Agriculture and Information Technology

Fall 2011 AGRICULTURE AND INFORMATION TECHNOLOGY

The BRIDGE LINKING ENGINEERING AND SOCIETY

Information Technology and Agriculture: Global Challenges and Opportunities K.C. Ting, Tarek Abdelzaher, Andrew Alleyne, and Luis Rodriguez How Smart IT Systems Are Revolutionizing Agriculture Jason K. Bull, Andrew W. Davis, and Paul W. Skroch The Impact of Mechanization on Agriculture John F. Reid Agriculture and Smarter Food Systems Matthew Denesuk and Susan Wilkinson Multiscale Sensing and Modeling Frameworks: Integrating Field to Continental Scales Indrejeet Chaubey, Keith Cherkauer, Melba Crawford, and Bernard Engel

Promoting the technological welfare of the nation by marshalling the knowledge and insights of eminent members of the engineering profession. The BRIDGE

NATIONAL ACADEMY OF ENGINEERING

Irwin M. Jacobs, Chair Charles M. Vest, President Maxine L. Savitz, Vice President Thomas F. Budinger, Home Secretary Venkatesh Narayanamurti, Foreign Secretary C.D. (Dan) Mote Jr., Treasurer

Managing Editor: Carol R. Arenberg Production Assistant: Penelope Gibbs The Bridge (ISSN 0737-6278) is published quarterly by the National Aca demy of Engineering, 2101 Constitution Avenue, N.W., Washington, DC

20418. Periodicals postage paid at Washington, DC. Vol. 41, No. 3, Fall 2011 Postmaster: Send address changes to The Bridge, 2101 Constitution Avenue, N.W., Washington, DC 20418. Papers are presented in The Bridge on the basis of general interest and timeliness. They reflect the views of the authors and not necessarily the position of the National Academy of Engineering. The Bridge is printed on recycled paper. © 2011 by the National Academy of Sciences. All rights reserved.

A complete copy of The Bridge is available in PDF format at http://www.nae.edu/TheBridge. Some of the articles in this issue are also available as HTML documents and may contain links to related sources of information, multimedia files, or other content. The Volume 41, Number 3 • Fall 2011 BRIDGE LINKING ENGINEERING AND SOCIETY

Editor’s Note 3 Agriculture and Information Technology Andrew Alleyne 5 Innovation and Urgency: Toward a New Green Revolution Mike Baroni

Features 6 Information Technology and Agriculture: Global Challenges and Opportunities K.C. Ting, Tarek Abdelzaher, Andrew Alleyne, and Luis Rodriguez Information technology could have as big an impact on agriculture in the next half-century as mechanization had in the previous century. 14 How Smart IT Systems Are Revolutionizing Agriculture Jason K. Bull, Andrew W. Davis, and Paul W. Skroch Rapid advances in biotechnology, breeding, and agronomics require equally sophisticated information systems. 22 The Impact of Mechanization on Agriculture John F. Reid In the future, agricultural machines will become data-rich sensing and monitoring systems. 30 Agriculture and Smarter Food Systems Matthew Denesuk and Susan Wilkinson Requirements for greater transparency about where food comes from and how it is treated along the way are becoming more stringent. 39 Multiscale Sensing and Modeling Frameworks: Integrating Field to Continental Scales Indrejeet Chaubey, Keith Cherkauer, Melba Crawford, and Bernard Engel Models informed largely by sensor-supplied data can provide insights into the role of agriculture in carbon, energy, nutrient, and water cycles.

NAE News and Notes 47 NAE Newsmakers 49 2011 Japan-America Frontiers of Engineering Symposium Held in Osaka, Japan

(continued on next page)

The BRIDGE

50 Second China-America Frontiers of Engineering Symposium 51 Penn Engineering and NAE Explore “Engineered Networks” at NAE Regional Meeting and Symposium 52 New Study of Integrated STEM Education 53 EngineerGirl! Essay Contest Winners 54 Projects by the Center for the Advancement of Scholarship on Engineering Education (CASEE) 55 Presentations at the Directors’ Guild of America 55 NAE-IOM “Go Viral” Challenge 56 New Volume of Memorial Tributes Available 56 Calendar of Upcoming Events 57 NAE Annual Meeting, October 16–17, 2011 57 In Memoriam

59 Publications of Interest

The National Academy of Sciences is a private, nonprofit, self- The Institute of Medicine was established in 1970 by the National perpetuating society of distin guished scholars engaged in scientific Academy of Sciences to secure the services of eminent members of and engineering research, dedicated to the furtherance of science and appropriate professions in the examination of policy matters pertaining technology and to their use for the general welfare. Upon the author- to the health of the public. The Institute acts under the responsibility ity of the charter granted to it by the Congress in 1863, the Academy given to the National Academy of Sciences by its congressional char- has a mandate that requires it to advise the federal government on ter to be an adviser to the federal government and, upon its own scientific and technical matters. Dr. Ralph J. Cicerone is president of the initiative, to identify issues of medical care, research, and education. National Academy of Sciences. Dr. Harvey V. Fineberg is president of the Institute of Medicine.

The National Academy of Engineering was established in 1964, The National Research Council was organized by the National under the charter of the Nation al Academy of Sciences, as a parallel Academy of Scienc es in 1916 to associate the broad community of organization of outstand ing engineers. It is autonomous in its adminis- science and technology with the Academy’s purposes of furthering tration and in the selection of its members, sharing with the National knowledge and advising the federal government. Functioning in Academy of Sciences the responsibility for advising the federal gov- accordance with general policies determined by the Academy, the ernment. The National Academy of Engineer ing also sponsors engi- Council has become the principal operating agency of both the neering programs aimed at meeting national needs, encourages edu- National Academy of Sciences and the National Academy of Engi- cation and research, and recognizes the superior achievements of neering in providing services to the government, the public, and the engineers. Charles M. Vest is president of the National Academy scientific and engineering communities. The Council is administered of Engineering. jointly by both Academies and the Institute of Medicine. Dr. Ralph J. Cicerone and Charles M. Vest are chair and vice chair, respectively, of the National Research Council. www.national-academies.org FALL 2006 3 Editor’s Note

the amount of protein in people’s diets, water use will increase dramatically as well. Urbanization also leads to so-called secondary effects in overall agricultural demand. Populations concen- trated in cities put a strain on the agricultural supply chain by increasing the geospatial separation between food production and food consumption. In addition, the increased need for transportation of agricultural products, either crops or animals, for processing and eventual consumption adds logistical burdens and increased costs Andrew Alleyne for fuel, infrastructure, and time. Stretching a supply chain also increases its vulnerabilities to disruption and decreases its efficiency. Agriculture and With this background in mind, the articles in this Information Technology issue examine ways to meet increases in demand using the knowledge and natural resources currently at our The topics addressed in this issue of The Bridge disposal, including information technology (IT), which respond to two major trends that affect our planet: we feel has been underused in the agricultural context. population growth and urbanization. The predicted IT is defined here as the use of information to enable or population growth for the first half of this century is improve products or processes. daunting. Depending on the estimate, there will be 9 to IT has transformed many other aspects of human 10 billion people by mid-century. The current popula- endeavor and has helped create systems for responding tion is just under 7 billion, meaning that there will be to a wide range of societal needs. Indeed, transporta- about a 50 percent increase from the beginning to the tion, communication, national security, and health sys- middle of this century. One may debate the relative tems are completely reliant on IT to perform even basic accuracy of particular models, but they all agree that functions. However, information, and its automated there will be many, many more mouths to feed in the technological embodiments, has not impacted agricul- coming decades. ture to the same level. In addition to the change in sheer numbers, there Modern urban public transportation systems have will also be a change in demographics by mid-century become remarkably reliable and predictable with the —a global trend toward urbanization—a shift from an incorporation of information and automation. Indeed, agrarian lifestyle to a city-based lifestyle for the majority they would fail to operate without information flow. of people in the world. For the first time in human his- The same can not be said for modern farming. tory, more people today live in cities than in rural areas. Many National Research Council (NRC) studies have It should be obvious to everyone that population growth been published related to agriculture. A recent report on coupled with increasing urbanization will require a dis- sustainable agriculture explored motivations for sustain- proportionate increase in agricultural output. ability and even recommended agricultural procedures One consequence of urbanization is a rising middle rooted in the concept of sustainability (NRC, 2010). class and an accompanying change in diet, specifically Other studies on the role of technology for agriculture in a higher demand for protein. Protein requires signifi- developing nations (NRC, 2008) have explored the suit- cantly greater agricultural resources to produce than ability of technologies for improving agricultural yields diets based directly on staple crops such as maize or rice. in global regions with high rates of poverty. To date, In addition to land and water requirements, protein however, neither NRC nor NAE has conducted a study requires feedstock crops, thereby creating a multiplier explicitly on the role of IT in the agricultural landscape. effect. Agriculture already consumes a large fraction of A fundamental premise of this issue of The Bridge is the total amount of water used by people. By increasing that more could be done with information to tighten The 4 BRIDGE up the supply chain associated with agricultural produc- agricultural productivity. He also discusses the role of tion. The goal is to wring out inefficiencies and maxi- information and communication technology (ICT) in mize the productivity potential of our current resources current agricultural systems. These ICT-integrated sys- to meet predicted demand. tems started on machines (tractors, combines, etc.) but Conversations, technical and otherwise, about spe- are rapidly spreading to the entire agricultural produc- cific aspects of IT as it relates to particular areas of tion chain. Machine-to-machine communication and agriculture have been initiated. One subject of these machine-to-field communication have become current conversations is the increased use of sensing modalities practice. The author ends with a look to the future farm- and the ubiquity of sensors across the agricultural spec- site, which will be automated and information rich with trum, from overall crop performance and animal health agricultural mechanization systems acting as both infor- to DNA reports on individual genomes. However, con- mation gatherers and physical implements. This level versations relating IT to agriculture in a broader sense of information would influence the production value have not been as prevalent. The modest goal of this chain before and after the farm-site itself and afford us issue of The Bridge is to initiate broader conversations an opportunity to meet future agricultural demands. by illustrating the reality and potential of IT to impact The fourth article, by Denesuk and Wilkinson of the tremendous challenge of feeding a growing planet. IBM, describes how information is used to track and The first article, by Ting, Abdelzaher, Alleyne, and trace food in a supply chain. The authors emphasize Rodriguez of the University of Illinois, provides a broad benefits that accrue when the “physical” world meets perspective on opportunities and challenges associ- the “digital” world. The article also provides a list of ated with integrating IT and agriculture. They begin key priorities for supporting a maximum impact on agri- by identifying individual challenges associated with cultural business productivity and efficiency throughout meeting future agricultural goals. They then break the the value chain discussion down into a taxonomy and show how differ- The final article, by Chaubey, Cherkauer, Crawford, ent elements of IT can be used to address these chal- and Engel of Purdue University, is an overview of sens- lenges. They also offer suggestions for accelerating the ing methods and modalities for agriculture. The authors integration of IT and agriculture and point out potential describe the specific physical quantities that must be challenges associated with the close coupling of IT and converted into information, as well as the transduction agriculture. Much like the unintended consequences mechanisms that perform the conversion. In addition, of introducing a non-native species into a new habitat, they discuss the need for integrating agricultural sens- the accessibility and malleability of information must ing systems that operate on a wide range of spatial and be carefully considered during large-scale deployment temporal scales. of IT in agriculture. One clear message that permeates the collection of The second article, by Bull, Davis, and Skroch of articles is the complexity associated with integrating Monsanto, describes how IT is used in plant breeding. agriculture and IT. Another message is the imperative The use of biotechnology to isolate specific desired traits that we move forward from where we are. In fact, we in crops is akin to techniques used in the pharmaceutical really have no other choice. industry. A combinatorial approach is used to identify We recognize that this issue of The Bridge will not novel genes for traits, such as insect resistance, herbi- be the final word on the topic, but we hope it will be a cide tolerance, and drought tolerance. This information valuable part of the conversation between technologists is then coupled to large-scale trials for determining the and policy makers working to ensure a secure future for most viable candidates. The process generates many our nation and our planet. terabytes of data each day that must be sifted through, correlated, and used in a decision-making process to cull less effective options. IT is a crucial enabler for turning these data into useful product-development decisions. Andrew Alleyne The third article, by John Reid of John Deere, Ralph and Catherine Fisher Professor of Engineering describes the history and state of the art in agricultural and Associate Dean for Research mechanization. Reid begins with a historical perspective University of Illinois, Urbana-Champaign on the importance of mechanization to increasing FALL 2011 5

Innovation and Urgency: Toward a New Green Revolution

Of all the factors underly- technology is already furthering progress toward each ing agriculture’s productivity of these goals. gains in the past 60 years— Applying existing knowledge and processes in regions better seeds, irrigation, fer- where agriculture has been slow to realize its potential tilizer, crop protection, soil could produce additional gains. For example, the UN management, more sophis- Food and Agriculture Organization has demonstrated ticated machinery—the that farmers in the developing world can reduce post- most critical may have been harvest losses from 15 percent of what they grow to near a sense of urgency. When zero just by using basic metal silos to protect harvested Norman Borlaug, the father crops. Results such as this demonstrate why we have Mike Baroni is vice president, of the Green Revolution, established the ADM Institute for the Prevention of Economic Policy, Archer Daniels exported his disease-resistant Postharvest Loss at the University of Illinois. Midland Company. wheat from Mexico to India To make better use of crops farmers already grow, and Pakistan in the 1960s, we are teaching cattle feeders to replace grain in their he was driven by a desire to achieve “a temporary suc- animals’ rations with corn stalks, cobs, and leaves cess in man’s war against hunger and deprivation.” treated with slaked lime—a process first described The revolution that followed was based both on sci- in scientific journals in the late 1950s. And in west entific innovation and a determination to alleviate Africa, we’re helping cocoa farmers improve yields human suffering. and quality simply by providing training in sustainable This same sense of purpose, coupled with new agronomic practices, access to basic market informa- technology, is inspiring individuals and organizations tion, and other services often taken for granted in the throughout the agricultural value chain to meet the developed world. challenges of a global population headed toward 9 bil- In our view, these simple innovations and interven- lion by 2050. To ensure that agriculture can meet future tions—along with the information technology, seed needs, the world will have to (1) reduce the millions of technology, and equipment being developed today— metric tons of post-harvest crops lost to pests, disease, suggest that the world might be on the verge of a new and inferior storage methods each year; (2) make bet- Green Revolution. Despite constraints on water, arable ter use of the crops and biomass already grown while land, and other resources, we believe that with ongoing using water and other inputs sparingly; and (3) increase innovation, investments in agricultural research and yields on existing land in ways that minimize the need infrastructure, and partnerships throughout the value to bring additional acres into production. As you will chain agriculture can meet the needs of a growing world learn in this edition of The Bridge, better information sustainably and responsibly. Information technology could have as big an impact on agriculture in the next half-century as mechanization had in the previous century.

Information Technology and Agriculture Global Challenges and Opportunities

K.C. Ting, Tarek Abdelzaher, Andrew Alleyne, and Luis Rodriguez

K.C. Ting Tarek Abdelzaher Andrew Alleyne Luis Rodriguez

The world population is on track to increase by approximately one-third— to more than 9 billion people—in the next 40 years. Feeding a population that large, which may require doubling agricultural productivity, will only be feasible with significant, even revolutionary, improvements in agricultural processes and equipment. In the past, agriculture has benefitted from, and often driven, improvements in technology. For example, in this country, increased mechanization on modern farms led to the change from an agrarian to an urban demographic (NAE, 2000).

