Nutrient Enrichment Drives Gulf of Mexico Hypoxia Grid Computing For
Total Page:16
File Type:pdf, Size:1020Kb
Earth Science in a Controlled Environment, p. 120 Meetings: The Ocean’s Response to Climate Change, p. 121 Meetings: Connecting Scientific Drilling and Human Evolution, p. 122 About AGU: Shelly, Lagroix, and Carlson Honored, pp. 123, 124 About AGU: New Editors Appointed for Three Sections of JGR, p. 124 VOLUME 90 NUMBER 14 7 APRIL 2009 Nutrient Enrichment Drives Gulf of Mexico Hypoxia During most summers over the past reducing nitrogen loading [Rabalais et al., 30 years, bottom dissolved oxygen across 2007]. a large area of the Louisiana and upper Texas continental shelf declined to con- Consensus and Skepticism centrations too low (hypoxia) for most fi sh and large invertebrate animals to survive. As part of a reassessment of the Action This area is one of the best known “dead Plan completed in 2008, the U.S. Environ- zones” proliferating around the world [Diaz mental Protection Agency’s Science Advisory and Rosenberg, 2008]. During July 2008, Board (SAB) conducted another evaluation hypoxic bottom waters extended across of the state of the science regarding hypoxia. 20,720 square kilometers (Figure 1), but An SAB panel (including three article coau- they were probably even more extensive thors, Walter R. Boynton, Robert W. How- because winds from Hurricane Dolly mixed arth, and James G. Sanders) concluded, Fig. 1. High chlorophyll a concentrations along the inner shelf off Louisiana and Texas (orange the waters off Texas before the survey “Recent science has affi rmed the basic con- and red) on 11 June 2008 (Indian Space Research Organization Oceansat 1 Ocean Color Moni- could be completed. clusion that contemporary changes in the tor image provided by the Louisiana State University (LSU) Earth Scan Laboratory; http:// www Increased inputs of nutrients (princi- hypoxic area in the northern Gulf of Mexico . esl . lsu . edu/ imagery/ ocm/) in relation to an overlay of the bottom hypoxia (<2 milligrams per pally nitrogen and phosphorus) from the are primarily related to nutrient fl uxes from liter, gray area outlined by dashes) during 21–27 July 2008. In June, chlorophyll a concentrations U.S. agricultural heartland within the the MARB” [SAB, 2008, p. 2]. The recon- ranged from 22 micrograms per liter (inshore) to 7 micrograms per liter (offshore) (N. N. Raba- Mississippi- Atchafalaya River Basin ( MARB) fi rmed consensus is that anthropogenic nutri- lais, unpublished data, 2008). are implicated in the development and ents stimulate the production of planktonic spread of hypoxia in the Gulf of Mexico. organic matter, the decomposition of which Consequently, the causes of, and solutions depletes dissolved oxygen in bottom waters Organic Matter Sources suggesting that a signifi cant amount of this for, hypoxia have been subjects of exten- on the seasonally stratifi ed inner shelf. marsh- derived organic matter enters the sive debate and analysis. An integrated sci- Despite these two major scientifi c assess- Seasonally recurring hypoxia developed bottom waters of the hypoxic zone [Turner entifi c assessment led to a 2001 Action Plan ments supporting this consensus, skeptics on the shelf from the 1970s through the et al., 2007]. Most of the organic matter is [Mississippi River/Gulf of Mexico Water- [Dagg et al., 2007; Bianchi et al., 2008] have 1990s, coinciding with a tripling of nitrate likely redeposited or metabolized within the shed Nutrient Task Force, 2001] with a goal suggested alternative causes of hypoxia, loading from the MARB between the 1950s estuarine- wetland complex or is transported of reducing the area of the hypoxic zone including (1) oxidation of organic matter and 1990, but not with changes in stratifi ca- offshore in surface waters in dissolved form to less than 5000 square kilometers by not derived from phytoplankton production, tion or external sources of organic matter [SAB, 2008]. Little wetland- derived organic (2) physical processes affecting water col- [SAB, 2008]. While there has been a mas- carbon is found in sediments beyond the umn stability, and (3) coastal wetland loss sive loss of coastal wetlands since the 1930s, coastal bays [Gordon and Goñi, 2003]. The BY D. F. BOESCH, W. R. BOYNTON, L. B. CROWDER, and river controls. This article addresses the loss rate peaked in the 1960s and has nonmarine organic matter in shelf sediments R. J. DIAZ, R. W. HOWARTH, L. D. MEE, S. W. NIXON, these criticisms and demonstrates why they since declined [Barras et al., 2008]. Further- is highly refractory and resistant to further N. N. RABALAIS, R. ROSENBERG, J. G. SANDERS, do not challenge the consensus on nutrient more, there is no corroborating spatiotem- D. SCAVIA, AND R. E. TURNER enrichment. poral, biochemical, or isotopic evidence Hypoxia cont. on page 118 Earth science computing and data man- Grid Computing for Earth Science agement needs traditionally have been pro- eGY vided for by local and national institutions. Electronic Geophysical Year The fundamental challenges facing and model simulations of a host of phenom- The