Biodiversity and Conservation

, Volume 20, Issue 6, pp 1225–1237

Country-based patterns of total species richness, endemicity, and threatened species richness in African rodents and insectivores

Authors

    • CNR-Institute of Ecosystem Studies
  • Federica Chiozza
    • Department of Biology and BiotechnologySapienza Università di Roma
  • Carlo Rondinini
    • Department of Biology and BiotechnologySapienza Università di Roma
  • Luca Luiselli
    • Eni Spa Environmental DepartmentEnvironmental Studies Centre Demetra s.r.l
Original Paper

DOI: 10.1007/s10531-011-0024-1

Cite this article as:
Amori, G., Chiozza, F., Rondinini, C. et al. Biodivers Conserv (2011) 20: 1225. doi:10.1007/s10531-011-0024-1

Abstract

Country-based patterns of total species richness, endemicity, and threatened species richness in African rodents and insectivores are studied in this paper. We found several patterns which were similar between insectivores and rodents. Indeed, in both groups we observed: (i) a significantly uneven distribution of species richness across countries and geographic regions with highest species richness peaks being in Middle Africa and lowest peaks in Northern Africa, (ii) species richness increasing with rainfall but being independent on a country’s surface area, (iii) in each country, the insectivore total species richness and endemic species richness increases were positively correlated with rodent total species richness and endemic species richness increases. However, number of endemics peaked in South Africa and D.R. Congo in both groups, but also in Tanzania for Insectivores and in Ethiopia for rodents. In addition, the highest numbers of threatened species occurred in D.R. Congo, Rwanda and Uganda for rodents and in South Africa, Tanzania and Cameroon for insectivores. The conservation implications of these results were discussed.

Keywords

RodentsInsectivoresAfricaSpecies richnessConservationEndemicityIUCN red list

Introduction

Biologists have long recognized the striking geographical variability of species richness among geographic regions and latitudes (Currie and Paquin 1987). A primary goal of contemporary ecology is to identify the factors responsible for this variability (May 1986), but for many groups of organisms there are still no available analysis of the species richness patterns at a country-based scale, thus making difficult our progress in the understanding of the geographical variability of species richness (Myers 1998; Hawkins et al. 2003; Ruggiero and Kitzberger 2004). In addition, also conservation based policies may be negatively influenced by an unsatisfactory knowledge of the country-based species richness of threatened taxa (Soulé and Orians 2001).

In particular, small mammals (rodents and insectivores) have been overlooked in this regard compared to other mammals, despite that they account for over 50% of the total number of species (Amori and Gippoliti 2001), and despite several species are currently threatened at a global scale (Amori and Gippoliti 2000, 2003). In this paper, focusing on the African continent, we analyze by country (i) the patterns of total species richness, (ii) the patterns of endemic species richness and (iii) the patterns of threatened species richness, using data for rodents and insectivores categorized on the IUCN Red List. Overall, we collate such vital but overlooked information for a total of 397 rodent and 181 insectivore species. More specifically, we identify species-rich countries and threatened species-rich countries with the aim of determining not only the patterns of distribution richness at the continental scale, but also the main priority areas for conservation of these animals.

Materials and methods

We extracted from the IUCN Red List data for rodents and insectivores (Soricomorpha, according to Wilson and Reeder 2005) occurring in Africa. We excluded from the analysis Mus musculus, Rattus norvegicus and Rattus rattus because of their cosmopolitan distribution, and the allochtonous species (for instance Sciurus carolinensis introduced to South Africa).

In order to identify eventual regional differences in species richness by country and frequency of occurrence of threatened species, Africa was subdivided in five macro-regional zones according to the main vegetation zones as in White (1983): (i) Northern, (ii) Western, (iii) Eastern, (iv) Middle and (v) Southern (Fig. 1). Madagascar and the other islands were not considered in our analyses. West African countries were separated from central African countries despite their vegetation affinities according to White (1983) because Myers et al. (2000) identified the West African forests as a biodiversity hotspot richer than central Africa’s rainforests.
https://static-content.springer.com/image/art%3A10.1007%2Fs10531-011-0024-1/MediaObjects/10531_2011_24_Fig1_HTML.gif
Fig. 1

