(C) PLOS One [1]. This unaltered content originally appeared in journals.plosone.org.
Licensed under Creative Commons Attribution (CC BY) license.
url:https://journals.plos.org/plosone/s/licenses-and-copyright

------------



The witchweed Striga gesnerioides and the cultivated cowpea: A geographical and historical analysis of their West African distribution points to the prevalence of agro-ecological factors and the paras

['Abou-Soufianou Sadda', 'Umr Dap', 'Université Dan Dicko Dankoulodo De Maradi', 'Maradi', 'Ird', 'Umr Iees-Paris', 'Su Ird Cnrs Inra Upec Univ. Paris Diderot', 'Centre Ird De France Nord', 'Bondy Cedex', 'Geo Coppens D Eeckenbrugge']

Date: 2021-09

To validate this model, we have confronted it to the expert information presented in the report of the Organization for Economic Co-operation and Development (OECD) [ 3 ], observing only partial correspondence. In fact, cowpea cultivation areas correspond very well to our Maxent model in the semi-arid Sahelian to Sudano-Sahelian zone and the northern Guinean savannah zone. However, the two maps differ significantly for the southern and western Guinean savannah and forest zones of West Africa. To the west, the OECD map also shows no significant cowpea cultivation not only from Guinea Bissau, through Sierra Leone, to Liberia, but also in the highly favorable region of northeastern Guinea, where our occurrence data document frequent cowpea cultivation in the latter region of Guinea. Along the shores of the Gulf of Guinea, the OECD map only shows little cowpea cultivation in the south of Côte d’Ivoire, Ghana, Togo, and Benin. For Nigeria, it appears limited to the north of the country, and the OECD distribution does not include most of the light green areas reported for this country in Fig 1B . In contrast, our occurrence data indicate limited cultivation in Côte d’Ivoire, but frequent cultivation in Ghana, Togo, Benin and Nigeria, even in the south, all regions where, according to the Maxent model, cowpea benefits from good to excellent climatic conditions.

Fig 1B presents the modeled distribution obtained with the Maxent software for cultivated cowpea in West Africa. Very good to excellent climatic conditions dominate in the northeast zone (Niger and northern Nigeria) and in the southern central zone, in and around the Dahomey gap (southern Nigeria, Benin, Togo, Ghana, southern Burkina Faso, and most of Côte d’Ivoire). In Côte d’Ivoire, these highly favorable conditions do not appear to match extensive cowpea cultivation. To the west, we find a highly favorable area in Guinea, and another one in Senegal, where the slight contrast between climatic adaptation and relative datapoint dispersal can be explained by observation methodology and/or a recent extension of the crop. In fact, cowpea has much increased in importance as a food crop in the northern peanut basin of Senegal after the repeated droughts of the late 20 th century, partly substituting more traditional staples, due to its ability to produce food in infertile soils under these conditions [ 24 ].

The distribution of the 1747 retained datapoints ( Fig 1A ) represents the geographic range of cultivated cowpea at the West African scale. The datapoint density appears heterogeneous across countries of interest. It is very high from northern Cameroon to Ghana and Burkina Faso, passing through Nigeria, southern and western Niger, Benin, and Togo. More to the west, datapoints appear less dense, yet relatively well distributed between southern Mali, northwestern Côte d’Ivoire, Guinea, and western Senegal, whereas observations are very sparse in Gambia, Sierra Leone, and Liberia. In the eastern half of Senegal and the western half of southern Mali, the uniform and relatively low density is linked to the source of data, mostly the CoEx project, which followed a particular sampling strategy. Otherwise, this relative east-west contrast probably reflects the economic importance of the crop among West African countries, with the relative exception of Nigeria, the main producer country, where the datapoint density is lower than for the second and third producers, Niger and Burkina Faso. More to the east, datapoint density in Mali may not represent its substantial contribution to world production [ 1 ].

From the 425 records of S. gesnerioides, 275 correspond to infestations on cultivated cowpea. As shown in Fig 1B , most of them were from regions with highly suitable climates for cowpea cultivation. This is particularly true in the center and to the east of the study area (i.e., from northern Cameroon to Burkina Faso and Togo). In the south, along the Gulf of Guinea, records of S. gesnerioides infestations were limited to the Dahomey gap, with presence in Benin and Togo. We have found no infestation records from southern Ghana to Guinea Bissau. In the case of Côte d’Ivoire, Kouakou et al. [ 44 ] ascertain the absence of S. gesnerioides in cowpea fields. To the northwest, occurrences were only recorded in southeastern Mali and western Senegal. In the latter country, most data concern the groundnut basin, and the Striga infestation appears less severe, as compared to the center and east of West Africa [ 24 ].

