SCIENCE science.org 10 DECEMBER 2021 • VOL 374 ISSUE 6573 1333
and reception ( 1 , 2 ). Yet ocean soundscapes
have aroused great concern ( 3 ), whereas
little attention has been paid to anthropo-
genic noise pollution in riverine habitats.
Unlike the marine environment, riverine
ecosystems are constricted to areas that
often face inevitable competition with
humans. All cetaceans that live in rivers,
including five dolphin species and a por-
poise, have declined substantially both in
range and numbers, especially in Asia ( 4 ).
The Yangtze finless porpoise (Neophocaena
asiaeorientalis asiaeorientalis), the sole
freshwater porpoise in the world, which
is only found in two Chinese lakes and
some parts of the Yangtze River, is listed as
Critically Endangered by the International
Union for Conservation of Nature ( 5 ).
Noise pollution affects freshwater ceta-
ceans across regions. The Yangtze has
been the world’s busiest inland river in
terms of shipping since 2005 ( 6 ), and its
daily average vessel flow and annual cargo
movement have increased over the past
decade ( 7 ). Noise levels at approximately
88% of the river’s porpoise distribution
area are sufficient to disrupt the behav-
ior of the finless porpoise, and 40% of
the river has noise levels high enough to
cause reduction in the porpoise’s hearing
sensitivity ( 8 , 9 ). Underwater noise expo-
sure from vessels in the Ganges River in
India elevated metabolic stress in Ganges
river dolphins (Platanista gangetica
gangetica) ( 10 ). Busy vessel traffic in the
Indus River in Northern Pakistan threat-
ens the Indus River dolphin (P. g. minor)
( 11 ). Compared with coastal Irrawaddy
dolphins (Orcaella brevirostris), fresh-
water Irrawaddy dolphins in Indonesia’s
Mahakam River showed a hypersensitivity
to intensive boat traffic ( 12 ).
Addressing noise pollution will
require improvements to shipping
lanes, ports, and ships. Shipping lanes
should be regulated to protect channel
resources. Optimal navigation speeds
and restrictions of ship speeds in ecologi-
cally important areas would help mitigate
noise pollution in cetacean habitats. The
government should identify and elimi-
nate illegal ports, especially those that
have been built along the Yangtze in
China. Ships that are docked should be
required to use power from land rather
than continuing to run their engines. The
construction and renovation of environ-
mentally friendly and economical ships
should be developed and put to use.
Subsidies should be provided to compa-
nies using outdated vessels to help them
replace the ships with newer, quieter tech-
nology, such as more efficient propellers.
Finally, governments should require all river
traffic to meet operation and management
standards that reduce noise pollution.
Zhi-Tao Wang, Peng-Xiang Duan, Ke-Xiong Wang*,
Ding Wang*
The Key Laboratory of Aquatic Biodiversity and
Conservation of the Chinese Academy of Sciences,
Institute of Hydrobiology, Chinese Academy of
Sciences, Wuhan 430072, China.
*Corresponding author. Email: [email protected]
(K.-X.W.); [email protected] (D.W.)REFERENCES AND NOTES- S. L. King, V. M. Janik, Proc. Natl. Acad. Sci. U.S.A. 110 ,
13216 (2013). - S. D. Simpson et al., Nat. Commun. 7 , 10544 (2016).
- C. Duarte et al., Science 371 , eaba4658 (2021).
- Marine Mammal Commission, “Freshwater
dolphins and porpoises” (2021); http://www.mmc.
gov/priority-topics/species-of-concern/
freshwater-dolphins-and-porpoises/. - D. Wang, S. T. Turvey, X. Zhao, Z. Mei, “Neophocaena
asiaeorientalis ssp. Asiaeorientalis” (The IUCN Red List
of Threatened Species, 2013); http://www.iucnredlist.org/
species/43205774/45893487. - X. Y. Zhu, “The annual cargo throughput of the Yangtze
River trunk line has ranked first in the world for many
y e a r s ,” China News (2019); http://www.chinanews.com/
cj/2019/07-05/8885178.shtml [in Chinese]. - Editorial Committee of the Yangtze River Yearbook,
Yangtze River Yearbook in 2020 (Yangtze River Yearbook
Press, Wuhan, 2020) [in Chinese]. - Z. T. Wang et al., Environ. Pollut. 262 , 114310 (2020).