K.C. Ting is professor and head of the Department of Agricultural and Biological Engineering; Tarek Abdelzaher is a professor in the Department of Computer Science; Andrew Alleyne is Ralph and Catherine Fisher Professor of Engineering and associate dean for research; and Luis Rodriguez is an assistant professor of agricultural and biological engineering and has an appointment in the Information Trust Institute. All are at the University of Illinois, Urbana-Champaign. FALL 2011 7

In this article we explain how information technolo- the semiconductor fabrication industry, we immediately gy (IT), a relatively recent field that began a little more understand that, because of external forces, the levels than 60 years ago (Shannon, 1948), could have at least of precision in crop or herd yield are far lower than in as big an impact on agriculture in the next half-century other industries. as mechanization had in the previous century. We present a systems-level perspective on the challenges and Large Geospatial Dispersion opportunities afforded by the integration of agriculture Various points in an agricultural supply chain are and IT. widely dispersed, and the overall agricultural system can be segmented into interconnected sub-processes Agricultural Challenges (Figure 1). Because agriculture is land-intensive, it In this section we outline five challenges associated requires significant acreage for the in-field parts of the with agriculture that must be overcome to achieve the process chain, which are usually not collocated with desired increases in productivity: inherent heterogene- upstream or downstream processes. ity; unanticipated disturbances; large geospatial disper- Another challenge related to geospatial dispersion sion; security and safety requirements; and constraints. is that the processing of some raw materials such as livestock and perishable crops is time critical. Thus, Inherent Heterogeneity long distances between processing points in the supply By its very nature, agriculture works with biological chain introduce risks into the overall viability of the systems that are inherently heterogeneous in compo- entire process. sition and processes. Fields can vary in soil type and moisture content down to a resolution of a square Security and Safety Requirements meter. Weather patterns can vary spatially and tem- Security and safety are paramount for agricultural sys- porally in terms of sunlight and rain. Raw materials tems on two separate time scales. On a short time scale themselves have basic genetic variations from plant to of days or weeks, the safety of food is critical because plant and animal to animal. Indeed, genetic variation many products are eventually ingested by humans. Pro- is often biologically useful for increasing resistance to tecting human health requires tight process manage- diseases and pests. But compare this heterogeneity with ment of the agricultural supply chain on a global scale. the homogeneity of other industrial processes, such as On a longer time scale of years or decades, sustain- Henry Ford’s assembly line, and the challenges associ- ability of the natural environment is critical to long- ated with maximizing product yield using a minimum of term societal health. Agricultural enhancements, resources become apparent. such as pesticides or fertilizers, must be used in ways

Unanticipated Disturbances Agricultural processes are much more vulnerable to unanticipated disturbances than many other industrial processes. The weather can cause floods or bring hail storms that can devastate crops. Pest or disease infestations can rapidly affect, if not wipe out, large quantities of raw material. When we contrast this environment to the carefully controlled (tem- FIGURE 1 Chain of processes for agricultural crops. In the pre-field state, much of the work in crop science and genetics is perature, humidity, etc.) associated with bio-informatics. In the in-field state, much of the work depends on agricultural mechanization. In the post-field clean-room environment of state, logistics, large-scale processing, and supply chain management are key factors. The 8 BRIDGE that increase productivity without adversely affecting of existing processes, rapid responses to changes in the the overall quality of life. Similarly, resources, such supply chain, and optimal allocation of resources more as land and water, must be used in ways that can be feasible. These approaches can be summed up as infor- sustained indefinitely. mation gathering, information processing, and decision making, the same principles that apply to management Constraints of supply chains in other industries. All of these challenges must be met within the constraints inherent in the agricultural process. For exam- Information Gathering ple, the amount of arable land is relatively fixed globally, The ability to gather agricultural information through particularly in more developed countries. Time is also advanced sensing systems has been improving steadily a key constraint, particularly for time-sensitive systems, over the past two decades, and the unit cost per bit such as livestock and perishable produce. There is a of information has decreased in line with similar cost finite window of time during which these agricultural decreases in other industries. For agriculture, there are products are viable during processing. Finally, the first three major types of sensing systems: physical, biologi- four challenges must be met within the constraints of cal, and human. economic viability (i.e., cost). Physical Sensors Opportunities Enabled by Physical sensors, the most prevalent of the three Information Technology types of sensing systems, are used throughout the chain The challenges described above also provide signifi- shown in Figure 1. DNA sequencers for crops gather cant opportunities for using IT. Indeed, if agriculture genetic information about crop varieties and provide can be seen as a supply chain, then many of the exist- input for bio-informatics algorithms that can identify ing paradigms for managing supply chains are already promising strains (e.g., herbicide-resistant soy or wheat) available. (Gale and Devos, 1998). Moisture sensors correlate Besides upgrading physical infrastructure, the effi- soil-to-capacitance changes for the in-field monitoring ciency of supply chains can be improved by the acqui- of moisture. In addition, nitrates and other chemical sition and exploitation of information resident in the compositions can be sensed by in-field devices that are chain itself. Access to these data makes monitoring commercially available (Ehsani et al., 1999). Remote sensing is used to collect broad geospatial information about in-field systems (Nowatzi et al., 2004). For example, Figure 2 shows different electromagnetic reflections for two health states of sugar beet crops as the result of complex absorption and reflection of solar energy. A thorough understanding of this spectrum for specific crops can be used to calibrate a sensor that can provide health or growth data on particular agricultural systems. A similar technique can be used to monitor livestock by sens- FIGURE 2 Spectral signatures of two contrasting states of sugar beet crops (Nowatzi et al., 2004). The different wavelengths ing, for example, methane of the electromagnetic spectrum range from short (deep UV) to long (infrared). Source: Adapted from Nowatzi et al., 2004. emissions. FALL 2011 9

Information gathering in agricultural processing facil- that is more useful than any individual unit of infor- ities is done with many of the same sensing modalities mation. The combination of simultaneously separating used for monitoring conditions (e.g., temperature, mass and aggregating information is the real power behind flow, chemical composition, etc.) in chemical process- information gathering (Sonka et al., 1999). ing facilities. Global information systems (GIS) can pinpoint geo- Information Processing spatial locations of agricultural units in a supply chain Processing of information entails manipulating data using radio frequency identification and global position- to change it from a less useful form to a more useful ing systems (Attaran, 2007). These systems can track form. This computational activity turns the binary 1s and monitor crops and livestock as they progress “from and 0s of the data world into information that can be farm to fork.” used by humans. In this section we describe modeling and data mining, two major aspects of information pro- Biological Sensing cessing related to agriculture. Biological sensing, sometimes called indirect sensing, is physical sensing augmented by biological cues. The state of produce can be inferred based on the monitoring of a biological agent, such as an insect or bacterial popu- The simultaneous lation. This can be done in-field or during post-field separation and aggregation transport and processing. An illustrative example would be a model for the relationship between crop yields and of information is the the presence of pollinators. This model would both monitor pollinator counts and provide a basis for infer- key to effective ring future crop yields (Garibaldi et al., 2011). information gathering. Humans in the Loop Not all data about agricultural systems come from Modeling engineered instrumentation and sensors, such as those The purpose of modeling is to represent the physical mentioned above. Humans in the loop are an impor- world in computational simulations, which can provide tant data source that should not be overlooked. Indeed, a basis for making predictions. Currently, many differ- networking advances in this age of IT can significantly ent types of models are being developed, each one tai- improve our understanding of the workings of highly lored to a specific use and community. For example, distributed, poorly automated socio-physical systems models of genetics are associated with bio-informatics. that may nevertheless significantly impact the planet’s Models of nutrients and water flow (Robinson et al., food supply. This is especially pertinent in develop- 2000) throughout a field may be cued to in-field growth ing countries, where cell phone penetration is exten- models and used to predict yields. Models can also be sive, especially compared to other technologies such used to predict the behavior of the overall supply chain as computers or personal vehicles. Although “crowd- (Stutterheim et al., 2009). Even though the accuracy sourcing” has been used effectively in other domains,1 of these models is often debated, they are useful for its application in the agricultural domain is still largely scenario-based planning and decision making. unexplored. Opportunities for information gathering in agricul- Data Mining ture are primarily in networking individual pieces of Data mining is a technique for extracting knowl- information from many sources throughout the supply edge and information from unstructured data sets. chain. The individual pieces of sensed data contain Data-mining tools have been used in limited ways for local information at time and spatial resolutions that agriculture, but there are still enormous untapped oppor- allow for increasing levels of “separability” or granular- tunities. Much like modeling, data mining has so far ity. Taken together, these data provide a global view been focused largely on individual uses and communities. Today, data-tagging protocols and search patterns 1 See http://www.google.org/flutrends/. are being used to tease out hidden relationships to The 10 BRIDGE predict trends in agricultural processes (Mucherino et on a space and time scale relevant to their particular al., 2009). A pure data-mining approach would make interests. For example, manufacturers of agricultural possible inferences based solely on data, thereby elimi- equipment have adopted systems that allow GIS sensor nating the need for developing infrastructure associated data to flow to machines and control the site-specific with modeling. applications of seeds, fertilizers, or pesticides. Similarly, seed companies have the ability to monitor crop growth Opportunities for the Future trials in several different fields and select useful genetic Although data mining based on specific knowledge varieties for market in a time frame suitable for future of a particular agricultural system usually provides more planting seasons. insight than completely unstructured searches, model- However, agriculture has several domains and several ing tools, used in conjunction with data mining, provide stakeholders within these domains (Figure 1), and our a framework for searching for hidden correlations or current ability to gather information across the supply relationships. The coupling of models and data frame- chain outpaces our ability to synthesize and make time- works also has the potential to provide insights into the critical decisions that affect the system as a whole. entire supply chain as a system. By using a combination of these technologies, we Levels of Decision Making should be able not only to predict behavior, but also to Decision making in agricultural systems occurs on gather enough information to determine whether the many levels. On the micro-scale, individual economic predictions are accurate. Moreover, macro-scale mod- and technical decisions can be made down to the level els and data frameworks can be incorporated into local, of an individual animal or plant. For example, a deci- high-resolution models and data frameworks to clarify sion can be made about how much of a fertilizer to apply overall supply chain behavior. to a single plant based on the cost of the fertilizer and Bringing together disparate types of information the potential yield of the plant. Currently, micro-scale processing will be challenging but could provide many decisions at one point in the supply chain are made benefits. For example, we would be able to “zoom in” independently of micro-scale decisions at other points to see how micro-scale activities, such as individual in the supply chain. systems in the supply chain, might be affected by On the so-called macro-scale, global decisions that macro-scale processes acting as boundary conditions affect multiple aspects of the overall supply chain must to those systems. Simultaneously, this modeling and be made about agricultural policy. Decisions at this data-mining capability would allow us to “zoom out” level include how much surplus of a staple crop to store and see how macro-scale processes influence, or even in reserve or whether to allow genetically modified dominate, interconnected networks of micro-scale foods into a given country. Currently global decision sub-systems. making and local decision making are only loosely coupled, particularly in developing countries where there is relatively little use of IT in agriculture.

Currently, global decision The Need for Coordinated Decision Making making and local Opportunities for decision making in agriculture are enhanced and enabled by the ubiquity of information decision making are gathering and the power of modern information processing. If, for example, data analytics associated with loosely coupled. data mining tools can be combined with contextual or situational information from models or sensor data, the combined data could provide input for powerful com- Decision Making puting algorithms that could provide support for the Moving effectively from data to decisions can often right decision choice at the right time. Moreover, such mean the difference between success and failure for an decisions could be made scalable so decisions for one enterprise. Many individual organizations have inte- geospatial location or one agricultural product would grated information systems to assist in making decisions not conflict with other decisions. FALL 2011 11

This level of coordinated decision making on a regional basis is currently lacking in modern agricultural practice, despite the significant benefits it would provide in yield and efficiency. Lessons can be learned from other fields, such as the U.S. Depart- ment of Defense (DOD), which often has to make large-scale, time-critical, distributed decisions that affect a value chain or supply chain. An illustrative example (Figure 3) shows how DOD fuses information from multiple intelligence, surveil- lance, and reconnaissance (ISR) platforms to come up with near real-time predictions and decisions to com- FIGURE 3 A visualization by the Thunderstorm Program showing multiple sources of information that are fused to support rapid mit resources (Lemnios, decision making and resource deployment over a large geospatial area. 2010). The goal of this program, named Thunderstorm, is to demonstrate the Exploiting Opportunities and identification of fast-moving threats (drug runners) from Meeting Challenges large data sets and make decisions about where to place Capitalizing on the kinds of opportunities described interception points. This requires massive amounts of above will require using technologies in a coordinated information from maritime and satellite ISR platforms, way. Information gathered must be fed to processing intensive computing based on oceanographic models systems housed in data centers of respective agribusi- and data-mining techniques, and decision-making tools nesses. The data can then be mined and used in deci- for detecting threshold levels for taking action. sion making, either automated or by humans. Relevant An analogy to agriculture would be the sudden decisions could include the type and quantity of infor- emergence of crop pests or livestock disease that must mation to be gathered in the first place. be detected (information gathering), damage levels By taking advantage of the combined capabilities and transmission paths that must be predicted (infor- of information gathering and information processing mation processing), and global or local interdiction techniques, decisions can be made that address the strategies that must be created and implemented (deci- challenge of inherent heterogeneity by accommodating sion making). it, and even exploiting it when this would be advanta- Although current capabilities in agriculture are not geous. For example, GIS and site-specific application on the level of the DOD example given above, the can exploit heterogeneities in soil in a field by spatially future of agriculture and IT lies in this direction. Spe- varying the application of seed or fertilizer to minimize cifically, the integration of information can help address the total amount used. many of the challenges identified in the opening section Unanticipated disturbances, such as changes in of this article: inherent heterogeneity; unanticipated weather or outbreaks of disease, are inevitable occur- disturbances; geospatial dispersion; security and safety rences in the agricultural setting. As genetic modifi- requirements; and constraints. cations to crops and livestock produce increasingly The 12 BRIDGE monocultural, biological systems, susceptibility to such tion in the IT space. We believe that agriculture would disturbances increases. One can react to these dis- benefit similarly if industry-government partnerships turbances, or one can predict and anticipate them. were to define agriculture-specific goals and standards In either case, information can inform and accelerate for information thereby introducing uniformity and the response. unleashing innovation similar to innovation in the On a shorter time scale, such as months, an unexpected telecom and computing industries. This could be par- flood might cause local food shortages creating subse- ticularly important in developing countries where mod- quent food security issues. In this situation, resources ern IT infrastructure (e.g., the cellular communication can be re-routed based on supply chain information. network) tends to leapfrog established frameworks. On a longer time scale, computational modeling can be used to predict climatic changes and their effects, Summary and sensing can be used to validate and calibrate mod- Agriculture is a critical human activity that will eling predictions and extrapolations. This information become increasingly important as the world population capability can then be used in scenario-based planning grows. In fact, given the necessity of feeding a grow- to guide decisions about the types of genetic enhance- ing planet, we have no choice but to increase our out- ments of crops or livestock that would be most suitable put with available resources, and IT, which offers many for coping with predicted environmental changes. opportunities for addressing challenges to increasing Currently, the challenge of geospatial dispersion is global agricultural output, must play a central part in addressed by using IT to “tag” individual crops or animals meeting that goal. Although IT has influenced many in the supply chain and geo-locate them in space and industries and greatly improved the management of sup- time. As the price of tags and readers comes down, the ply chains, to date it has only influenced agriculture in a ubiquity of sensed information will certainly increase. relatively localized way. It will take buy-in by all stake- Security and safety challenges can be met by improv- holders and coordinated, communal efforts on a global ing awareness of all aspects of the supply chain. Contin- level to integrate agriculture and IT to meet the needs uous monitoring of crops or livestock can alert us to the of a hungry world. presence of potentially threatening biological elements, enabling a prompt response. If a malicious activity References occurs, data-mining techniques coupled with appropri- Attaran, M. 2007. RFID: an enabler of supply chain opera- ate information gathering along the supply chain can tions. Supply Chain Management: An International Jour- be used to pinpoint the location of the security breach nal 12(4): 249–257. and provide information about how it propagates. This Ehsani, M.R., S.K. Upadhyaya, D. Slaughter, S. Shafii, and information can then be used to make a decision about M. Pelletier. 1999. A NIR technique for rapid determina- the appropriate response. tion of soil mineral nitrogen. Precision Agriculture 1(2): As IT is increasingly used to exploit opportuni- 219–236. ties and address challenges, agricultural output will Gale, M., and K. Devos. 1998. Comparative genetics in the increase, despite the challenging, inherent constraints. grasses. Proceedings of the National Academy of Sciences Higher output per unit time, unit cost, or square meter 95(5): 1971–1974. of land will follow, along with faster and better time- Garibaldi, L., M. Aizen, A. Klein, S. Cunningham, and L. critical decisions. Harder. 2011. Global growth and stability of agricultural yield decrease with pollinator dependence. Proceedings of Future Needs the National Academy of Sciences 108(14): 5909–5914. Many of the technologies and techniques discussed Lemnios, Z.J. 2010. Transforming U.S. Defense R&D to above are already being pursued by individual organi- Meet 21st Century Challenges. Plenary presentation to zations and companies. What we need now is closer SPIE Defense Security and Sensing Symposium, Orlando, collaboration and integration among stakeholders Fla., April 2010. to accelerate the introduction of IT into agricultural Mucherino, A., P. Papajorgji, and P. Pardalos. 2009. Data processes. Mining in Agriculture. Springer Optimization and its Industries such as telecommunications have benefit- Applications, Vol. 34. New York: Springer Verlag. ted greatly from collaboration leading to standardiza- NAE (National Academy of Engineering). 2000. Greatest FALL 2011 13

