How Many Species Are There on Earth and in the Ocean?

Camilo Mora, Derek P. Tittensor, Sina Adl, Alastair G. B. Simpson, Boris Worm

PLOS | Biology August 23, 2011

http://www.plosbiology.org/article/info%3Adoi%2F10.1371%2Fjournal.pbio.1001127#s1

The article contents:

Abstract

Author Summary

Introduction

Results

Discussion

Materials and Methods

Supporting Information

Acknowledgments

Author Contributions

References

Here I transcript only the introduction and discussion.

For the rest, go to the direction in the web, above.

Introduction

Robert May recently noted 

that if aliens visited our planet, 

one of their first questions would be, 

“How many distinct life forms

—species—does your planet have?” 

He also pointed out that we would be 

“embarrassed” by the uncertainty in our answer. 

This narrative illustrates 

the fundamental nature of knowing 

how many species there are on Earth, 

and our limited progress 

with this research topic thus far. 

Unfortunately, limited sampling 

of the world's biodiversity to date 

has prevented a direct quantification 

of the number of species on Earth, 

while indirect estimates remain uncertain 

due to the use of controversial approaches 

(see detailed review of available methods, 

estimates, and limitations in Table 1). 

Globally, our best approximation 

to the total number of species 

is based on the opinion of taxonomic experts, 

whose estimates range between 

3 and 100 million species; 

although these estimations 

likely represent the outer bounds 

of the total number of species, 

expert-opinion approaches 

have been questioned 

due to their limited empirical basis 

and subjectivity (Table 1). 

Other studies have used 

macroecological patterns 

and biodiversity ratios 

in novel ways to improve estimates 

of the total number of species (Table 1), 

but several of the underlying assumptions 

in these approaches have been the topic 

of sometimes heated controversy, (Table 1); 

moreover their overall predictions 

concern only specific groups, 

such as insects, deep sea invertebrates, 

large organisms, animals, fungi, or plants. 

With the exception 

of a few extensively studied taxa 

(e.g., birds, fishes), 

we are still remarkably uncertain 

as to how many species exist, 

highlighting a significant gap 

in our basic knowledge of life on Earth. 

Here we present a quantitative method 

to estimate the global number of species 

in all domains of life. 

We report that the number of higher taxa, 

which is much more completely known 

than the total number of species, 

is strongly correlated to taxonomic rank 

and that such a pattern allows the extrapolation 

of the global number of species 

for any kingdom of life (Figures 1 and 2).

Discussion

Knowing the total number of species 

has been a question of great interest 

motivated in part by our collective curiosity 

about the diversity of life on Earth 

and in part by the need to provide 

a reference point for current 

and future losses of biodiversity. 

Unfortunately, incomplete 

sampling of the world's biodiversity 

combined with a lack of robust extrapolation 

approaches has yielded highly 

uncertain and controversial estimates 

of how many species there are on Earth. 

In this paper, 

we describe a new approach 

whose validation against 

existing inventories 

and explicit statistical nature 

adds greater robustness 

to the estimation of the number 

of species of given taxa. 

In general, the approach 

was reasonably robust to various caveats, 

and we hope that future improvements in data quality 

will further diminish problems 

with synonyms and incompleteness of data, 

and lead to even better 

(and likely higher) estimates 

of global species richness.

Our current estimate of ~8.7 million species 

narrows the range of 3 to 100 million species 

suggested by taxonomic experts 

and it suggests that after 250 years 

of taxonomic classification 

only a small fraction of species on Earth (~14%) 

and in the ocean (~9%) have been indexed 

in a central database (Table 2). 

Closing this knowledge gap may still take a lot longer. 

Considering current rates of description 

of eukaryote species in the last 20 years 

(i.e., 6,200 species per year; ±811 SD; Figure 3F–3J), 

the average number of new species 

described per taxonomist's career (i.e., 24.8 species) 

and the estimated average cost to describe animal species 

(i.e., US$48,500 per species) and assuming 

that these values remain constant 

and are general among taxonomic groups, 

describing Earth's remaining species 

may take as long as 1,200 years 

and would require 303,000 taxonomists 

at an approximated cost of US$364 billion. 

With extinction rates 

now exceeding natural background rates 

by a factor of 100 to 1,000, our results 

also suggest that this slow advance 

in the description of species 

will lead to species becoming extinct 

before we know they even existed. 

High rates of biodiversity loss 

provide an urgent incentive 

to increase our knowledge 

of Earth's remaining species.

Previous studies have indicated 

that current catalogues of species 

are biased towards conspicuous species 

with large geographical ranges, 

body sizes, and abundances. 

This suggests that the bulk of species 

that remain to be discovered 

are likely to be small-ranged 

and perhaps concentrated in hotspots 

and less explored areas 

such as the deep sea and soil; 

although their small body-size 

and cryptic nature suggest 

that many could be found 

literally in our own “backyards” 

(after Hawksworth and Rossman). 

Though remarkable efforts 

and progress have been made, 

a further closing of this knowledge gap 

will require a renewed interest 

in exploration and taxonomy 

by both researchers and funding agencies, 

and a continuing effort to catalogue 

existing biodiversity data 

in publicly available databases.