Map of Africa showing the main geographic regions

In order to explore whether small mammal species richness was linked to rainfall, we ranked the various countries by average rainfall per year (data from Africa Rainfall Estimate Climatology (CPC ARC), available at http://www.cpc.noaa.gov/products/fews/AFR_CLIM/afr_clim.html), according to the following scheme: score 0 = 0–500 mm (Algeria, Lybia, Egypt, Western Sahara, Mauritania, Mali, Sudan, Djibouti, Somalia, Namibia); score 1 = 501–1500 mm (Morocco, Tunisia, Eritrea, Senegal, Niger, Tchad, Ethiopia, Tanzania, Kenya, Angola, Zambia, Malawi, Mozambique, Botswana, Zimbabwe, South Africa—including Lesotho and Swaziland), score 2 = 1501–3000 mm (Gambia, Ivory Coast, Burkina Faso, Ghana, Togo, Benin, Rwanda, Burundi, Uganda), and score 3 = more than 3000 mm (Guinea-Bissau, Guinea, Sierra Leone, Nigeria, Cameroon, Congo, DR Congo, CAR, Equatorial Guinea, Gabon).

The distribution frequencies of the number of species across countries was analysed by observed-versus-expected χ2 test. One-way ANOVAs were used to identify differences among geographic regions in terms of mean number of species by country and mean number of threatened taxa by country; multivariate analysis of variance (MANOVA) models were used to investigate the effects of (i) total number of species by country, (ii) geographic region and (iii) number of endemic species by country, on the number of threatened species by country. Relationships between species richness by country and degree of rainfall were assessed by Kendall’s Tau statistics. In order to explore whether species richness by country was related to a country’s size, we analyzed by Kendall tau coefficient the number of species by country against the relative area surface of each country in km2. In all cases, analyses were performed independently for insectivores and rodents. All tests were two-tailed, with alpha set at 5%.

Results

The species richness by country is given in Fig. 2a for insectivores and b for rodents. The distribution of the number of species across countries was significantly uneven (at observed versus expected χ2) in either insectivores (χ2 = 3639.6, df = 46, P < 0.000001) or rodents (χ2 = 639.9, df = 46, P < 0.000001). More in detail, there were statistically significant differences in the mean number of insectivore species by country across geographic regions (one-way ANOVA: F4,43 = 4.357, P < 0.0048), and Tukey HSD post-hoc tests revealed that these statistical differences were mainly due to a higher species richness in Middle versus Northern (P < 0.008) and Middle versus Western (P < 0.022) (Fig. 3a). As for rodents, the same pattern was marginally significant (one-way ANOVA: F4,43 = 2.578, P < 0.051) (Fig. 3b).
https://static-content.springer.com/image/art%3A10.1007%2Fs10531-011-0024-1/MediaObjects/10531_2011_24_Fig2_HTML.gif
Fig. 2

Number of a insectivore species by country, and b rodent species by country

https://static-content.springer.com/image/art%3A10.1007%2Fs10531-011-0024-1/MediaObjects/10531_2011_24_Fig3_HTML.gif
Fig. 3

Regional variation in the mean number of species by country. a Insectivores, b rodents. For statistical details, see the text

There was a highly significant positive relationships between the numbers of rodent species by country and the number of insectivore species by country (r = 0.893, adjusted r2 = 0.793, P < 0.00001; regression equation: Insectivores = −3.178 + 0.408 × Rodents; see Fig. 4).
https://static-content.springer.com/image/art%3A10.1007%2Fs10531-011-0024-1/MediaObjects/10531_2011_24_Fig4_HTML.gif
Fig. 4

Relationships between number of species of rodents and number of species of insectivores by country. For statistical details, see the text

As for insectivores, 67 species (36.6% of the total n = 183) were endemic. Their numbers by country are given in Fig. 5a. According to this figure, the main countries in terms of number of endemic taxa were South Africa, Tanzania, D.R. Congo, all with at least nine endemic species. The mean number of endemic species by country did not vary significantly among geographic regions (one-way ANOVA: F4,41 = 1.776, P = 0.0152), but depended on the total number of species by country (r = 0.667, adjusted r2 = 0.432, n = 43, P < 0.02). Concerning rodents, 99 species (25%) were endemic. Their numbers by country (Fig. 5b) revealed that South Africa (17 species), Ethiopia (14 species) and D.R. Congo (9 species) lead the list of countries. Overall, the frequency of endemic species was significantly higher in insectivores than in rodents (χ2 = 24.43, df = 1, P < 0.000001), but their numbers were significantly positively correlated (r = 0.738, adjusted r2 = 0.534, P < 0.00001; regression equation: Insectivores = 0.019 + 0.664 × Rodents; see Fig. 6).
https://static-content.springer.com/image/art%3A10.1007%2Fs10531-011-0024-1/MediaObjects/10531_2011_24_Fig5_HTML.gif
Fig. 5