Engineering Achievements of the 20th Century. Available Shannon, C. 1948. A mathematical theory of communica- online at http://www.greatachievements.org/. tion. Bell System Technical Journal 27(July and October): Nowatzi, J., R. Andres, and K. Kyllo. 2004. Agricultural 379–423 and 623–656. Remote Sensing Basics. NDSU Extension Report AE Sonka, S., R. Schroeder, S. Hoving, and D. Lins. 1999. Pro- 1262. Available online at http://www.ag.ndsu.edu/pubs/ duction agriculture as a knowledge creating system. Inter- ageng/gis/ae1262.pdf. national Food and Agribusiness Management Review: Robinson, B., H. Viswanathan, and A. Valocchi. 2000. Effi- 165–178. cient numerical techniques for modeling multicomponent Stutterheim, R., C.N. Bezuidenhout, and P.W.L. Lyne. 2009. ground-water transport based upon simultaneous solu- A framework to simulate the sugarcane supply chain, tion of strongly coupled subsets of chemical components. from harvest to raw sugar. International Sugar Journal Advances in Water Resources 23(4): 307–324. 111(1324): 250–255. Rapid advances in biotechnology, breeding, and agronomics require equally sophisticated information systems.

How Smart-IT Systems Are Revolutionizing Agriculture

Jason K. Bull, Andrew W. Davis, and Paul W. Skroch

Jason K. Bull Andrew W. Davis Paul W. Skroch

Corn production in the United States has doubled in the last 40 years (Figure 1), primarily as a result of improved cultivars that exploit agronomic advances in soil management and increased planting density (Hallauer et al., 1988). Monsanto, together with other key industry partners, is com- mitted to helping farmers double the yields of corn, soy, and cotton again by 2030 through a mix of advances in biotechnology traits, breeding, and agronomics.1

Jason K. Bull is head of Technology Pipeline Solutions, the R&D IT Division of Monsanto. Andrew W. Davis is a leader in Emerging Strategies, and Paul W. Skroch heads the Pipeline Advancement and Analytics Team, both in Technology Pipeline Solutions.

1 See http://www.monsanto.com/ourcommitments/Pages/sustainable-agriculture-producing-more. aspx. FALL 2011 15

This commitment is based on the success of the current generation of genetically improved crops that have changed the face of agriculture by increasing yield, reducing yield variability, reducing the use of pesticide and insecticide, and raising the profit per acre for farmers (Park et al., 2011). Approximately 95 percent of soybeans and 75 percent of corn grown in the United States have been improved through FIGURE 1 Increase in yields since 1970. Doubling yields by 2030 will require advances in biotechnology, breeding, and agronomics. biotechnology. More than 95 percent of soy beans in Argentina and 50 percent in partnerships to develop the next generation of “smart Brazil have also been improved.2 IT” systems in agriculture. When given a choice, farmers have consistently adopted biotechnology-improved products because of Information, the Heart of Modern Agriculture the advantages listed above. These products, the result Information is essential to modern agriculture, from of sophisticated technological breakthroughs in genom- the creation of new hybrid and varietal products to the ics, breeding, and agronomy, have changed agricultural placement of products in the correct management zones practices and expectations. For example, insect-resistant to capturing value at harvest time. Choosing the cor- crops have reduced the use of pesticides globally com- rect products for a given farm field requires accurate pared with conventional crops (Carpenter, 2010). information about traits in the germplasm, as well as the Agricultural companies continue to develop seed overall performance of those products. As the combina- products with a variety of traits and characteristics tions of biotechnology traits in products become more adapted to specific environmental regions. The next complex, companies must provide more information to generation of products promises further improvements farmers about traits, product performance, and product in yield and stress traits (e.g., improved drought tol- fit to agronomic practices. erance), better use of nitrogen fertilizer by the plant, and broader spectrum insect protection. Breakthrough advances in breeding methodologies (e.g., molecular markers and sequencing) promise higher yielding cultivars, and targeting the resultant seed products to optimal management practices promises additional yield gains (Figure 2). Rapid advances in these complex scientific areas require equally sophisticated information systems. In this article we discuss the changing role of information technology (IT) systems in the evolution of research on seeds and traits as well as on the future of agriculture. We also discuss how systems are evolving to accelerate decisions and opportunities for cross-industry

FIGURE 2 Information connects the technologies required to create the next 2 See http://www.monsanto.com/newsviews/Pages/do-gm-crops- increase-yield.aspx. generation of products. The 16 BRIDGE

FIGURE 3 Candidate products are culled as they advance through phases in the R&D pipeline. Information must be aggregated and reviewed to make the best possible decisions for each phase.

Recommending the right product for the right man- Pipelines for Biotechnology and Breeding agement environment or zone involves analyzing many Demand for biotechnology traits and locally adapted variables (e.g., soil types, drainage patterns, yield in past germplasm has led to the development of agricultural years, insect pressure, disease pressure). Such analyses R&D pipelines similar in overall form to those used require that companies collect information about prod- in the pharmaceutical industry (Figure 3). Product uct performance at refined environmental scales and advancement decisions are made in distinct phases. translate this information into recommendations for In each phase, candidate products are culled, leaving farmers. Increasingly, the product is information itself only the most promising to move on to the next phase. as much as it is seeds and traits. Testing in subsequent phases is increasingly stringent to The development of new products has become a race ensure that the future commercialized product will meet to integrate and leverage the vast amount of informa- the needs of customers. tion produced by industrial research and development Information is generated and consumed by thousands (R&D) pipelines and industry partners. The increasing of scientists in high-throughput product development complexity of information required for product creation, and assessment pipelines. Monsanto has two primary placement, and value capture requires sophisticated IT pipelines, one for biotechnology and one for breeding systems engineered to handle the scale of information (Figure 4). The biotechnology pipeline identifies new and to interrogate that information to support intelli- and novel genes for traits, such as insect resistance, her- gent decisions. bicide tolerance, and drought tolerance. The breeding Success in the marketplace is directly linked to the pipeline creates germplasms with advantageous native efficiency and accuracy of translating research informa- genes that are locally adapted to specific geographic tion into decisions about products. Thus, IT systems regions across the globe. merit significant investment and are considered a key Laboratory and field work, each of which contributes factor in a company’s competitive advantage. to screening and the understanding of novel candidate FALL 2011 17

FIGURE 4 Overview of the Monsanto R&D Pipeline. Biotechnology and breeding form parallel R&D pathways, linked by shared tools and technology. IT platforms are major links that support research, product characterization, and product advancement decisions. genes and germplasm, are integrated in both pipelines. field testing network in the world. Seeds are planted in Information collection is industrial in scale and often millions of field plots every year in thousands of loca- highly automated. tions around the globe, and information about multiple In the biotechnology pipeline, for example, genom- characteristics of each resulting plant is collected (e.g., ics and molecular biology tools are used to select and yield, moisture content, plant height, disease reactions). characterize thousands of genes every year. The pipe- Evaluations are conducted in differing geographies many line is organized into a series of steps, including gene times per year. nomination, sequencing, cloning, transformation, and The logistical operation of this network involves greenhouse and field testing. Each step involves inte- management of seed inventory, distribution, and exper- grated laboratory or field processes that together pro- imental design for every plot planted, harvested, and vide a comprehensive view of gene function and quality. analyzed. Millions of data points are analyzed for the Marker-assisted breeding, a technique in which DNA North American harvest alone. markers are used to identify individuals with favorable gene combinations, is used in the breeding pipeline. Individual seeds are then “chipped,” and the shav- ings are used to analyze the DNA genotype of each seed (Figure 5). Millions of samples are processed and billions of data points are generated every year. The chipping, geno typing, and planting of predicted best selections is extremely time sensi- FIGURE 5 Industrial-scale information collection and management for R&D pipelines. IT systems enable global interchange of tive and highly automated. breeding germplasm. Automation and technological advancements, such as seed chippers and molecular markers for genotyping, Monsanto runs the largest require IT systems for managing analysis and interpretation. The 18 BRIDGE

Finally, the products advanced through both the bio- genes of interest, led to a massive change in the incor- technology and breeding pipelines are integrated (again poration of novel genes into adapted germplasm. That using genetic crosses, field plot evaluations, and molec- change could only be made thanks to wholesale innova- ular markers) to create superior products with the new tions in IT systems, which established high-throughput, genes of interest in the best locally adapted germplasm. high-capacity DNA genotyping laboratories, optimized Creating novel biotechnology products and assessing workflows that automatically select plants with the product performance on this massive scale can only be genes of interest in the best backgrounds, and integrated done with advanced IT systems. these workflows into the context of an already massive, global breeding pipeline. The IT Pipeline Like most companies, Monsanto’s R&D pipelines It can take 10 to 15 years to discover and commer- rely on a mix of third-party and highly customized pro- cialize new genetically enhanced products. The process prietary IT systems that integrate information across requires standardized procedures and regulated testing workflow steps in each pipeline. These systems make to verify trait expression and farm value. To orchestrate it possible to eliminate many manual steps and replace this multistep, multistage, multiyear, interdependent them with repeatable automation. Having IT systems evaluation process, IT systems must model the pipeline do much of the work makes it possible for science to be so candidate products can be tracked and capacity opti- automated and innovations to be leveraged on a mas- mized, results analyzed and interpreted, and compliance sive scale at “the speed of an integrated circuit.” ensured throughout. In other words, it takes an IT pipe- Building such systems at the desired speed and scale line to manage an R&D pipeline. requires creating a new type of IT organization that melds scientific expertise in breeding and biotechnology with computer and industrial engineering. Thus, IT systems are a core the IT organization is comprised of expert scientists in addition to computer engineers—creating a power- component in the transition ful combination of specialized skills in biotechnology, breeding, genomics, bioinformatics, and computer sci- from breakthrough ence in a single organization. This new model for IT enables a rapid transition from science to industrial-scale emerging R&D needs and breakthroughs to engineered implementation. software solutions that industrialize the application of innovations by standardizing workflow steps, automating information production and analysis, and using For example, managing a global-scale plant biotech- standardized algorithms to recommend the best seed nology pipeline requires an IT workflow system that placement decisions possible. connects laboratory assessment points (gene nomina- In the DNA molecular-marker genotyping example, tion, gene construction, transformation) seamlessly to results from genotyping assays must be scored for each track detailed information to understand and evaluate individual seed. With millions of samples, this task can- genes for various traits. The system brings transparency not be done cost effectively by manual inspection of the to the pipeline, enables decisions about gene advance- scores. Automating the process eliminates a workflow ments, and eliminates laborious manual tracking. bottleneck and a source of human error. Another example in the breeding pipeline is the Science at the Speed of IT analysis of field-trial information during harvest, when An organization with the ability to quickly leverage many thousands of designed experiments must be sta- new scientific breakthroughs on a massive scale has a tistically analyzed to determine which candidate prod- substantial competitive advantage, and IT systems are ucts are the best performers in terms of traits of interest, a core component in the transition from breakthrough geographies of interest, and years of evaluation. Because science to industrial-scale implementation. For exam- every harvest takes a period of weeks and advancement ple, breakthroughs more than a decade ago in molecular decisions are time sensitive, analyses must be repeated marker technology, which enabled accurate “tagging” of many times as more information is collected from the FALL 2011 19

field. The data are automatically quality checked and other biotechnology trials nearby? What is the likeli- analyzed within hours of harvesting as new information hood of the stress of interest occurring (e.g., drought)? becomes available. These are just a few of the most straightforward questions that must be answered for any one trial or trait. Smart R&D Pipelines Rely on Smart IT Systems Now consider testing thousands of genes in multiple To accelerate the creation of the next generation of backgrounds in thousands of locations each year, glob- seed products, the next generation of IT systems must ally, in an environment in which product delays can do more than enable decisions and manage complex cost hundreds of millions of dollars in lost opportunity. workflows. These smart IT systems must embody the Designing smart IT systems for decision management processes they support, transform scientific intuition has at least three characteristics: capturing the rules and into repeatable decisions, and enrich and accelerate criteria for making a decision; capturing the context in resource-constrained pipelines. From the perspective of which decisions are made; and capturing the outcome an R&D pipeline, the focus must change from generat- of those decisions so the process can be improved upon. ing and managing information per se to explicitly defin- (Figure 6). ing the decisions made at key points in the pipeline. The creation of smart IT systems requires a para- Defining and Capturing Decisions digm shift in the role of IT systems in the generation The first step in developing a “smart system” is to of advanced seed products. Traditionally, IT systems expose and capture the “decision” itself. The key is have been considered tools for collecting and managing to identify decisions in the pipeline that are repeat- information used by individuals to make decisions. This able and that can be understood as a system of rules mindset focused more on the aggregation of information that can be applied to similar information repeatedly than on how that information could be used to make optimal decisions and how systems themselves could learn from those decisions in a loop of continuous improvement (Figure 6). As the volume and variety of information to be aggregated has increased exponentially, decision making based on information aggregation followed by manual decisions by experts has begun to exceed human capacity. For example, selecting a site to conduct a regulated biotechnology field trial seems simple enough until one considers the variety of questions that must be answered to proceed. Do I have the required regula- FIGURE 6 Building intelligence into a machine requires smart IT systems. A. Traditionally, experts depended on their experience tory permits for the trial? to make the best decisions possible, and their knowledge of the rules and criteria for those decisions, as well as any learning, What was grown the previ- remained with the individual. Thus developing feedback loops could take years, making the accumulation of knowledge a slow ous year in the same field process. B. By building intelligence into IT systems, rules and criteria can be turned into corporate assets that can be optimized that I should monitor? Are and improved upon. The 20 BRIDGE