The vast majority of Striga infestations are observed in this densest and driest part of the cowpea climate envelope, particularly under highly contrasted climates. Clearly, S. gesnerioides is more xerophytic than cowpea and its aggressiveness is favored by aridity. The identification of datapoints by country ( Fig 2B ) helps interpreting the geography of S. gesnerioides impact. Niger and northern Senegal are the places where cowpea is confronted to the most arid and contrasted climates, closely followed by Mali, Burkina Faso, northern Nigeria and northern Cameroon. Along the Dahomey gap, climates of Ghana, Togo and Benin present slightly less arid and less contrasted climates, and these countries are less severely exposed to Striga attacks. In contrast, cowpea cultivation areas from Guinea, Liberia, southern Cameroon and southern Senegal, in the lower half of the principal plane, i.e., under much less arid climates, do not appear to be threatened by Striga parasitism. The case of Côte d’Ivoire is particular, as its northeastern region presents climates similar to those of Benin, Togo and Ghana, but no cases of Striga parasitism on cowpea.

Cultivated cowpea occurrences clearly accumulate at high values of the second principal component, i.e., under semi-arid climates, characterized by a severe dry season of 7–10 months and only 1–2 months of rainfall in the north to 4–5 months in the south, with a high spatial, inter- and intra-annual variability ( Fig 2A ). This accumulation of datapoints in the upper part of the principal plane seems to draw a saddle-shaped aridity limit to the cowpea climatic envelope; only ca. 60 occurrences, mostly from Senegal ( Fig 2B ), lie above this limit.

Table 1 and Fig 2 present the results of the PCA characterizing the ecoclimatic envelopes of cultivated cowpea. Bioclimatic variables 1 (annual mean temperature), 2 (mean diurnal range), 6 (minimal temperature of coldest month), 9 (mean temperature of driest quarter), 11 (mean temperature of coldest quarter), 14 (precipitation of driest month), 17 (precipitation of driest quarter) and 19 (precipitation of coldest quarter) were excluded from the analysis because of their poor specific contribution to the ecoclimatic model and/or because they can be deduced directly from other bioclimatic variables. The two principal components with an eigenvalue superior to 1 are related to (i) seasonality of temperatures and precipitation (explaining 51% of the total variance), and (ii) precipitation and temperature of the wettest month and quarter (33% of the total variance).

Striga gesnerioides in West Africa: Hosts, geography and evolution of collections

In addition to the 275 records of cultivated cowpea parasitism, 91 cases were reported on other species, and 59 cases without host information were recovered, for a total of 425 occurrences of S. gesnerioides (Table 2). Thus, in the study area, S. gesnerioides parasitizes many dicotyledons, mostly wild members of Fabaceae (Alysicarpus ovalifolius, Cassia mimosoides, Indigofera astragalina and Indigofera sp., Tephrosia elegans and T. pedicellata, and Zornia glochidiata) and Convolvulaceae (Ipomoea coptica, I. coscinosperma, I. eriocarpa, I. nil, I. pescaprae, I. pes-tigridis, I. pileata, I. vagans, Jacquemontia tamnifolia, Merremia pinnata and M. tridentata). It has even been reported on an introduced legume, Arachis repens, in Nigeria and Togo [45]. More rarely, it has been found on species of the Primulaceae (Anagalis pumila). The single observation on cultivated tobacco (Nicotiana tabacum, Solanaceae) is not surprising, as S. gesnerioides is a major pest of this crop in Zimbabwe [21]. More surprising are the four cases on monocotyledons from genera Andropogon and Pennisetum (P. pedicellatum) in Mali and Benin. Once compiled from the available databases and literature, this host range appears wider than previously published lists [19–24, 44, 46, 47].