Number of endemic species by country: a insectivores b rodents

https://static-content.springer.com/image/art%3A10.1007%2Fs10531-011-0024-1/MediaObjects/10531_2011_24_Fig6_HTML.gif
Fig. 6

Relationships between number of endemic species of rodents and number of endemic species of insectivores by country. For statistical details, see the text

In terms of species of conservation concern (classified as at least Vulnerable in IUCN Red List), there were a total of 52 insectivore species, with more than 50% occurring in just three countries, i.e. South Africa (11 species), Tanzania (10 species), and Cameroon (8 species) (Table 1). The frequency of distribution of threatened taxa by geographic region was significantly uneven (observed versus expected χ2 = 44.59, df = 4, P < 0.000001), with Eastern Africa and Middle Africa housing the highest frequency of threatened taxa (Fig. 7a). A general MANOVA model (with number of threatened species by country as dependent variable, and (i) total number of species by country, (ii) geographic region, and (iii) number of endemic species by country as covariates), revealed that all of these three independent variables influenced significantly the number of threatened species by country (F24,6 = 168.313, P < 0.000001), with Duncan post-hoc tests revealing significant effects of each of the three covariates taken alone (in all cases, P < 0.000001).
Table 1

List of African insectivores categorized as critically endangered, endangered or vulnerable according to the IUCN Red List

Genus

Species

IUCN category

Countries of occurrence

Amblysomus

marleyi

EN

South Africa

Amblysomus

robustus

VU

South Africa

Carpitalpa

arendsi

VU

Mozambique, Zimbabwe

Chlorotalpa

duthieae

VU

South Africa

Chrysospalax

trevelyani

EN

South Africa

Chrysospalax

villosus

VU

South Africa

Congosorex

phillipsorum

CR

Tanzania

Crocidura

wimmeri

CR

Ivory Coast

Crocidura

harenna

CR

Ethiopia

Crocidura

ansellorum

EN

Zambia

Crocidura

baileyi

EN

Ethiopia

Crocidura

lanosa

EN

DR Congo, Rwanda

Crocidura

phaeura

EN

Ethiopia

Crocidura

stenocephala

EN

DR Congo, Uganda

Crocidura

tansaniana

EN

Tanzania

Crocidura

telfordi

EN

Tanzania

Crocidura

usambarae

EN

Tanzania

Crocidura

bottegoides

EN

Ethiopia

Crocidura

desperata

EN

Tanzania

Crocidura

picea

EN

Cameroon

Crocidura

tarella

EN

DR Congo, Uganda

Crocidura

eisentrauti

VU

Cameroon

Crocidura

glassi

VU

Ethiopia

Crocidura

kivuana

VU

DR Congo

Crocidura

lucina

VU

Ethiopia

Crocidura

manengubae

VU

Cameroon

Crocidura

allex

VU

Kenya, Tanzania

Crocidura

fumosa

VU

Kenya

Crocidura

macmillani

VU

Ethiopia

Cryptochloris

wintoni

CR

South Africa

Cryptochloris

zyli

EN

South Africa

Micropotamogale

lamottei

EN

Ivory Coast, Guinea, Liberia

Myosorex

eisentrauti

CR

Equatorial Guinea

Myosorex

geata

EN

Tanzania

Myosorex

blarina

EN

DR Congo, Uganda

Myosorex

okuensis

EN

Cameroon

Myosorex

rumpii

EN

Cameroon

Myosorex

kihaulei

EN

Tanzania

Myosorex

longicaudatus

VU

South Africa

Myosorex

zinki

VU

Tanzania

Neamblysomus

julianae

CR

South Africa

Neamblysomus

gunningi

EN

South Africa

Neamblysomus

julianae

VU

South Africa

Ruwenzorisorex

suncoides

VU

Burundi, DR Congo, Rwanda, Uganda

Suncus

aequatorius

CR

Kenya

Surdisorex

norae

VU

Kenya

Surdisorex

polulus

VU

Kenya

Sylvisorex

howelli

EN

Tanzania

Sylvisorex

isabellae

EN

Cameroon, Equatorial Guinea

Sylvisorex

morio

EN

Cameroon, Kenya

Sylvisorex

lunaris

VU

Burundi, DR Congo, Rwanda, Uganda

Sylvisorex

camerunensis

VU

Cameroon, Nigeria

https://static-content.springer.com/image/art%3A10.1007%2Fs10531-011-0024-1/MediaObjects/10531_2011_24_Fig7_HTML.gif
Fig. 7

