villages spatially distributed across the area of the interest, choosing a range of madd patch sizes.
Locate madd patches with villagers nearby as in number 1, but have the village leaders select a responsible person to record the yield of the patch. The leaders, with the help of the Wula Nafaa project liaison, explain to the villagers the importance of informing or better yet, allowing the responsible person to count or weight the fruit. The amount of fruit will then be put into a relationship with the surface area that produced it. The results will be the same as option 1 with a yield of fruit per unit of surface area.
For small patches, especially in the savanna areas, it may be possible to directly count the fruit if the vines are growing on only one or maybe two trees exposing the sides of the tree to direct sight. Convert the numbers into kilograms or another marketing unit, then to a yield per surface unit area.
Use the same method as described above for baobab to estimate the amount of madd vines outside of the madd vine patches.