Mapping the occurrences of S. gesnerioides (Fig 3A) first shows its quite general presence in the Sahelian and Sudano-Sahelian belts and its absence to the south, except in Benin and relatively close areas of Togo and Nigeria, in relation to the dry climate of the Dahomey gap. Second, no clear geographic differentiation can be detected between the distribution of S. gesnerioides hosted by cowpea and that of records on other hosts (wild plants in their great majority). However, again we find distinct trends in eastern and western parts of West Africa. To the east, in Niger, Nigeria, and Burkina Faso, observations on cowpea appear more widespread than on wild plants; only in Benin do the reports on wild plants cover a significant area of the country, comparing well with that observed on cowpea. To the west, the cases on cowpea and on other plants appear much better balanced, both in their geographic distribution and in their numbers. An extreme case is that of northern Côte d’Ivoire, where all observations concern wild plants. One likely reason for the higher prevalence of cowpea specific strains in eastern West Africa is that cowpea itself is more present there, which may cause a collecting bias in our host data. A second one is that S. gesnerioides has become much more common there because of this earlier extension and intensification of cowpea cultivation to the east. The latter explanation is consistent with the historical effect of monocropping and loss of soil fertility on the severity of the parasitism [16, 17, 20, 25, 30, 31]. Thence it is important to understand both the spatial and temporal dynamics of the S. gesnerioides problem aggravation. To address this question, we have categorized the observations by bi-decadal periods (Fig 3B).

PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 3. Distribution of S. gesnerioides records and derived distribution model. A. Occurrence records categorized by host species: cowpea (pink circles), wild Fabaceae (red squares), wild Convolvulaceae (blue squares), Poaceae (yellow squares, for Andropogon, and circle, for Pennisetum), Anagalis pumila and Cyanotis lanata (orange squares), indeterminate wild hosts (grey squares) and indeterminate hosts in indeterminate environment (white squares). B. Occurrence records categorized by decadal or bidecadal period of observation: 1899–1909 (black circles), 1960–1980 (green circles), 1980–2000 (yellow circles), 2000–2020 (red crosses), unknown date (grey crosses). C. Maxent ecoclimatic distribution model derived from these observations (color code as for Fig 1B). https://doi.org/10.1371/journal.pone.0254803.g003

Up to 1980, it seems that S. gesnerioides was quite neglected by botanists and agronomists. Indeed, the five earliest collections of our dataset, dating from 1899 to 1909, are all from Mali, and their host was not identified. The next 15 were collected in Cameroon, Niger, Nigeria, Togo, Côte d’Ivoire, Mali and Senegal, from 1960 to1980. The host was identified only in six cases, four on cowpea and two on wild plants (determined to family or genus). For the 1980–2000 bidecadal period, the number of records increases very significantly, with a total of 186, in the same seven countries mentioned above, plus Burkina Faso and Benin. The host was often recorded too, cultivated cowpea being most common: four out of five in northern Cameroon, 14 out of 18 in Niger, 22 out of 29 in Nigeria, 28 out of 32 in Burkina Faso, and 17 out of 26 in Mali. The picture was slightly different for Benin, where the numbers of cowpea (19) and wild hosts (12) were more balanced, thanks to a phytosociological inventory by a group from the Abomey-Calavi University. It was even more different for Togo, with only six cases, two of which on cowpea, and for Senegal, with 15 records on wild hosts versus five on cowpea. More globally, the sudden increase in record numbers was related to a series of systematic field surveys, themselves justified by the increasingly severe impact of S. gesnerioides parasitism on cowpea cultivation, such as the FAO study in northern Cameroon, Nigeria, Niger, Benin and Togo, [45, 48], that of Wade [24] in Senegal, and the extensive collections of herbarium materials by A. Raynal-Roques in Burkina Faso and Mali, from 1985 to 1991. Efforts to develop genetic resistance also provided information on the distribution of S. gesnerioides races [27].

From 2000 to present, collecting efforts have further increased, in Niger (Maradi and Zinder regions; [49, 50], Benin, northern Ghana (nine sites; [51], Burkina Faso (58 sites; [47], and in the groundnut basin of Senegal [52]. Our dataset includes 207 records (after validation and filtering), with 156 cases on cultivated cowpea, and 32 on other plants, mostly contributed from a survey in Côte d’Ivoire, where S. gesnerioides was observed only on wild hosts [44].

While the chronological dynamics of data accumulation clearly indicates that S. gesnerioides had become a very significant pest on cowpea around 1980, more precise information on the spatio-temporal sequence of this development is quite scarce and difficult to scan, because of the successive delays between the occurrence of the problem, the related research programs, and the publication of their results.