Number of species categorized as critically endangered, endangered or vulnerable according to the IUCN Red List by geographic region. a insectivores, b rodents

As for rodents, a total of 37 species were classified as threatened (Critically Endangered, Endangered or Vulnerable) by IUCN (Table 2), with the majority of them being found in Eastern and Middle African regions (D.R. Congo, Rwanda, Uganda; Fig. 7b). The same general MANOVA design performed above, provided a statistically significant model (F10,35 = 4.463, P < 0.00044), with total number of species per country (F36,9 = 9.04, P < 0.0008) and number of endemic species per country (F10,35 = 3.46, P < 0.0005), but not geographic region (F4,41 = 2.14, P = 0.093), influencing the number of threatened species per country. Overall, there was no statistical difference between insectivores and rodents in terms of frequency of threatened species by geographic region (observed versus expected χ2 = 15.96, df = 9, P = 0.068).
Table 2

List of African rodents categorized as critically endangered, endangered or vulnerable according to the IUCN Red List

Genus

Species

IUCN category

Countries of occurrence

Allactaga

tetradactyla

VU

Egypt, Lybia

Cryptomys

kafuensis

VU

Zambia

Dasymys

montanus

EN

RD Congo, Uganda

Delanymys

brooksi

VU

Burundi, RD Congo, Rwanda, Uganda

Dendromus

kahuziensis

CR

Congo, The Democratic Republic of the

Dendromus

oreas

VU

Cameroon

Desmomys

yaldeni

EN

Ethiopia

Gerbillus

hesperinus

EN

Morocco

Gerbillus

hoogstraali

VU

Morocco

Grammomys

minnae

VU

Ethiopia

Hybomys

badius

EN

Cameroon

Hybomys

basilii

EN

Equatorial Guinea

Hybomys

lunaris

VU

RD Congo, Uganda

Hylomyscus

grandis

CR

Cameroon

Hylomyscus

baeri

EN

Côte d’Ivoire, Ghana, Sierra Leone

Lamottemys

okuensis

EN

Cameroon

Lemniscomys

mittendorfi

VU

Cameroon

Lophuromys

dieterleni

EN

Cameroon

Lophuromys

eisentrauti

EN

Cameroon

Lophuromys

rahmi

EN

RD Congo, Rwanda, Uganda

Lophuromys

medicaudatus

VU

RD Congo, Rwanda, Uganda

Lophuromys

melanonyx

VU

Ethiopia

Mastomys

awashensis

VU

Ethiopia

Meriones

sacramenti

VU

Egypt

Mystromys

albicaudatus

EN

Lesotho

Nilopegamys

plumbeus

CR

Ethiopia

Otomys

barbouri

EN

Kenya, Uganda

Otomys

lacustris

VU

Kenya, Malawi, Tanzania, Zambia

Otomys

occidentalis

VU

Cameroon, Nigeria

Paraxerus

vincenti

EN

Mozambique

Praomys

hartwigi

EN

Cameroon

Praomys

morio

EN

Cameroon, Eq. Guinea

Praomys

obscurus

EN

Nigeria

Praomys

degraaffi

VU

Burundi, Rwanda, Uganda

Tachyoryctes

macrocephalus

EN

Ethiopia

Thamnomys

venustus

VU

RD Congo, Rwanda, Uganda

Thamnomys

kempi

VU

Burundi, RD Congo, Rwanda, Uganda

Both insectivores and rodents showed a significantly positive relationship between number of species by country and rank of average country rainfall (Table 3), whereas both insectivores and rodents did not show any relationship between species richness by country and country area surface (Table 3).
Table 3

Relationships between rainfall degree, country area surface (km2), and species richness by country in African small mammals. Significance is highlighted in boldface

 