limitations of cost and the number of human kind at the beginning of the 21st cen- ena relating to the Earth and its space envi- computer central processing units (CPUs) tury require an effective response to the ronment. These capabilities require that the available at any one site can be overcome by needed and a logical name for data to be massive changes that are putting increasing Earth science community utilize, both in geographically distributed systems for access- accessed via the grid storage. Because of its architecture, a grid can pressure on the environment and society. real and remote time, massive amounts of ing data, computing resources, and Web ser- effi ciently tackle a large ensemble of com- The worldwide Earth science community, data, which are usually distributed among vices. Geographically distributed computing putations running independently. A grid is with its mosaic of disciplines and players many different organizations and data began by catering to specialized purposes or also ideally suited for analyzing and pro- (academia, industry, national surveys, inter- centers. particular user groups (e.g., the Distributed ducing large data sets and for sharing data national organizations, and so forth), pro- The Earth science community can bene- European Infrastructure for Supercomput- within large teams. Several grid infrastruc- vides a scientifi c basis for addressing issues fi t greatly from technology that can provide ing Applications). The Berkeley Open Infra- tures have already been deployed around such as the development of new energy ready access to computing resources and structure for Network Computing (BOINC), the world, for example, in North and South resources; a secure water supply; safe stor- services, easily managed data and metadata introduced in 1999, pioneered the use, on a America, Asia, Australia, North Africa, and age of nuclear waste; the analysis, modeling, storage in distributed systems or in data cen- volunteer basis, of the enormous processing in 2008 in Senegal. and mitigation of climate changes; and the ters, clearly defi ned data policy, authentica- power of personal computers (PCs) around The largest grid deployment to date, assessment of natural and industrial risks. In tion, confi dentiality, and electronic collabo- the world. Earth scientists adopted BOINC Enabling Grids for E- Science (EGEE; http:// addition, the Earth science community pro- ration. Grid infrastructure and systems meet only for specifi c climate and hydrology appli- w w w . e u - e g e e . o r g / ) , i s d e s i g n e d f o r a n a l y z - vides short- and medium- term prediction of these requirements as a distributed resource cations because it is not general enough and ing petabytes of data that will be produced weather and natural hazards in real time, system. Grid computing permits the sharing does not handle the problems of PC heteroge- by the European Organization for Nuclear of resources between institutions and allows neity and confi dentiality requirements. Research’s (CERN) Large Hadron Collider BY P. RENARD, V. BADOUX, M. PETITDIDIER, for scaling up computing power and stor- experiment in Geneva, Switzerland. Access AND R. COSSU age capacity in a way that is impossible for a Grid Computing to EGEE is not restricted to high- energy phys- single institution to do. Also, grid computing ics and is currently used by other scientifi c offers a transparent collaborative platform Grid computing emerged more than communities including bioinformatics, Earth for users, allowing them to have access to 10 years ago [Foster and Kesselman, 1998] sciences, and astronomy. As of March 2009, more resources at a given time. This access as one type of distributed resource sys- EGEE is deployed at more than 300 sites. is especially important for exploiting large tem. Grid computing consists of a network EGEE provides more than 80,000 CPUs and data sets scattered in several locations, for infrastructure comprising loosely coupled more than 20 petabytes of storage, and it running large statistical jobs, and for sharing heterogeneous data storage and comput- is capable of running up to 100,000 jobs data and algorithms among many partners ing resources connected via the Internet concurrently. without the need for conversions. and controlled for management and access Grid computing has become a basic by software (middleware) such as gLite, tool for particle physics and biotechnology UNICORE, Globus Toolkit, and GRIA. A grid researchers, but it still is not used routinely system is based on long- term and dynamic Why Is This Here? by Earth scientists. collaboration among grid partners (resource Beginning with this issue, AGU has providers and user communities) with a trust Earth Science Community Needs redesigned the front page of Eos to accom- agreement to guarantee security and con- modate a new post offi ce requirement for fi dentiality. A user must be authorized by a Since the International Geophysical Year publishers and business mailers regarding certifi cation authority and must belong to a in 1957–1958, the Earth science community the placement of U.S.