- Z. T. Wang et al., Ecotoxicol. Environ. Safe. 226 , 112860
(2021). - M. Dey et al., Sci. Rep. 9 , 15426 (2019).
- F. Perveen, S. A. Mehmood, S. Ahmed, Z. U. Reman, J. Life
Sci. 5 , 39 (2011). - D. Kreb, K. D. Rahadi, Aquat. Mamm. 30 , 363 (2004).
10.1126/science.abf0222
STEM disparities we
must measure
The National Science Foundation’s (NSF’s)
Survey of Earned Doctorates, an annual
census of all newly minted US PhDs since
1957, has been a cornerstone in science,
technology, engineering, and mathemat-
ics (STEM) diversity efforts ( 1 ). Along
with NSF’s other national workforce
surveys, the Survey of Earned Doctorates
provides critical data needed to document
underrepresented groups in STEM and
determine their eligibility for funding and
federal resources. However, NSF’s surveys
do not include sexual orientation and
gender identity (SOGI) questions, despite
the alarming disparities that lesbian, gay,
bisexual, transgender, and queer (LGBTQ)
people face in US STEM fields ( 2 ).
LGBTQ people are estimated to be 17 to
21% less represented in STEM than statis-
tically expected ( 3 , 4 ) and are less likely
to major in STEM, persist in STEM, earn
STEM degrees, and be in STEM occupa-
tions ( 2 , 5 , 6 ). Harmful biases and unsup-
portive STEM environments appear to be
partly at fault. For instance, LGBTQ scien-
tists experience more career barriers and
workplace harassment than non-LGBTQ
scientists, even when controlling for other
demographic and career-related factors ( 7 ).In April, after years of delays, NSF
began piloting SOGI questions for its
surveys, and the agency presented its
first results on 2680 respondents in
November ( 8 ). Although NSF had long-
standing concerns that the questions
would be perceived as too sensitive, their
results showed instead that break-off
and nonresponse rates (which indicate
discomfort with survey questions) were
virtually 0%. Moreover, LGBTQ and non-
LGBTQ respondents alike overwhelmingly
reported feeling comfortable providing
SOGI data to a federal agency like NSF.
However, NSF stopped short of any indi-
cation that it would consider adding SOGI
questions to its surveys now, suggesting
instead that more research is needed.
Any incremental knowledge NSF will
gain by additional research is outweighed
by the urgent need for nationwide SOGI
data on the STEM workforce. Agencies
from the Department of Education to
the Department of Labor have already
been collecting SOGI data on hundreds of
thousands of respondents for years ( 9 ). In
2016, the Department of Education began
collecting SOGI data on its national
surveys of high-schoolers and college
students ( 10 ). Five years later, NSF is
increasingly out of sync with its federal
counterparts.
Demographic questions are never per-
fect. NSF’s race and ethnicity questions
reflect federal-wide standards established
in 1977 and later revised in 1997 ( 11 ). The
questions have been criticized for being
insufficiently nuanced ( 12 ), but NSF has
never hesitated to include them, because
race and ethnicity data are critically
needed to resolve urgent STEM opportu-
nity gaps. The same logic applies to SOGI
data. NSF should always work to improve
how it collects demographic information,
but striving for more perfect survey meth-
ods should never trump its congressio-
nally mandated responsibility to address
the challenges faced by underrepresented
groups in STEM. LGBTQ people in STEM
should be no exception.
Jonathan B. Freeman
New York University, New York, NY 10003, USA.
Email: [email protected]REFERENCE AND NOTES- NSF, Survey of Earned Doctorates (2021); http://www.nsf.gov/
statistics/srvydoctorates/. - J. B. Freeman, Pol. Insights Behav. Brain Sci. 7 , 141 (2020).
- J. B. Freeman, Nature 559 , 27 (2018).
- E. Cech, M. Pham, Soc. Sci. 6 , 12 (2017).
- B. E. Hughes, S c i. A d v. 4 , eaao6373 (2018).
- D. Sansone, C. S. Carpenter, PLOS ONE 15 , e0241596
(2020). - E. Cech, T. Waidzunas, Sci. Adv. 7 , eabe0933 (2021).
- R. L. Morrison, J. Chandler, F. Lan, K. Jamrick, “Session
B-1: Measuring sexual orientation & gender identity:
INSIGHTS