Regression equation

Kendall Tau

P-value

Rainfall

 Insectivores

Number of species = 10.105 + 4.19 × Rainfall degree

0.311

0.0018

 Rodents

Number of species = 37.535 + 6.83 × Rainfall degree

0.246

0.014

Country area

 Insectivores

Number of species = 13.59 + 0.0004 × Country area

0.142

0.155

 Rodents

Number of species = 39.3 + 0.0001 × Country area

0.190

0.056

Discussion

Our analyses, performed independently for insectivores and rodents, nonetheless revealed important similarities in the patterns emerging from both groups. Notably, both rodents and insectivores showed: (i) an uneven distribution richness across countries and geographic regions, (ii) highest species richness peaks were observed in Middle Africa and lowest peaks in Northern Africa, (iii) species richness increased with rainfall but was independent on the country size, (iv) in each country, the insectivore species richness increases were positively correlated with rodent species increases, (v) the number of endemic insectivores was positively correlated with the number of endemic rodents by country. Apart from these major similarities, we also highlighted some (relatively minor) differences: (a) number of endemics peaked in South Africa and D.R. Congo in both groups, but also in Tanzania for Insectivores and in Ethiopia for rodents, (b) the highest numbers of threatened species occurred in D.R. Congo, Rwanda and Uganda for rodents, and South Africa, Tanzania and Cameroon for insectivores.

Overall, the wide similarities in the patterns observed for both rodents and insectivores suggest that species richness is similarly influenced by macro-environmental and climatic factors, which means that both groups respond in a similar way to the ecological conditions available all throughout Africa. In particular, rainfall was for sure a main factor positively influencing species richness patterns. Areas with high humidity (especially rainforests) were also showed to be important correlates of species richness in other groups of organisms in Africa, including e.g., amphibians (Behangana and Luiselli 2009), turtles (Iverson 1992a), and birds (Pomeroy and Lewis 1987). Obviously, also the fact that Middle Africa was reacher than Northern Africa in terms of species richness is well demonstrated in many other organisms (Iverson 1992b, 1992c; Vié et al. 2009), hence not meriting further discussion.

In terms of endemic species richness, insectivores and rodents showed both similarities and differences. Once again, the positive relationships between numbers of endemic species by country revealed an overall similarity between the two groups, also confirmed by the fact that South Africa and D.R. Congo were important countries for endemicity in both groups. These two regions are well known to be important endemicity centres for other groups as well (Picker and Samways 1996; Rosenzweig and Sandlin 1997; Cowling et al. 2003; Holden and Levine 2009). Interestingly, however, Ethiopia and Tanzania were, respectively, important for rodents and insectivores. As for Ethiopian rodents, the important endemicity region is the Bale mountains (Sillero-Zubiri et al. 1995a), which is also important for other animal groups (Hillman 1986; Gottelli and Sillero-Zubiri 1992). For rodents, there a few endemic genera of the area which are also seriously endangered (i.e., Megadendromus, Muriculus, Nilopegamys), in addition to the various endemic species which are not threatened (e.g., the giant molerat Tachyoryctes macrocephalus; Sillero-Zubiri et al. 1995b). The endemic and threatened genera of rodents of Bale Mountains allowed to identify a special priority ecoregion for rodent conservation: the Ethiopian Highlands ecoregion (Amori and Gippoliti 2001). As for insectivores, Tanzania appeared an important area for endemicity, with one Congosorex, four Crocidura, three Myosorex, and one Sylvisorex being endemic of this country. The presence of these species (particularly Congosorex) in the Eastern Arc Mountains supports the theory that these mountains are refuges of ancient lineages of organisms with ties to the Guineo-Congolian region of Africa and that there was once a continuous swath of forest connecting these two areas (Stanley et al. 2005).

In terms of conservation implications, our study suggests that different regions should be considered of priority relevance for insectivores and rodents. For insectivores, South Africa, Tanzania and Cameroon are particularly important. More in detail, the moist deep forests of Cameroon are important for the conservation of threatened Sylvisorex and Myosorex species (Nowak 1999), as well as the Eastern Arc mountains in Tanzania for Congosorex, Crocidura and Myosorex species (Novak 1999), and the north-northeastern regions (e.g. for Amblysomus, Cryptochloris and Neamblysomus species) and the Cape area (Chlorotalpa) in South Africa.

For rodents, D.R. Congo, Rwanda and Uganda should be considered the main areas for conservation. Primary rainforests should be the main target habitats. This is true in both mountain regions in the east of the country (several species of potential conservation concern, including e.g. Delanymys brooksi and Dendromus kahuziensis, Stuart and Adams 1990) as well as along the coastal forests where species of the genera Myosciurus, Lamottemys, and Zenkerella can be threatened, especially by deforestation (Amori and Gippoliti 2001).

Copyright information

© Springer Science+Business Media B.V. 2011