(Taylor and Raden, 2007). Exposing these rules and terms of scope and scale. Take, for example, the scale criteria requires detailed knowledge about the purpose of DNA sequencing, which can generate terabytes of of experiments and their role in the R&D pipeline. information every day. In addition, projections call for exponential growth every 2 to 3 years (Kahn, 2011). Capturing Decision Context Turning these data into information requires analyzing A smart IT system must do more than enumerate and complex DNA sequences and combining that informa- apply a list of rules that lead to a decision. It must also tion with other information on gene function to iden- capture the context—a snapshot of temporally relevant tify new markers and genes. information—in which the decision is made. Capturing Facing these challenges and others will require opti- the context is important for learning which information mized infrastructure, including specialized computing is relevant and how it can be optimally weighted. platforms (Schadt et al., 2010), optimized data storage, Historical context can be provided by information and networks that support global operations. Designing warehouses that store not only what decisions were and implementing these capabilities will be beyond the made but also the exact information that was used to capacity of a single entity and will require partnerships make them at the time they were made. In this way, among public and private institutions that specialize a record can be built of specific information used for a in each of these areas and have experience in applying decision, which, over time, can be leveraged to recon- industry-standard technologies at scale. struct the decision process. Over time, a system “mem- Building partnerships with IT providers that have ory” is created that can be used to improve and refine demonstrated expertise in a given computing domain future iterations of that decision. will enable new systems to leverage industry-standard solutions. Partnerships in highly technical areas (e.g., Capturing Decision Outcomes sequence assembly optimization) will also enable the A smart IT system must also be able to evaluate and development of sophisticated or custom solutions. “learn” from the outcome of a decision. What was the Monsanto, for example, has partnered with sequencing impact of the decision? Did the products that advanced equipment manufacturers, universities, specialized hard- at one stage continue to advance? Did this candidate ware manufacturers, and large IT companies to tackle product actually perform better, as predicted, when a the production and analysis of sequencing information certain agronomic practice was used? to enable delivery of continually improving seed.

Information, the Next Agriculture Frontier Agriculture is on the threshold of realizing an excit- A smart IT system must be able ing and dynamic opportunity to apply and benefit from to evaluate and “learn” from innovation in IT. The next generation of seed innovations, which will be necessary to double yields by 2030, the outcome of a decision. will depend on how effectively the industry can collect, analyze, and use the explosion of new information to make decisions for the R&D pipeline and the farm. As Humans learn, at least in part, from the consequences products become more complex and their selection is of actions, and a smart IT system must learn in a similar driven by more detailed characteristics of local man- way. By capturing outcomes, the decision model can agement zones, companies will have to provide more be benchmarked and improved upon, thus transforming specific on-farm product information. In fact, this expert decisions into corporate assets that can be learned information will be as essential as the product itself from, measured, modeled, and subsequently improved (i.e., seeds and traits) to realizing maximal yields and upon again by the next generation of scientists. economic benefit. Realizing the goal of doubling agricultural yields Partnerships across Industry by 2030 will require that product performance both Supporting the generation of information from high- improve and become more predictable as multiple traits throughput pipelines and integrating that information are targeted to specific stresses and management zones. to enable smart IT systems are both massive tasks in Optimization of agricultural inputs will become more FALL 2011 21

feasible with the next generation of biotechnology and Realizing this vision will require new partnerships among breeding traits, which will reduce the environmental seed companies, IT providers, and other key players in impact of agriculture by reducing the use of herbicides, the agricultural industry to bring cutting-edge IT to the insecticides, fungicides, and fertilizers. Combined with product development pipeline as well as to the farm. advances in farm machinery and precision agriculture, farmers will be able to cultivate the same or more acres References more profitably. Carpenter, J.E. 2010. Peer-reviewed surveys indicate positive These advances mean that agriculture will also impact of commercialized GM crops. Nature Biotechnol- become increasingly technology driven, and this “tech- ogy 28(4): 319–321. nification” will depend on how we use information. Hallauer, A.R., W.A. Russell, and K.R. Lamkey. 1988. Corn Advances in the extraction and use of information from Breeding. Pp. 463–554 in Corn and Corn Improvement, genomics and molecular breeding will create opportuni- 3 ed., Agronomy Monograph 18, edited by G.F. Sprague ties for identifying new genes for biotechnology traits. and J.W. Dudley. Madison, Wisc.: American Society of Advances in how we use information to define and use AGRONOMY. management zones will lead to optimization of product Kahn, S.D. 2011. On the future of genomic data. Science use and seed performance. Advances in information 331(6018): 728–729. about weather and potential insect infestations will cre- Park, J.R., I. McFarlane, R.H. Phipps, and G. Ceddia. 2011. ate opportunities to manage risk. The role of transgenic crops in sustainable development. All phases of the agricultural pipeline will be trans- Plant Biotechnology Journal 9: 2–21. formed and will require innovations in IT systems to Schadt, E.E., M.D. Linderman, J. Sorenson, L. Lee, and G.P. realize the full potential of the seed products under Nolan. 2010. Computational solutions to large-scale data development and to achieve the goal of doubling yields management and analysis. Nature Reviews Genetics 11: by 2030. Success will be linked to the “intelligence” of 647–657. smart IT systems, because time-to-market for new prod- Taylor, J., and N. Raden. 2007. Smart (Enough) Systems: ucts will be linked, in turn, to how quickly and effec- How to Deliver Competitive Advantage by Automating tively information can be transformed into decisions. Hidden Decisions. Boston, Mass.: Prentice Hall. In the future, agricultural machines will become data-rich sensing and monitoring systems.

The Impact of Mechanization on Agriculture

John F. Reid

Significant challenges will have to be overcome to achieve the level of agricultural productivity necessary to meet the predicted world demand for food, fiber, and fuel in 2050. Although agriculture has met significant challenges in the past, targeted increases in productivity by 2050 will have to be made in the face of stringent constraints—including limited resources, less skilled labor, and a limited amount of arable land, among others. John F. Reid is director, Product The metric used to measure such progress is total factor productivity Technology and Innovation, (TFP)—the output per unit of total resources used in production. According John Deere Moline Technology to some predictions, agricultural output will have to double by 2050 (GHI, 2011), with simultaneous management of sustainability. This will require Innovation Center. increasing TFP from the current level of 1.4 for agricultural production systems to a consistent level of 1.75 or higher. To reach that goal, we will need significant achievements in all of the factors that impact TFP. Mechanization is one factor that has had a significant effect on TFP since the beginning of modern agriculture. Mechanized harvesting, for example, was a key factor in increasing cotton production in the last century (Figure 1). In the future, mechanization will also have to contribute to better management of inputs, which will be critical to increasing TFP in global production systems that vary widely among crop types and regional economic status. For example, a scarce, basic resource that will have to be managed much better is water, a critical input in agricultural production. Both the FALL 2011 23

FIGURE 1 Improved farm equipment and mechanization has been essential to increasing total factor productivity (TFP) as illustrated in this example of the Old Rotation Study for cotton production. efficiency and effectiveness of water use will have to enabled by technologies that created value in agricul- improve dramatically. tural production practices through the more efficient Today, approximately 70 percent of withdrawals of use of labor, the timeliness of operations, and more fresh water are used for agriculture (Postel et al., 1996). efficient input management (Table 1) with a focus on By 2025, 1.8 billion people are expected to be living in sustainable, high-productivity systems. Historically, areas with absolute water scarcity (UN FAO, 2007), and affordable machinery, which increased capability and two-thirds of the world population will live in water- standardization and measurably improved productivity, stressed areas. Improving water management will have was a key enabler of agricultural mechanization. Figure to be achieved by more efficient irrigation technology 2 shows some major developments since the mid-1800s and higher efficiencies in whatever technologies farmers are currently using. TABLE 1 Primary Methods for Productivity In this article, I define the current state of agri- Enhancement in Agriculture cultural systems productivity and demonstrate how information and communication technologies (ICT) 1. Efficient use of labor by: are being integrated into agricultural systems. I also a. removing bottlenecks describe how the integration of ICT will create oppor- b. making efficient use of time tunities for increasing agricultural-system productivity 2. Timeliness of operations by: and influencing productivity beyond the agriculture a. hitting optimum agronomic or business windows value chain. b. reducing spoilage and harvest losses of agricultural products The Impact of Mechanization on Productivity 3. Efficient use of inputs (including water, seeds, nutrients, Agricultural mechanization, one of the great pesticides, etc.) achievements of the 20th century (NAE, 2000), was 4. Enabling sustainable production systems The 24 BRIDGE

For most of the 20th century, four key factors influenced increases in the rate of crop production: more efficient use of labor; the timeliness of operations; more efficient use of inputs; and more sustainable pro- ductions systems (Table 1). These four drivers played out at different rates in different crop production systems, but always led to more efficient systems with lower input costs. Technologi- cal innovations generally increased mechanization by integrating functional processes in a machine or FIGURE 2 A conceptual diagram showing how agricultural mechanization has contributed to increased productivity. crop production system and by making it possible for a by John Deere, a major innovator and developer of farmer to manage increasingly large areas of land. machinery technology. By the late 20th century, electronically controlled In the 19th century, as our society matured, a great hydraulics and power systems were the enabling tech- many innovations transformed the face of American nologies for improving machine performance and pro- agriculture. Taking advantage of a large labor base ductivity. With an electronically addressable machine and draft animals, farmers had been able to manage architecture, coupled with public access to global reasonable areas of land. This form of agriculture was navigation satellite system (GNSS) technology in the still practiced in some places until the middle of the mid-1990s, mechanization in the last 20 years has been 20th century. focused on leveraging information, automation, and Early innovations were implements and tools that communication to advance ongoing trends in the pre- increased the productivity of draft animals and assisted cision control of agricultural production systems. farmers in preparing land for cultivation, planting and In general, advances in machine system automation seeding, and managing and harvesting crops. The have increased productivity, increased convenience, origins of the John Deere Company, for example, and reduced skilled labor requirements for complex were based on the steel-surfaced plow developed by tasks. Moreover, benefits have been achieved in an its founder. This important innovation increased the economical way and increased overall TFP. productivity of farmers working in the sticky soils of the Midwest. From Mechanization to A major turning point occurred when tractors began Cyber-Physical Systems to replace draft animals in the early decades of the 20th Today’s increasingly automated agricultural produc- century. Tractors leveraged a growing oil economy to tion systems depend on the collection, transfer, and significantly accelerate agricultural productivity and management of information by ICT to drive increased output. Early harvesting methods had required sepa- productivity. What was once a highly mechanical rate process operations for different implements. With system is becoming a dynamic cyber-physical system tractors, the number of necessary passes in a field for (CPS) that combines the cyber, or digital, domain with specific implements was reduced, and eventually, those the physical domain. The examples of CPS reviewed implements were combined through innovation into below suggest the future potential of ICT for achieving the “combination” or combine harvester. the target TFP of 1.75 and beyond. FALL 2011 25

Precision Agriculture Precision Guidance Precision agriculture, or precision farming, is a sys- Around the turn of the 21st century, GNSS technol- tems approach for site-specific management of crop ogy had become so precise and accurate that it had out- production systems. The foundation of precision farm- paced the requirement for the early phases of precision ing rests on geospatial data techniques for improving farming and become commercially viable for enabling the management of inputs and documenting produc- a number of automatic-guidance applications (Han et tion outputs. al., 2004). Advances in GNSS technologies include As the size of farm implements and machines decimeter to centimeter accuracy by using signals from increased, farmers were able to manage larger land areas. a geospatially known reference point to correct satellite At first, these large machines typically used the same signals. One premium example is a real-time kinematic control levels across the width of the implement, even global positioning system (RTK-GPS) technology though this was not always best for specific portions of (Figure 3a) that reduces fatigue and lowers the skill the landscape that might have different spatial and other level required to achieve high-performance accuracy characteristics (Sevila and Blackmore, 2001). in field operations. A key technology enabler for precision farming In short, in less than 20 years, GPS technology went resulted from the public availability of GNSS, a tech- from being an emergent technology to a robust, mature nology that emerged in the mid-1990s. GNSS pro- technology that has optimal capabilities for produc- vided meter, and eventually decimeter, accuracy for tion agriculture. A number of solutions are emerging mapping yields and moisture content. A number of today (Figure 3a) for achieving high-precision accuracy ICT approaches were enabled by precision agriculture, through various reference-signal configurations (e.g., but generally, its success is attributable to the design of RTK-GPS, multiple satellite systems, sensor fusion machinery with the capacity for variable-rate applica- with complementary sensors, and multiple sources of tions. Examples include precision planters, sprayers, corrections). fertilizer applicators, and tillage instruments. The predominant control strategies for these systems are based on management maps developed by farmers and their crop consultants. Typically, mapping is done Sensor capabilities turn using a geographic information system (GIS), based on characteristics of crops, landscape, and prior harvest agricultural vehicles into operations. mobile recording systems Sources of data for site-specific maps can be satellite imaging, aerial remote sensing, GIS mapping, field of crop attributes. mapping, and derivatives of these technologies. Some novel concepts being explored suggest that management strategies can be derived from a combination of Operator-guidance aids that provide feedback to the geospatial terrain characteristics and sensed informa- operator about required steering corrections through tion (Hendrickson, 2009). All of these systems are audio and visual cues were the first systems on the enabled by ICT. market for precision guidance. This feature allowed a A competitive technology for map-based precision vehicle system to follow paths parallel to prior opera- farming is on-the-go sensing systems, based on the con- tions across a field. These types of systems worked well cept of machine-based sensing of agronomic properties at decimeter accuracy and required no major control- (plant health, soil properties, presence of disease or system integration into the vehicle. weeds, etc). The immediate use of these data drives The major benefits of these systems were to reduce control systems for variable-rate applications. These overlap/underlap in field operations with extremely sensor capabilities essentially turn the agricultural wide implements, typically for spraying chemicals and vehicle into a mobile recording system of crop attri- fertilizers. The decrease in overlap meant the parsimo- butes measured across the landscape. In fact, current nious use of resources. The decrease in underlap meant production platforms are increasingly becoming tools that chemicals and fertilizers were applied to every part for value-added applications through ICT. of the field. The 26 BRIDGE