At the latest, the problem appeared in the late 1970s in the cowpea cultivation basin of Niger and northern Nigeria, as well as in southern Benin. Indeed, following a five-year survey in Niger, from 1981 to 1985, Adam [12] identified S. gesnerioides as one of the three major causes of the decreasing cowpea yield, in relation to its abundance in the central-eastern (Maradi and Zinder) and western (Niamey) regions of the country. In the proceedings of the Second International Workshop on Striga, held in 1981, Obilana [53] commented on the serious threat posed by S. gesnerioides in northern Nigeria, with very severe crop losses in 1980, and mentioned field screening for resistance or tolerance to this pest, in 1977, 1978, and 1979. According to the report of Parkinson [45] on his 1984–1985 survey, S. gesnerioides attacks on cowpea were limited to the Sudano-Sahelian region of Nigeria, although it was observed much further south on other hosts. He further mentioned cowpea as the host of S. gesneroides in southern Benin, farmers claiming “that most of the cowpea consumed in Benin used to be produced in Zou Province but in the last five years, this is no longer true because of Striga infestation.” Thus, S. gesnerioides infestation was very severe before 1980 in this region too.

Countries in the center and west of the studied region suffered attacks later: in the above-mentioned Second International Workshop on Striga of 1981, Reneaud [54] commented that in the Volta valleys (Burkina Faso), S. gesnerioides was still considered of minor importance, in a context of sporadic cultivation of cowpea on limited surfaces. For northern Ghana, Stoop et al. [55] did not even mention cowpea among the Striga-threatened crops, and Parkinson [45] found no incidence of infestation of cowpea in Togo in 1984–1985, despite the observation of S. gesnerioides parasiting Arachis repens.

This eastward delay in S. gesnerioides infestation was confirmed in 1990 by Cardwell and Lane [30], whose systematic survey revealed more general and severe attacks on cowpea in Niger, northern Nigeria and southern Benin, than in Burkina Faso and Mali. Of notice too is the observation of Hoffmann et al. [56] that cowpea infestations were still localized and limited to small areas in Mali in 1991. Even further west, in Senegal, the impact of S. gesnerioides was marginal in the 1970s [57]. It became increasingly significant after 1985, following the collapse of peanut production and the extension of cowpea cultivation [24, 52]. For the period 1993–1996, Wade [24] documented the presence of S. gesnerioides on wild plants and, increasingly, on cultivated cowpea in most of the peanut production basin (14% of cowpea fields infested), and more particularly in its central and northern portions, developing important populations in departments where cowpea was more prominent (50% of infested fields in certain villages). Furthermore, Wade [24] demonstrated that the parasite was able to shift successfully between wild hosts and cowpea, in both directions, and predicted that it would soon become a considerable threat for the crop. The studies of Wade [24] and Tonessia et al. [52] constitute the clearest field demonstration of the relationship between the expansion of cowpea cultivation (from 12,000 ha in 1970 to 90,000 ha in 2001) and its parasitism by S. gesnerioides, further aggravated by the introduction of susceptible cowpea cultivars and the development of virulence against the most common cultivars.

This spatiotemporal analysis of the literature leads us to use a similar narrative to explain the spread of severe S. gesnerioides parasitism on cowpea at the scale of the whole region. Before the rise of infestations, both the crop and the parasite had the same distribution as today in West Africa, and the general intensification of cowpea production has led to the current situation, first in two early cowpea basins of (i) northern Cameroon, Niger, Northern Nigeria and (ii) southern Benin, and later westward to Togo, Ghana, Burkina Faso, Mali and Senegal. Contrary to the vision of a wide-scale diffusion of new and aggressive forms of the parasite, following the spread of their host, the multilocal vision derived from Wade’s interpretation implies that the distribution of S. gesnerioides was already in equilibrium well before the expansion and intensification of cowpea cultivation, which can be verified by ecoclimatic modeling and a comparative study of the bioclimatic envelopes on wild hosts and cowpea.

[END]

[1] Url: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0254803

(C) Plos One. "Accelerating the publication of peer-reviewed science."
Licensed under Creative Commons Attribution (CC BY 4.0)
URL: https://creativecommons.org/licenses/by/4.0/


via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/

You are viewing proxied material from magical.fish. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.