The ultimate in unmanned automation is the capability of driving complete field patterns under autonomous management of the tractor- implement functions without frequent operator inter- vention. Figure 3c shows one commercial example of the execution of this concept. The figure shows a very rudimentary form of path planning, integrated with automatic guidance, that can increase productivity by managing the paths a vehicle must follow. Path management can be programmed to reduce time loss caused by navigation (e.g., turning around) and implement management. Like precision agricul- FIGURE 3 Examples of automation in control applied to agricultural mechanization systems: (a) RTK guidance through AutoTrac ture, precision guidance and a GPS reference station; (b) Swathcontrol Pro; (c) iTech Pro enabling precision patterned guidance across a field; and (d) creates data from its preci- HarvestLab for sensing crop properties. sion operations that could be used in crop manage- On the next level of evolution, automatic guidance ment. Examples of these data include information on systems appeared that managed steering for an opera- the “as-applied” state of operations, vehicle paths, and tor through automatic control. Automatic guidance operational state variables. The data can then be used systems enabled precision operations depending on to meet the needs of other ICT in systems automation the type of GNSS signal and how it was integrated into and optimization. the requirements of the agricultural operations. GNSS technology enabled the management of inputs System Automation and Control such as seed, pesticides, and fertilizers with precision Until recently, automation has been focused on func- across the field. For example, the chemical applications that depend on GNSS or direct sensing. How- tion to buffer zones and grassy waterways was reduced ever, processes that lend themselves to control based on based on sensing of the field location of these features. the attributes of soil and crop properties are also being John Deere’s software product, SwathControl Pro (Fig- investigated. Some initial applications of these, which ure 3b), enabled farmers to manage the definition and were coupled with GPS, mapped the yield and moisture execution of this capability. of harvested crop operations. GNSS technology provided the reference signal that It is also possible to use sensing of soil or crop enabled accurate vehicle location at the GNSS sensor, properties—such as controlling the cut-length of a but precision control of the machine required several self-propelled forage harvester (SPFH)—as part of additions to the system (e.g., attitude correction, iner- a combination of techniques to increase machine sys- tial sensors, implement control). With these features, a tem productivity. In this example, the cut-length is mobile CPS could correct the attitude of the vehicle on the section length into which a tree, or forage plant, uneven terrain and manage the vehicle system path for is cut. When an SPFH is operated with static cutting precision in the execution of complex functions. settings, independent of the size of the forage plant, it FALL 2011 27

can consume a significant amount of energy in cutting It is clearly within the vision of the industry to develop forage for ensiling (storage in silos). advanced capabilities (such as those listed below) that HarvestLab™, a sensing technology, uses near infra- leverage these ICT innovations: red (NIR) reflectance sensing to detect the moisture • machine knowledge centers that enable improved content of forage and adjust the cut-length of harvested design, faster problem resolution, and higher system material (Figure 3d). This control strategy can signifi- productivity, increased uptime, and lower operat cantly reduce the energy consumption for harvesting ing costs forage with no degradation in the ensiling process. The results are a significant reduction in fuel consumption • stores of agronomic knowledge that can lead to opti- in the harvest operation and a high-quality cut, which mization of farm-site production systems enables proper forage preservation. • stores of social knowledge related to customer or con- NIR sensing has often been used in the laboratory sumer value-drivers and in grain processing and storage to measure properties (e.g., moisture oil and protein content) of biologi- As ICT continues to penetrate production systems, cal materials, which contributes to value-added uses of a massive network is being developed of machine sys- corn, cereal grains, and forage. As these technologies tems that are platforms for value creation—well beyond mature, ICT has the potential to connect information productivity from agricultural mechanization intended about constituent properties to downstream processes. for the farmer or the farm site. These systems are collecting and managing information with potential value Machine Communications in downstream value-chain operations that use crop or The automation methods described above generate drive systems to achieve environmental sustainability. massive amounts of data. However, the data are not limited to on-vehicle storage or even to on-the-go deci- Worksite and Value Chain Productivity sion making. Inter-machine communication greatly The next step in automation and control is to move increases the potential of these systems. beyond individual vehicle systems to the optimization In the last few years, the commercial application of of production systems and farm worksites. To achieve telematics devices on machines has been increasing this goal, we have developed the beginnings of vehicle in agriculture, thus empowering a closer connection and machine systems that can both sense and control between farmers and dealers in managing machine with precision. These systems can be driven by data uptime and maintenance services. Other applications from a variety of sources to provide precision control. for machine communication systems include fleet and For example, they are capable of collecting, storing, asset management. and transferring information about the crop, field, and In addition, inter-machine communications are machine state at the time of field operation. They can expanding machine system data applications, such as also receive data from public and private data sources. diagnosing and prognosticating machine health. Inter- machine communications can also include implements and tools (e.g., monitoring seeding rate in tractor implement applications). Functionally, a modern, high-end The next step in agricultural machine system is effectively a mobile, geospatial data-collection platform with the capacity to automation and control is the receive, use, sense, store, and transmit data as an inte- optimization of production gral part of its operational performance. As we strive for higher TFP levels, these high-end systems and farm worksites. applications are moving toward systems with increasingly advanced ICT capabilities, including data communication management from machine to off-machine Furthermore, data collected by machines can be data stores. Other ICT capabilities under development transferred to farm-management systems as well as to include vehicle-to-vehicle operations management in public and private sources that require information the field. about production management for quality, compliance, The 28 BRIDGE or value-added purposes. Thus, we are entering an era and simulation that can improve production efficiency of emerging field and farm optimization systems that by anticipating the impact of weather and various pro- can drive up TFP of the worksite, including machines, duction methods. geographies, and cropping systems. In the future, ICT will enable the development of new As intelligent mobile equipment for worksite solu- platforms that can provide more support to production tions has evolved over the last 20 years, agricultural agriculture by taking advantage of opportunities to con- mechanization has also evolved from a bottom-up nect farmers, the value chain, and society in ways that integration of the foundations of ICT applied to basic are beyond present capabilities. The German-funded mechanization systems required for crop production. iGreen project, for example, is working on location- The primary machine capabilities of precision sensing, based services and knowledge-sharing networks for advanced control systems, and communications have combining distributed, heterogeneous public and private created the potential for the emergence of CPS from information sources as steps toward future ICT systems production agricultural systems. (iGreen, 2011). Today, we are extremely close to having Although these advanced technologies are not uni- true CPS and control systems for measuring the “pulse” formly distributed among platforms and production of agricultural productivity on planet Earth. systems, where they exist, there are opportunities to leverage ICT to increase production systems capabili- Conclusion ties. Looking ahead, it is expected that the business Agricultural mechanization will be a key factor to value of ICT will expand to additional platforms. achieving our TFP goals and feeding a growing planet. Technologies integrated on vehicles must work seam- Looking ahead, agricultural machines will become data- lessly with other systems. Drawbacks of some initial rich sensing and monitoring systems that can map the attempts for ICT capabilities have been the signifi- performance of both machines and the environment cant time required for setup or management, the lack they work on with precision resolution and accuracy, of a common architecture, the lack of standardization and this capability will unlock levels of information among industries, and the lack of standardization with about production agriculture that were heretofore the farmer in mind as a user of ICT. Recently, several unavailable. organizations have been working to develop standards, and some improvements have already been developed References or are in process (ICT Standards Board, 2006; U.S. GHI (Global Harvest Initiative). 2011. GHI website. Avail- Access Board, 2010).1 able online at http://www.globalharvestinitiative.org/. Centers that store machine, agronomic, and social Han, S., G. Zhang, B. Ni, and J.F. Reid. 2004. A guidance knowledge will aggregate data to provide value-added directrix approach to vision-based vehicle guidance sys- services for machinery operation and farm manage- tems. Computers and Electronics in Agriculture 43(3): ment. Some of these data may be collected by farmers, 179–195. and some will be provided by public and private sources Hendrickson, L. 2009. Landscape Position Zones and of agricultural information. Some data sources, such as Reference Strips. PowerPoint Presentation. Available remote sensing, have been mentioned, but a number of online at http://nue.okstate.edu/Nitrogen_Conference2009/ others will emerge as the aggregated knowledge in effi- Hendrickson.ppt. cient production agriculture increases. ICT Standards Board. 2006. . . . to coordinate the standardiza- Centers with machine knowledge can help increase tion activities in the field of Information and Communica- equipment uptime and anticipate machine system tions Technology. Available online at http://www.ictsb.org/. failures based on vehicle state variables in operation. iGreen. 2011. Welcome to iGreen! Available online at Machine data that provide a better understanding http://www.igreen-projekt.de/iGreen/. of machine use can also lead to more efficient system National Academy of Engineering (NAE). 2000. Greatest designs that meet the needs of farmers. Agronomic data Engineering Achievements of the 20th Century. Available will create new opportunities for intensive modeling online at http://www.greatachievements.org/. Postel, S.L., G.C. Daily, and P.R. Ehrlich. 1996. Human appropriation of renewable fresh water. Science 1 See also http://asabe.org/, http://aem.org/, http://www.sae.org/, and http://www.iso.org/iso/home.html. 271(5250): 785. FALL 2011 29

Sevila, F., and S. Blackmore. 2001. Role of ICTs for an the Twenty-First Century. Available online at http://www. Appropriate World Market Development. Presentation fao.org/nr/water/docs/escarcity.pdf. at the 12th Members Meeting, Club of Bologna, Bologna, U.S. Access Board. 2010. Draft Information and Commu- Italy, November 18–19, 2001. Available online at http:// nication Technology (ICT) Standards and Guidelines. www.clubofbologna.org/ew/documents/Proc2001.pdf. Available online at http://www.access-board.gov/sec508/ UN FAO (United Nations Food and Agriculture Organiza- refresh/draft-rule.htm. tion). 2007. Coping with Water Scarcity: Challenge of Requirements for greater transparency about where food comes from and how it is treated along the way are becoming more stringent.

Agriculture and Smarter Food Systems

Matthew Denesuk and Susan Wilkinson

The growing world population is putting increasing strains on natural resources, including agricultural resources, and recent economic and environmental trends are making the problem even more acute. The unprec- Matthew Denesuk edented growth of the global middle class is accelerating demand for foods with higher agricultural footprints, such as meats. At the same time, limited arable land areas, increasing water shortages, rising energy costs, and urgent climate/environmental concerns are creating a desperate need for substantial increases in factor productivities.1 In addition, in response to issues related to public health and food supply chains, requirements for greater transparency about where food comes from and how it is treated along the way are becoming more stringent. To meet those requirements, we must find ways to quickly trace the sources of contamination or other health-related issues.

Susan Wilkinson Matthew Denesuk is manager of Natural Resources Modeling and Social Analytics, IBM Research. Susan J. Wilkinson is the global subject expert for Food Safety and Traceability, IBM Global Business Services.

1 For example, consider water as one factor in agricultural production. An increase in water productivity implies that the same level of output could be obtained with less water. Other factors could include labor, seeds, fertilizer, types of machinery, etc. FALL 2011 31

The Transformational Potential of Smarter Agriculture “Physical Meets Digital” Broadly speaking, IT is increasingly impacting agri- In many domains of human activity today, the infu- culture and the overall food system in myriad ways sion of information technology (IT)-enabled “intel- throughout the value chain. Impacts range from fun- ligence” capable of driving major improvements in damental inputs, such as genomics and computer mod- systems-level productivity and performance is accel- eling that can help drive the next generation of seed erating. These improvements will have the most pro- and planting technology, to food distribution, such as found impact on “highly physical” domains, that is, smarter logistics that can help deliver food more quickly domains that involve large-scale infrastructure (e.g., using less fuel and fewer machine resources and with less machinery, large vehicles, static structures, and pro- spoilage en route to the point of consumption. ductive land masses) and that are often exposed to The clear social need and inevitably growing mar- natural elements. Examples include traffic systems, ket demand are driving investment by governments, utilities management (water or electricity), trans- IT firms, and private equity and venture capital firms. portation and shipping, mining, oil and gas, and of Many venture capital investments have focused on course agriculture and the broader food system. In aspects of “physical meets digital,” such as traceability, these cases, the use and performance of such infra- sensing for reducing spoilage, and sensing for agricul- structure tend to substantially determine overall sys- tural optimization and/or water productivity. tem performance. “Agriculture 2.0,” which implies decisive shifts in the To varying degrees, many aspects of these kinds of ways and means of agricultural production, is becoming systems have traditionally been resistant to the infu- a term of art in this context. Because venture capital sion of intelligence, due largely to the difficulty and tends to flourish in environments that are experiencing cost associated with (1) obtaining detailed, timely data market or business model disruptions, such investments about the system components and surrounding envi- should be watched closely. ronments and (2) remotely controlling actions based on analyses of those data. However, manufacturers of heavy equipment are increasingly outfitting their products with deeper lev- Comprehensive data reflecting els of instrumentation—various types of transduction capabilities that produce data reflecting the status of the the state of infrastructure equipment, potentially enabling management or opera- and the environment can tional changes to be performed remotely. In addition, technologies for generating data that reflect environ- now be collected with mental conditions (e.g., in situ sensors, remote sensing satellites, airborne vehicles) are becoming increasingly tolerable latencies. capable and cost effective. Combined with cost-effective communications and interconnection technologies, comprehensive data reflecting the state of the The present article focuses predominantly on two infrastructure and environment can now be collected related areas (see Figure 1) in which “smarter” systems with tolerable latencies, enabling effective actions to enabled by physical-digital integration can have a posi- be executed. tive and global impact: (1) track-and-trace technolo- In addition, technology for effectively instrumenting gies to support food safety and ultimately optimize food “products,” such as food items, pharmaceuticals, and supply chains; and (2) increasing farm multifactor pro- so on, is also becoming more capable and less expen- ductivity by improving water logistics and application, sive. Think, for example, of the increased availability optimizing machine/fleet maintenance, and improving of radio-frequency identification and 2D barcode tech- farm operations/processes. Multifactor productivity nology, wireless environment/gas sensors in shipping includes agricultural business optimization as a result containers, and global positioning system and other of the integration of these approaches with additional location-tracking technologies. downstream business information. The 32 BRIDGE

FIGURE 1 Agricultural value chain and topics in this article.

Food Tracking and Tracing problems back to their sources; and (3) fears of ter- The increasing focus on food safety and security is rorism or other deliberate attempts to damage or con- being driven by a number of factors: (1) the number taminate food supplies. and/or visibility of food safety-related incidents (Fig- The purpose of food tracking and tracing is to be ure 2); (2) the increasing globalization and complex- able to follow food and food components as they flow ity of the food supply system, which contributes to through and are transformed along various pathways in both higher risks associated with the number and the value chain, and, when needed, to be able to fol- diversity of handling entities and diminishing vis- low the flow backward to identify the sources and/or ibility/transparency. The latter effect produces conditions to which the food was subjected along the uneasiness because of uncertainties about where food way. The latter capability is a critical component of comes from, which greatly complicates efforts to trace food safety systems.

FIGURE 2 A sampling of food contamination and recalls since 2006. FALL 2011 33

Tracking and tracing capabilities in food sys- organizations in 108 countries, and their well-known tems can be complicated and require coordination global trade item numbers (GTINs), including UPC among many firms and functions, which may have (Universal Product Code), SSCC (Serial Shipping substantially different business motivations. Broadly Container Code), and EAN (European/International speaking, motivation can be compliance driven, Article Number), have been used by retailers and sup- that is, dictated by some form of government regula- pliers of packaged goods for decades. The adoption tion or customer requirements, or based on business of GS1 standards varies by country and sector but has value, that is, focused on branding, overall risk miti- increased significantly every year, and efforts are under gation, or improving value-chain operational perfor- way to increase adoption by companies in the upstream mance (e.g., by cross leveraging tracking and tracing supply chain. capabilities). GS1 standards for product identification (product One clear benefit of tracking and tracing that applies type and lot numbers) are the basis of a major initiative throughout the value chain is the ability to minimize undertaken by the produce industry to enable traceabil- disruptions caused by a food-safety event. For example, ity back to the farm. The goal of the initiative, called if the cause of the event can be isolated to a particular the “Produce Traceability Initiative” (PTI), is to achieve food from a particular part of a particular farm, the prob- adoption of electronic traceability throughout the sup- lem can be effectively quarantined, and food sales should ply chain for every case of produce by 2012. Although return to normal levels as customer confidence recovers. participation in PTI is currently voluntary, U.S. food The alternative, which frequently occurs, is that con- retailers and their major produce suppliers are actively sumers stop buying the type of food (e.g., cucumbers) moving toward compliance. initially suspected as the cause. Even though the initial suspicion often turns out to be incorrect, sales of many types of food can be adversely impacted as health officials and food-chain participants work to find the culprit. GS1 Standards throughout Even after the culprit is found, it frequently takes a the food supply chain substantial amount of time to identify the location of, and remove the contaminated product from, the mar- would make traceability ket. The longer this takes, the more consumer confidence is eroded, and the longer it takes for sales to much more feasible. return to normal.

Standards and Regulations Execution Issues Under the Food Safety Modernization Act, signed by Tracking and tracing capability has three major pil- President Obama in January 2011, the Food and Drug lars: (1) establishing a premise ID; (2) establishing a Administration is required to execute traceability pilot product ID; and (3) establishing a means of tracking tests and use what they learn to develop new regulations movements and transformations (Figure 3). to improve the ability of the food industry to identify Establishing a premise ID. A unique identifier is the source of food-safety problems and quickly remove required for the source location and for each location contaminated products from the market. For tracking an agricultural good passes through on its journey from and tracing to succeed, however, there must be a will- the farm to the retail outlet. This identifier must be at a ingness to share information across the supply chain, as level of temporal and spatial granularity appropriate for well as agreements among trading partners on standard- the purpose and must include the identity of the firm or ized expressions of key data elements. organization that owns or operates each premise. GS1 standards could be implemented throughout On the compliance level, the identifier may, for the food supply chain to enable traceability. GS1 is example, relate to produce from a particular section of a a not-for-profit organization dedicated to the design field in a window of several weeks. On the value level, and implementation of global standards for identify- it may include information related to more sustainable ing goods and services to improve the efficiency and or socially responsible practices on the farm, special visibility of supply chains. There are GS1 member handling, and so on. The 34 BRIDGE

FIGURE 3 Three pillars of food tracking and tracing systems.

Establishing a product ID. The product ID relates to of tracking and tracing capabilities. However, although the type and batch of agricultural good produced. On the deployment of advanced tracking and tracing tech- the compliance level, it might relate to a particular type nology can significantly improve supply chain and other and subtype of vegetable linked to a particular prem- operational efficiencies, adoption has been slow, largely ise, such as a field-packed box of organic iceberg lettuce because of the number and complexity of food and food- from a particular section of a particular farm in central component pathways and the concomitant need for California. On the value level, it might include infor- multiple levels of process alignments and information mation on the seeds used, irrigation characteristics, and sharing (including adoption of a consistent semantic more precise time stamps. model and a system for conveying relevant information Establishing a means of tracking movements and and protecting otherwise sensitive information). transformations. As food moves through the value In addition, the governance mandate for agriculture chain, it is physically transported in various types of and the food supply industries is often split among juris- containers and through various local environments dictions (e.g., states and provinces) and national bodies. (temperature, humidity, various gas concentrations); As a result, solutions must support concurrent jurisdic- it sits for varying periods of time at transition points; tional and national policies. and it is transformed in various ways (e.g., blending, Therefore, it is essential that all stakeholders be canning, freezing, preserving). Thus, maintaining the identified, that relative benefits be defined, and that integrity of the data is a challenge. On the compli- buy-in and cooperation be established. Although all ance level, it may include tracking points of origin and parties will benefit to some degree from risk mitigation entry, type and ID of transport, basic information about and reduced liability, other benefits will vary substan- transformation, and basic information about dates and tially by stakeholder. Key benefits beyond compliance times. On the value level, it may include environmen- with regulations may include increased access to global tal characterizations (especially temperature history) markets, better quality control, improved demand vis- and geospatial history. ibility and forecasting, and improved brand image and increased sales. Implementation Issues For incremental adoption, it is important that one Government regulations and the requirements of busi- define “slices” of a hypothetical comprehensive track- ness customers will encourage a minimum level (at least) ing and tracing system that cover enough space to FALL 2011 35

provide significant benefits to all stakeholders, but are technology for efficient labeling and scanning is either also contained enough to be manageable (and finance- not available or too expensive. In such cases, there may able). It is often helpful to define such slices using be little or no tracking and tracing, and retailers and a so-called “cube model” that has three dimensions: consumers may have essentially no insight into where (1) food type; (2) value-chain position; and (3) geo- their food comes from or what the safety risks are. graphical/regulatory region (e.g., state, province). However, as mobile phones and related sensing and Manageable slices can often be defined in which one communications technologies reach higher levels of dimension is fixed and the other two vary across the penetration and continue to improve in capabilities, relevant values (e.g., packaged leafy vegetables from all they may help make cost-effective, easy-to-use systems relevant regions from inputs through retail). available to rural farmers. Mobile phones and so-called Another issue relates to the classic economic “agency “social web” technology also have the potential to con- problem,” that is, that costs and benefits can accrue sub- vey important, targeted information to rural farmers, stantially differently in different areas of the value chain. directly and through social contexts, that will help In practice, costs may be borne disproportionately by them increase output levels and productivity (e.g., food producers/farmers and early processors relative to Agarwal et al., 2010; Patel et al., 2010; Veeraraghavan later processors, distributors, and retailers. et al., 2007). There are signs, however, that the business ecosystem is evolving as third parties increasingly provide Enhancing Multifactor Productivity technology and process components, such as low- “Physical meets digital” innovation can help deci- cost labeling technology, tracking-infrastructure-as- sively in many respects to increase factor productiv- a-service, and food environmental monitoring and ity directly on the farm and interactions with early response capabilities. By leveraging scale and special- food processors. A few representative examples are ization, these changes can lower adoption and man- described below. agement costs. Optimizing Water Logistics and Use Country and Regional Issues The amount of water used in agriculture is staggering Structurally, food systems are globally distributed (Table 1). For example, it is estimated that producing and integrated in a variety of ways. Different structures 1 kilogram (kg) of wheat requires 500 to 4,000 liters of reflect economic and policy characteristics associated water, and 1 kg of meat requires 5,000 to 20,000 liters with the end destination, as well as with characteris- of water (e.g., Lundqvist et al., 2008). Thus it is not tics associated with production and distribution points surprising that agriculture dominates the human use of along the way. water—estimates are in the range of 70 percent. For example, food in the United States can come If, as expected, the global middle class, which eats from farms nearly anywhere in the world through a more meat and delicacies, continues to grow rapidly variety of distribution pathways with distinct profiles for well into the current century, water requirements will risk of contamination, spoilage, and additions to cost. increase much faster than population growth. Around These pathways cross through many different countries the world today, difficulty obtaining sufficient amounts and regions, each with potentially different availabili- of clean water for irrigation is already a critical issue. ties of relevant technology and deployment capabilities. And the relative economic motivations (e.g., based on labor costs, risk tolerance) for employing necessary pro- TABLE 1 Estimated Water Consumption cesses can vary widely from country to country and even Required to Produce Some Common Foods region to region. Nevertheless, companies and policy makers can insist on certain levels of supply chain trans- Food (1 kg) Water Required to Produce (liters) parency and can select vendors who can comply. Wheat 500–4,000 The food supply for a typical, large, rapidly growing city in a developing country, however, may receive Meat 5,000–20,000 the majority of its food from relatively undeveloped Chocolate 24,000 rural farms in remote parts of the country, where the The 36 BRIDGE

Increasingly, various types of sensor and communi- utilization. Thus leading firms are moving away from cations technology are being used to provide data and “fix on failure” and scheduled-maintenance policies perform analyses necessary to increase water productiv- toward condition-based maintenance and ultimately ity and reduce irrigation requirements (e.g., Aqueel-ur- predictive maintenance approaches. Rehman et al., 2011). In dry areas (e.g., North Africa, However, these approaches are still in their infancy: western Asia), where water is sometimes a more limit- accuracy and specificity issues often lead to false posi- ing resource than land, farmers may focus more on water tives, as well as a lack of targeted information as to productivity than on yield per unit of land (Oweis and what specifically needs to be fixed/done; and predic- Hacham, 2006). tion lead times still tend to be too short to take action In such areas, deficit irrigation is becoming more to optimize maintenance and/or production schedules prevalent (Geerts and Raes, 2009). This technique based on anticipated problems with equipment. Nev- involves using measurements and modeling to design ertheless, as the volume, diversity, and quality of data and execute strategies for supplying the minimum improve, and with increasingly creative modeling and amount of exogenous water that will support stable and analytics, condition-based maintenance and predictive acceptable yields. Successful deficit irrigation requires maintenance capabilities are expected to have a major field instrumentation that provides data reflecting impact on agricultural productivity and cost efficiency soil conditions, water inputs, evapotranspiration, and in the future. spatial crop distribution. Models reflecting such factors as crop stage/type and crop water stress response Biofuels Production for Optimizing the Harvesting and can then be used to optimize output based on the Transport of Highly Perishable Crops water restrictions. The increasing use of biofuels as an alternative or It is estimated that spoilage and other waste in the complement to fossil fuels has had a significant effect on food chain cause about a 30 percent food loss. There- agriculture both in terms of changing demand and pro- fore, smarter supply chains—both in terms of logistics duction characteristics and in terms of political/social and the use of environmental monitoring and response engagement. When traditional food crops such as corn systems—can substantially increase the effective “water and sugarcane are used as feedstock, upstream biofuels productivity” of agriculture. production is similar to food production.2 Downstream processes, however, resemble more traditional industrial production processes. The need for detailed tracking and tracing of biofuels is generally lower than for other Because of spoilage and other products, but optimizing the process from planting to production can be a priority. waste in the food chain, about Ethanol production from sugarcane is an illustrative 30 percent of food is lost. example. Many factors affect the level of sugar in sugarcane (e.g., McLaren, 2009), which is a key factor in the ultimate fuel yield. In addition, because of certain natural biophysical processes, sugar levels tend to decrease Equipment Health and Fleet-Level Uptime and rapidly after the sugarcane is harvested. Thus, minimiz- Maintenance Optimization ing the harvest-to-mill time interval is a high priority. Uptime and utilization of heavy equipment in “highly Solutions to this problem include optimizing crop physical” industries, whether on the farm, at a mine site, planning and logistics. Using GIS technology and or on an oil rig, are often key factors in business perfor- various modeling and analytics approaches to balance mance. Fortunately, heavy equipment is increasingly water availability, soil characteristics, distance-to-mill, being instrumented with sensors that collect data that road and weather conditions, traffic, and mill schedules, can be used to model the health status of the equipment. producers can optimize their planting, harvesting, and Combined with other types of inspection data and envi- transport activities to maximize ethanol yields. ronmental data, these models may be improved to the point that they ensure the optimization of maintenance 2 Non-food stocks have many advantages, but this topic is beyond the of the equipment and maximization of uptime and scope of this article. FALL 2011 37

End-to-End Operational and Financial that data be shared as much as possible to accelerate the Management for Farms development of models and analytics that are accurate True end-to-end business optimization is often enough to support critical business decisions. Because impeded by siloed IT architectures and a lack of important data are generated by many firms and other timely, reliable information. Typically, different data stakeholders, policies and procedures must engender models are used for different functions. In addition, mutual trust, and security must be provided as needed. data are often of varying quality and freshness making Ecosystem thinking. The kinds of improvements it difficult to optimize processes that require trading- discussed in this paper will require that stakeholders off factors associated with different functions/silos in collaborate more closely in mutually beneficial ways the business. based on a sense of shared destiny. Relatively speaking, Nevertheless, cross-functional optimizations, such agricultural and food industry players have tended to be as aligning harvesting specifics with current market less inclined to adopt this ecosystem mindset. dynamics, can be highly beneficial for agricultural Tailoring. Approaches and technologies must be firms. Thus, farms are increasingly instituting capa- adapted for the unique needs and capabilities associ- bilities for collecting timely, consistent data about their ated with widely differing sociopolitical and economic assets and business processes and using analytics and development conditions around the world. modeling to enable end-to-end optimization and pro- Business structure and models. New vendor busi- active decision making. ness models will be needed to enable the widespread By tracking such factors as per unit costs and rev- adoption by agricultural businesses of necessary tech- enues, water usage, energy consumption, the health nologies and processes. These models should include status of equipment, soil characteristics, and crop types/ third-party vendors who can offer key capabilities at states, and then relating them to yields, metrics of mar- lower cost based on scale and specialization. Financing ket demand, and current and predicted distribution/ and/or cloud-oriented “as-a-service” models should also logistics characteristics, such optimizations are possible, be used to reduce adoption and management costs. and managers can have access to decision-support infor- References mation for balancing business objectives such as cash flow, profit, customer satisfaction, risk levels, and pre- Agarwal, S.K., K. Dhanesha, A. Jain, A. Kumar, S. Menon, N. dicted future output capacity. Rajput, K. Srivastava, and S. Srivastava. 2010. Organiza- tional, Social and Operational Implications in Delivering Conclusions/Action ICT Solutions: A Telecom Web Case-study. Paper present- Information technology will be increasingly impor- ed at ICTD2010, London, U.K., December 13–16, 2010. tant for improving agricultural business productivity Aqeel-ur-Rehman, A.Z. Abbasi., N. Islam, and Z.A. Shaikh. and cost efficiency and for providing safe food (and fuel) 2011. A review of wireless sensors and networks’ applica- for a growing and increasingly wealthy and demanding tions in agriculture. Computer Standards and Interfaces world population. “Physical meets digital” technolo- (April). doi:10.1016/j.csi.2011.03.004. gies and processes, which are well suited to deployment Geerts, S., and D. Raes. 2009. Deficit irrigation as an on-farm throughout the food value chain, will have important strategy to maximize crop water productivity in dry areas. near-term and longer term impacts. The following pri- Agricultural Water Management 96(9): 1275–1284. orities are key to making this happen. Lundqvist, J., C. de Fraiture, and D. Molden. 2008. Saving Data capture and management policies. Necessary Water: From Field to Fork—Curbing Losses and Wastage data are becoming increasingly available, but policies in the Food Chain. SIWI Policy Brief. Stockholm: SIWI. and processes to ensure that they are captured and man- Available online at http://www.siwi.org/documents/Resourc- aged electronically in a way that ensures their quality, es/Policy_Briefs/PB_From_Filed_to_Fork_2008.pdf. availability, and integration (especially in federated McLaren, J. 2009. Sugarcane as a Feedstock for Biofuels: models) have not been put in place. An Analytical White Paper. NCGA Whitepaper. Avail- Sharing and scale. Useful analytics and modeling able online at www.ncga.com/files/pdf/SugarcaneWhitePaper often require large amounts of detailed data about many 092810.pdf. different kinds of assets and processes in the context of Oweis, T., and A. Hachum. 2006. Water harvesting and actual use. It is in the interest of everyone involved supplemental irrigation for improved water productivity of The 38 BRIDGE

dry farming systems in West Asia and North Africa. Agri- Veeraraghavan, R., N. Yasodhar, and K. Toyama. 2007. cultural Water Management 80(1-3): 57–73. Warana Unwired: Replacing PCs with mobile phones in Patel, N., D. Chittamuru, A. Jain, P. Dave, and T.S. Parikh. a rural sugarcane cooperative. Presented at the Interna- 2010. Avaaj Otalo—A Field Study of an Interactive Voice tional Conference on Information and Communication Forum for Small Farmers in Rural India. Pp. 733–742 in Technologies and Development, 2007. ICTD 2007. Avail- Proceedings of ACM Conference on Human Factors in able online at http://people.ischool.berkeley.edu/~rajesh/ Computing Systems (CHI 2010). Available online at index_files/WaranaUnwired. http://www.stanford.edu/~neilp/pubs/chi2010_patel.pdf. Models informed largely by sensor-supplied data can provide insights into the role of agriculture in carbon, energy, nutrient, and water cycles.

Multiscale Sensing and Modeling Frameworks Integrating Field to Continental Scales

Indrejeet Chaubey, Keith Cherkauer, Melba Crawford, and Bernard Engel

Indrejeet Chaubey Keith Cherkauer Melba Crawford Bernard Engel

Agriculture, which has traditionally been considered a source of food, feed, and fiber, is increasingly being identified as a source of energy and ecosystem services, such as biodiversity and climate, water, and pest reg- ulation (MEA, 2005). The growing world population has significantly increased pressure on agriculture in both areas. Historically, agriculture has met societal demands by expanded irrigated and non-irrigated crop- production areas, increased automation, genetic selection and modifica- tion of plants, and better management and pest control, among other approaches. Some of these approaches provide few opportunities today for

Indrajeet Chaubey is a professor in the Departments of Agricultural and Biological Engineering and Earth and Atmospheric Sciences; Keith Cherkauer is an associate professor of agricultural and biological engineering; Melba Crawford is a professor of agronomy and civil engineering and associate dean of engineering for research; and Bernard Engel is a professor and head of the Department of Agricultural and Biological Engineering. All are at Purdue University. The 40 BRIDGE increasing agricultural production, but others are in from various sources, including sensors. As ecosystem the early stages of development and provide tremen- services provided by agriculture become more impor- dous opportunities for further expansion. tant, models and sensor-supplied data in these models Food production in greenhouses provides a glimpse that can provide insights into understanding the role of of the possibilities for meeting increasing demands on agriculture in carbon, energy, nutrient, and water cycles agriculture. Modern greenhouse production involves will also become more important. large numbers of sensors and controls, as well as infor- Thus, expanding sensing and IT will be critical to mation technologies (IT), to optimize the production of meeting the food, feed, and fiber requirements of a specialty crops, such as fruits and vegetables. Although growing global population, as well as meeting energy these specific technologies may not be directly appli- and ecosystem needs. cable to crop production on a broad scale of watersheds, In this article, we propose a multiscale framework regions, and even continents, they do provide evidence that can meet the need for in situ point-scale sensing of the value of sensing and IT in food production. For in continuous time to develop an understanding of fun- broader applications, these approaches would have to be damental physical and biological processes, spatially greatly expanded and include multiscale sensing. continuous sampling over extended areas by remote Computational models and computer-based decision- sensing technologies to represent phenomena at field, support systems are increasingly being used to assist with watershed, and regional scales, and models of processes a range of agriculturally related environmental and that link across these scales and represent associated resource conservation issues. These models rely on data dynamic processes (Figure 1).

FIGURE 1 Multi-scale framework integrating in situ and remote sensing data collection, multi-scale modeling, and process representation to develop agricultural watershed management strategies. FALL 2011 41

Field-Scale Opportunities: Real-Time Sensing the Cooperative Observer Program (COOP), which Soil and climate conditions are key components in was launched in 1890. Participants in COOP collect growing healthy and productive crops. With advances daily precipitation and air temperature measurements, in precision agriculture, farmers now know more about which are distributed through the National Climatic their soils and crop yields than they have at any time Data Center and many state climatologist offices (http:// since the mechanization and commercialization of agri- dss.ucar.edu/datasets/ds510.0/). culture began in the early 20th century (Lowenberg- Much of this information is now available to farmers DeBoer and Erickson, 2000). For example, yield maps in near-real time on the Internet, and many weather- now provide high-resolution information on inter- related sites provide real-time storm tracking and fore- annual field-scale production variability, enabling casts ranging from hourly to 10-day blocks. Many farmers to adjust application rates of seed, fertilizer, and commercial organizations have also begun to market pesticides to maximize crop yields. meteorological-observation systems, which can be pur- However, these are static sources of information gen- chased by individuals and installed locally, with direct erated after a crop has been harvested. They cannot transmission of data into the corporate system. provide immediate input about conditions that might These technologies not only provide farmers with lead to reduced productivity, nor can they be used to site-specific meteorological information, they also determine preventive solutions that could increase yield contribute to new markets for companies that use in a given season. these observations to create user-friendly agricultural decision-support products. Examples of such services In Situ Technologies include forecasts of good weather windows for plant- Opportunities and Challenges ing, spraying, or harvesting and site-specific climate information for evaluating crop rotation or pest man- Inexpensive sensor technologies and wireless com- agement strategies. munication are increasingly being used in modern agriculture to provide real-time information on soil and meteorological conditions. Soil temperature and moisture are standard measurements that are impor- First-generation biosensors, tant to crop yield. Low temperatures can slow seed germination and cause stress on young plants, thus such as the root oxygen increasing the likelihood of disease and smaller yields. Late-season planting can avoid cold soil temperatures bioavailability sensor, are but also decrease crop yields by delaying growth and leaving the laboratory and development until drier weather later in the summer. Traditionally, soil temperature has been measured at a being used in the real world. handful of sites in a state; in recent years daily values have been provided online. Water stress, induced by limited water availability, is Future Possibilities perhaps the biggest factor in reducing crop yield. Soil Biosensor technology is still in its infancy, but first- moisture measurements can provide information on generation biosensors are now leaving the laboratory plant water availability, but these data are not regularly and being used in the real world. One such technol- collected. However, a growing number of inexpensive ogy, the root oxygen bioavailability sensor (Liao et al., soil moisture and temperature sensors are available for 2004; Porterfield, 2002), mimics the consumption of use in farm fields to provide information on current oxygen by a root, and therefore the transport of water conditions, and even to coordinate with irrigation and nutrients into the plant. Oxygen consumption is as scheduling systems (Oshaughnessy and Evett, 2008; important to crop yield as soil moisture because the rate www.agmoisture.com). of consumption is sensitive to both dry and overly wet Measurement of local weather conditions, especially (oxygen-limited) conditions (Drew and Stoltzy, 1996). precipitation and air temperature, as well as wind speed, Biosensors have also been developed that are sen- humidity, air pressure, and solar radiation, has been the sitive to nutrients and contaminants of interest in responsibility of the National Weather Service through The 42 BRIDGE agriculture, including nitrate and phosphate (McLamore for commercially available sensor systems (e.g., http:// et al., 2009). These sensors are currently limited in www.campbellsci.com/communications). However, there operational applicability by their short lifespan (days to are still significant difficulties in initial installation of months) and because they are designed for use in liquid many of these technologies. media. However, as scientists merge their knowledge of biosensors with modern manufacturing technologies, Advancing the State of the Art for such sensors will surely make their way into agricultural Production Agriculture applications, including monitoring the distribution of Self-organizing wireless hardware and software for nitrogen and phosphorous across fields and informing sensor networks have recently become commercially next-generation farm equipment about where more fer- available (Martinez et al., 2004), making the devel- tilizer should be applied to maximize production. opment and deployment of sensor networks possible. Measurement of carbon storage and fluxes will be Although these networks have great potential for agri- important for agricultural systems to offset emissions cultural applications, integrating communications, data in other parts of the economy. Current fluxes of car- acquisition, and sensor technologies into operational bon from soils are poorly understood, primarily because nodes in a stable, functional sensor network requires making accurate measurements is difficult and costly; significant effort and knowledge and will require over- it requires proper installation of chambers at the soil coming major challenges to their deployment. surface to capture fluxes or micrometeorological sta- Agricultural sensor networks must be robust enough tions taking rapid measurements to estimate fluxes just to survive environmental extremes including rain, above the soil surface. To be useful for climate-change sun, and flooded fields. They must also be minimally adaptation and mitigation plans, however, newer, more intrusive, so they do not hinder access to the field by cost-effective systems that require less operational over- heavy machinery, and they should be designed to sur- sight will be necessary for quantifying carbon fluxes and vive inadvertent run-ins with such machinery. In addi- eventually for monitoring carbon sequestration. tion, sensor nodes must be designed for easy installation and removal, so nonfunctioning nodes can be easily replaced and all nodes can be removed prior to tillage. Once these technological problems are resolved, Fluxes of carbon from soils probably in the coming decade, self-organizing networks could be integrated with inexpensive sensor technolo- are poorly understood, gies, making it possible for farmers to deploy networks primarily because making rapidly across their fields and giving them immediate access to spatial and temporal information about field accurate measurements is conditions. In addition, it should be possible for sensors to communicate directly with farm equipment and difficult and costly. personal electronic devices, such as smart phones. This would give farmers in the field access to both real-time and archival information about field con- Communications Technologies ditions as they survey or work in their fields. Smart Exploiting Multipurpose Communications Systems software or links to service providers could provide addi- The integration of new and existing sensor technolo- tional input about potential problem sites, potential gies with wireless communication systems has already ways to resolve those problems, and even suggestions begun, and many companies now offer wireless meteo- for where to find materials or equipment or bids from rological and soil moisture sensor systems. With these other service providers. tools, farmers can check the status of their fields before Remote Sensing: leaving the house. Technologies for Scaling Information Connections to cellular phone networks, satellite uplinks, long-distance radio frequency communica- Current Capabilities for Production Agriculture tions, and direct Internet or WiFi connections have “Remote sensing” refers to technologies that can all become relatively common and are often options make measurements without being in direct contact FALL 2011 43

with the target of interest. Often characterized not only Future Contributions to Agriculture by the acquisition of a measurement, but also by the spa- Imaging technologies can now simultaneously acquire tial and temporal resolution of data, information based hundreds of measurements in narrow windows (bands) on remote sensing provides capabilities for scaling in of the electromagnetic spectrum. New active sensing situ measurements and understanding physical and bio- technologies that emit energy in specific wavelengths logical processes in systems that operate on watershed, and measure associated responses are also becoming regional, continental, and global scales. These tech- commonplace. Three technologies that are evolving to nologies extend spatial coverage, provide information operational level on airborne and space-based platforms about inaccessible areas, and acquire unique measure- are particularly promising for advancing the science of ments via a variety of sensing modalities. agriculture: (1) hyperspectral imaging; (2) active and Remote sensing technologies on tractors and com- passive microwave systems; and (3) laser-based systems. bines and on airborne and space-based platforms are now integral to modern agriculture, with applications ranging from subsistence farms to large-scale mechanized production farms and high-value specialty crops. Remotely Soil moisture, a critical factor sensed precipitation, temperature, and wind speed are incorporated into weather predictions; Global Position- in crop condition, irrigation ing System (GPS) data are assimilated into guidance strategy, and predicting crop systems for field operations; and advanced sensors are incorporated into specialty equipment for use in preci- yield, cannot be measured sion agriculture. Imaging technologies most commonly used in pro- directly and is difficult to derive duction agricultural applications flown on satellites, from remote-sensing data. airplanes, and unmanned vehicles record reflected light or energy from land/water surfaces. Multispec- tral sensors, such as the advanced very-high-resolution Coupled with corresponding advances in agricultural radiometer (AVHRR) sensors on NOAA satellites, modeling and data-delivery systems, these technologies have provided daily global coverage at kilometer scale can revolutionize the characterization and prediction of for decades, resulting in widespread use of vegetation agricultural processes on multiple scales. Like in situ indices as indictors of crop vigor (Lillesand et al., sensors, these new remote sensing technologies provide 2007). In fact, these simple indices, which provide opportunities for entrepreneurs to develop and deliver free data globally, are the most widely used inputs for products to farmers. models of agricultural yield and for empirical indica- Hyperspectral imaging sensors, which mimic labora- tors of crop health. tory spectrometers that provide chemistry-based infor- More advanced products derived from the next mation related to reflectance and absorption, provide generation of these sensors, such as NASA MODIS two-dimensional images in hundreds of narrow bands. (modis-land.gsfc.nasa.gov/), are slowly being incorpo- These data can advance the science of agriculture in rated into models used in agricultural research and physically based and empirical models to estimate chlo- applications. Data products derived from the Landsat rophyll content (Haboudane et al., 2008) and nitro- series of missions (landsat.usgs.gov) have also signifi- gen content (Chen et al., 2010), monitor water stress, cantly improved a wide range of agricultural applica- detect invasive species, evaluate water quality, and so tions, although the 16-day repeat cycle is a limiting on (Thenkabail et al., 2000). Commercially flown air- factor, particularly in cloud-covered regions. The borne hyperspectral sensors monitor high-value agri- availability of products based on multispectral data cultural crops, and hyperspectral satellite missions are acquired by both space-based and airborne platforms being developed by the international community for has increased dramatically in the past decade, as satel- launch in the next decade. lites launched by both governments and the private Remote sensing-based soil moisture products are sector provide capability for timely products to support primarily based on measurements in the microwave agricultural applications. region of the spectrum, where dielectric properties of The 44 BRIDGE soils under different conditions can be related to soil application), crop production in high-relief areas, and moisture. Although soil moisture cannot be measured soil mapping and management. Information related to directly and is difficult to derive from remote sensing the vertical structure of vegetation (e.g., height, den- data, it is one of the highest priority measurements sity) can also be derived from LIDAR, providing a non- for agricultural remote sensing. In addition to being intrusive alternative to traditional destructive sampling a critical factor in crop condition, soil moisture is an (Dubayah et al., 2010; Selbeck et al., 2010). important parameter for characterizing atmospheric and These three advances in sensor technologies provide land-surface interactions and plays a role in regional clear evidence of the potential of new data sources to and global weather patterns. Soil moisture is also of support research and applications in agriculture in both paramount importance in developing agricultural man- the developed and developing worlds. Coupled with agement strategies (e.g., irrigation) and predicting crop decision-support models and rapidly evolving commu- yield, as well as detecting and monitoring drought. nication technologies, these new sources of data have Current operational space-based soil-moisture prod- much to contribute to next-generation agriculture. ucts have low resolution (tens of km), which is useful for regional and global applications but of limited use Closing the Gaps with Models for watershed-scale applications (Jackson et al., 2010). Enabled by the widespread availability of computa- However, the upcoming launch of the NASA Soil tional resources, significant advances have been made Moisture Active/Passive (SMAP) mission (smap.jpl. in agricultural models that can link and integrate across nasa.gov/Imperative) is widely anticipated by the agri- scales and processes. However, major challenges will cultural community because it is expected to provide have to be overcome before we can take advantage of advanced soil moisture products at resolutions that will the power of simulation modeling to evaluate agricul- be useful for watershed management. tural production, its impact on environment and ecosystem services, and the development of sustainable management strategies. A few of these challenges are Agricultural management described briefly below. decisions are made at field Integration of Models at Different Spatial and Temporal Scales scales, but policy decisions In situ observations have long been used for the development, calibration, and evaluation of models— apply on regional, national, from field-scale models such as DRAINMOD (Skaggs or even international scales. et al., 1995), which simulates water and nutrient move- ment through subsurface drainage, to the WEPP model (Laflen et al., 1991), which simulates soil erosion from Laser-based systems, such as LIDAR (light-detection hillsides, to the soil and water assessment tool (SWAT) and ranging) systems, emit laser pulses of given wave- (Arnold et al., 1998), one of the most widely used lengths and detect energy that is intercepted and scat- catchment-scale models that can evaluate impacts of tered back to the sensor. GPS-derived trajectories and agricultural management decisions on crop production, the platform motion are combined with the time to hydrology, and water quality (swatmodel.tamu.edu). interception of the backscattered energy to derive high- Agricultural management decisions are made at field resolution three-dimensional presentations of the land- scales, but policy decisions apply on regional, national, scape (lidar.cr.usgs.gov/). or international scales. Currently, simulation models for Airborne LIDAR products, which are widely used analyzing the impacts of agriculture on the field to river- for floodplain, bathymetric, and urban mapping, are basin scale work in isolation from models that work on being investigated specifically for agricultural applica- global scales. To enable agricultural production evalu- tions. LIDAR provides the most accurate remote sens- ations in a single modeling environment, these models ing-based estimates of topography (and therefore slope will have to be integrated. and sun exposure), which are relevant to agricultural Remote sensing and sensor networks are two tech- management decisions related to runoff (e.g., fertilizer nologies that are helping to bridge the gap by providing FALL 2011 45

spatial observations that can be used to scale processes modeling, and advanced applications in the agricultural from the field and watershed to larger scales. For exam- community. Coupled with the increasingly interdisci- ple, large-scale variability in soil moisture is controlled plinary nature of agricultural education, research, and largely by general conditions (e.g., when it last rained commercial activities, the future looks promising for the and vegetation type), whereas small-scale variability development of new capabilities for meeting both grow- depends much more on local conditions (e.g., Crow and ing demands for food, feed, and fiber, and for diverse eco- Wood, 1999). system services in a rapidly evolving world.

Process Representation to Evaluate Competing Demands References Evaluating competing demands for services from Arnold, J.G., R. Srinivasan, R.S. Muttiah, and J.R. Williams. agricultural lands (e.g., biomass production for food 1998. Large area hydrologic modeling and assessment–Part and fuel, water-quality improvement, minimum-flow 1: Model development. Journal of the American Water requirements to meet the needs of ecosystems, etc.) will Resources Association 34(1): 73–89. require an integrated modeling framework that includes Chen, P., D. Haboudae, N. Tremblay, J. Wang, P. Vigneault, ecosystem services, production of food and fuel, and the and B. Li. 2010. New spectral indicator assessing the use of water to support agricultural production and eco- efficiency of crop nitrogen treatment in corn and wheat. system demands. The development of such models and Remote Sensing of Environment 114: 1987–1997. frameworks will be essential for sustainable agricultural Crow, W.T., and E.F. Wood. 1999. Multi-scale dynamics of production in the future. soil moisture variability observed during SGP’97. Geo- physical Research Letters 26: 3485–3488. Sensors and Networks for Monitoring Water Quality Drew, M.C., and L.H. Stoltzy. 1996. Growth under Oxy- Monitoring water quality is very expensive and time gen Stress. Pp. 845–858 in Plant Roots: The Hidden Half, consuming. This is the primary reason for the limited edited by Y. Waisel, A. Eshel, T. Beeckman, and U. Kafkafi. availability of global water-quality data products. In New York: Marcel Decker. addition, in situ and remote sensing technologies for Dubayah, R.O., S.L. Sheldon, D.B. Clark, M.A. Hofton, J.B. monitoring water quality can only evaluate a limited Blair, G.C. Hurtt, and R.L. Chazdon. 2010. Estimation of number of indicators. For the next generation of water- tropical forest height and biomass dynamics using LiDAR quality models, we will need sensors that can easily and remote sensing at La Selva, Costa Rica. Journal of Geo- inexpensively monitor parameters such as nitrogen physical Research 115(3): 1–17. and phosphorus concentrations, pathogens, and sedi- Haboudane, D., N. Tremblay, J.R. Miller, and P. Vin ment in real or near-real time. Vigneault. 2008. Remote estimation of crop chlorophyll content using spectral indices derived from hyperspectral New Applications for Modern Communication Devices data. IEEE Transactions on Geoscience and Remote Sens- Applications to support daily decisions (e.g., the ing 46: 423–437. location of the nearest gas station) on mobile com- Jackson, T.J., M.H. Cosh, R. Bindlish, P.J. Starks, D.D. Bosch, munication devices have increased dramatically in the M. Seyfried, D.C. Goodrich, M.S. Moran, and J. Du. 2010. last few years, and the agricultural community also has Validation of Advanced Microwave Scanning Radiometer access to some real-time data (e.g., crop yield). How- soil moisture products. IEEE Transactions on Geoscience ever, we need applications that can be used to evaluate and Remote Sensing 48: 4256–4272. the impacts of agricultural production on hydrology, Joseph, A.T., R.van der Velde, P.E. O’Neill, R.H. Lang, water quality, and ecosystem services. The develop- and T. Gish. 2008. Soil moisture retrieval during a ment of such applications will facilitate education and corn growth cycle using L-Band (1.6 GHz) radar obser- decision making for sustainable agricultural production vations. IEEE Transactions on Geoscience and Remote and environmental quality. Sensing 46: 2365–2374. Laflen, J.M., L.J. Lane, and G.R. Foster. 1991. WEPP: A Looking to the Future new generation of erosion prediction technology. Journal Although the hurdles described above will be difficult to of Soil and Water Conservation 46(1): 34–38. surmount, attention is now focused on advanced sensing, Liao, J., G. Liu, O. Monje, G.W. Stutte, and D.M. Porterfield. data storage and retrieval, communication capabilities, 2004. Induction of hypoxic root metabolism results from The 46 BRIDGE

physical limitations in O2 bioavailability in microgravity. for automatic irrigation scheduling. Paper No. 083452. Advances in Space Research 34(7): 1579–1584. Proceedings of the American Society of Agricultural and Lillesand, T.M., R.W. Kiefer, and J. Chipman. 2007. Remote Biological Engineers International (ASABE), Providence, Sensing and Image Interpretation, 6th ed. New York, N.Y.: Rhode Island, June 29–July 2, 2008. John Wiley and Sons. Porterfield, D.M. 2002. Use of microsensors for studying the Lowenberg-DeBoer, J., and K. Erickson, eds. 2000. Preci- physiological activity of plant roots. Pp. 503–527 in Plant sion Farming Profitability. West Lafayette, Ind.: Purdue Roots: The Hidden Half, edited by Y. Waisel, A. Eshel, T. University Press. Beeckman, and U. Kafkafi. New York: Marcel Decker. Martinez, K., J.K. Hart, and R. Ong. 2004. Environmental Selbeck, J., V. Dworak, and D. Ehlert. 2010. Testing a vehicle sensor networks. IEEE Computer 37(8): 50–56. based scanning lidar sensor for crop detection. Canadian McLamore, E.R., D.M. Porterfield, and M.K. Banks. 2009. Journal of Remote Sensing 36: 25–34. Non-invasive self-referencing electrochemical sensors for Skaggs, R.W., M.A. Breve, A.T. Mohammad, J.E. Parsons, and quantifying real time biophysical flux in biofilms. Biotech- J.W. Gillliam. 1995. Simulation of drainage water quality nology and Bioengineering 102: 791–799. with DRAINMOD. Irrigation and Drainage Systems 9(3): MEA (Millennium Environment Assessment). 2005. Eco- 257–277. doi:10.1007/BF00880867. system and Human Well-Being: Synthesis. Washington, Thenkabail, P.S., R.B. Smith, and E. DePauw. 2000. D.C.: Island Press. Available online at http://www.millen- Hyperspectral vegetation indices and their relation- niumassessment.org/en/Products.aspx?. ships with agricultural characteristics. Remote Sens- Oshaughnessy, S.A., and S.R. Evett. 2008. Integration of ing of Environment 71: 158–182. wireless sensor networks into moving irrigation systems FALL 2011 47

NAE News and Notes NAE Newsmakers

Fourteen NAE members and one L. Gary Leal, Warren B. and recognition of outstanding accom- foreign associate have been elected Katharine S. Schlinger Distin- plishments in the field of aviation. to the American Academy of Arts guished Professor of Chemical The award will be presented on and Sciences. Induction of the new Engineering, University of Cali- October 21, 2011, at the club’s 69th class is scheduled for October 1, fornia, Santa Barbara annual dinner-dance at the Waldorf 2011, at the academy’s headquarters Chad A. Mirkin, director, Astoria Hotel in New York City. in Cambridge, Massachusetts. International Institute for Nano- John Werner Cahn, Department technology and Rathmann Profes- of Physics, University of Washing- Paul G. Allen, chairman, Vulcan sor of Chemistry, Northwestern Inc. ton, was awarded the 2011 Kyoto University Prize by the Inamori Foundation Frances H. Arnold, Dick and Alan R. Mulally, president and for his “outstanding contribution to Barbara Dickinson Professor of CEO, Ford Motor Company alloy materials engineering by the Chemical Engineering, Bioengi- establishment of spinodal decompo- neering and Biochemistry, Divi- Shree K. Nayar, T.C. Chang sition theory.” The highest private sion of Chemistry and Chemical Professor of Computer Science, award given in Japan, the Kyoto Engineering, California Institute Columbia University of Technology Prize is bestowed on an individual Patricia G. Selinger, retired vice whose achievements have had global Wanda M. Austin, president and president, Data Management, reach and have contributed to the chief executive officer, The Aero- Architecture and Technology, betterment of society. The prize space Corporation IBM Silicon Valley Laboratory includes a diploma, the Kyoto Prize Marsha J. Berger, professor, Cou- Howard A. Stone, professor, medal, and a cash gift of $625,000. rant Institute, New York University Department of Mechanical and Frans Kaashoek, professor, Aerospace Engineering, Princeton Edmund M. Clarke, FORE Sys- Computer Science and Artificial University tems University Professor of Com- Intelligence Laboratory, Massachu- puter Science, Carnegie Mellon NAE Foreign Associate Raghu- setts Institute of Technology, has University nath A. Mashelkar, CSIR Bhat- been chosen to receive the 2010 nagar Fellow, National Chemical ACM (Association for Computing Glenn H. Fredrickson, professor Laboratory, India Machinery)-Infosys Foundation of chemical engineering and mate- Award in the Computing Sci- rials and director of Mitsubishi Isamu Akasaki, professor, Meijo ences for his contributions to the Chemical Center for Advanced University, received the IEEE 2011 structuring, robustness, scalability, Materials, University of Califor- Edison Medal. Professor Akasaki and security of software systems nia, Santa Barbara was honored for his “pioneering underlying many applications. Dr. , Ransburg Dis- contributions to the development of Leah H. Jamieson Kaashoek’s use of information-flow tinguished Professor of Electrical nitride-based semiconductor mate- control techniques to address a and Computer Engineering and rials and optoelectronic devices, major security challenge in widely John A. Edwardson Dean of Engi- including visible wavelength LEDs used commercial systems has led neering, Purdue University and lasers.” to efficient, portable, and highly Norman R. Augustine, retired Michael I. Jordan, Pehong Chen distributed applications of soft- chairman and CEO, Lockheed Distinguished Professor, Univer- ware systems and wider use of sity of California, Berkeley Martin Corporation, will be hon- portable embedded and distrib- ored by the Wings Club with the Linda P.B. Katehi, chancellor, uted systems. The prize includes a 2011 Distinguished Achieve- University of California, Davis $150,000 award. ment Award, which is given in The 48 BRIDGE

The Association for Comput- IEEE International Conference on Machinery presented Gurindar ing Machinery (ACM) has named Communications in Kyoto, Japan. Sohi, John P. Morgridge Profes- Takeo Kanade, U.A. and Helen Priyaranjan Prasad, retired tech- sor and E. David Cronon Professor Whitaker University Professor of nical fellow, Safety Research and of Computer Sciences, University Computer Science and Robotics, Development, Ford Research Labo- of Wisconsin-Madison, with the Carnegie Mellon University, the ratory, received the SAE Internation- Eckert-Mauchly Award on June 7 winner of the 2010 ACM/AAAI al Arnold W. Siegel International in San Jose, California. Dr. Sohi Allen Newell Award for contribu- Transportation Safety Award during was honored “for pioneering widely tions to research in computer vision the SAE 2011 World Congress. The used micro-architectural techniques and robotics. The Newell Award winner of the award is chosen for for instruction-level parallelism.” is given in honor of career contri- his or her outstanding international Rao R. Tummala, Joseph M. Pet- butions that have contributed to research, innovation, and contri- tit Chair in Electronics Packaging computer science as a whole or that butions to crash injury protection, and director of NSF-ERC in SOP bridge computer science and other crash injury biomechanics, and crash Technology, Georgia Institute of disciplines. The award includes a injury design for all kinds of vehicles. Technology, is the recipient of the $10,000 prize and is supported by John Rogers, Lee J. Flory-Founder 2011 IEEE Field Award in Com- the Association for the Advance- Chair in Engineering, University ponents, Packaging and Manu- ment of Artificial Intelligence of Illinois, has been awarded the facturing Technologies, which is (AAAI) and individual contribu- Lemelson-MIT Prize. Rogers’ sponsored by the IEEE Compo- tors. Dr. Kanade received the award research has resulted in the creation nents, Packaging and Manufac- on June 4 at the ACM Awards Ban- of revolutionary products integral turing Technology Society. Dr. quet in San Jose, California. to human health, fiber optics, semi- Tummala was honored for pioneer- Alfred E. Mann, USC Board of conductor manufacturing, and solar ing and innovative contributions to Trustees, USC-AMI Chairman, and power. The prize is bestowed on package integration research, cross- chairman and CEO, MannKind an outstanding mid-career inventor disciplinary education, and glo- Corporation, was awarded the Med- dedicated to improving our world balization of electronic packaging. ical Design Excellence Awards through technological invention The award was presented on June 2 (MDEA) Lifetime Achievement and innovation. The prize includes at the IEEE Electronic Components Award presented by UBM Canon. a $500,000 cash award. and Technology Conference in Lake The award is given to an individual Bruce E. Rittmann, Regents’ Pro- Buena Vista, Florida. whose contributions during a long fessor of Environmental Engineer- Amnon Yariv, Martin and Eileen career have had a demonstrable ing and director, Swette Center for Summerfield Professor of Applied impact on technological, business, Environmental Biotechnology, Ari- Physics and professor of electrical and cultural advancements in medi- zona State University, was selected engineering, California Institute of cal devices. to receive the Environmental Engi- Technology, was named the recipi- IEEE has awarded H. Vincent neering Excellence Award from ent of the 2011 IEEE Photonics Poor, dean of engineering and the American Association of Envi- Award. The award, sponsored by applied science and Michael Henry ronmental Engineers. The organi- IEEE Photonics Society, was pre- Strater University Professor, Prince- zation presented the award to Dr. sented to Dr. Yariv on May 4 at the ton University, the 2011 IEEE Eric Rittmann on May 4 at the National Conference on Lasers and Electro- E. Sumner Award. Sponsored by Press Club in Washington, D.C. Dr. Optics in Baltimore, Maryland, Alcatel-Lucent Bell Labs, the award Rittman was honored for creating for fundamental contributions to was given in recognition of Dr. Poor’s new technology that removes dan- photonics science, engineering, pioneering contributions to multiple gerous contaminants from water. and education that have broadly access communications. Dr. Poor The IEEE Computer Society impacted quantum electronics and received the award on June 7 at the and Association for Computing light-wave communications. FALL 2011 49

2011 Japan-America Frontiers of Engineering Symposium Held in Osaka, Japan

on the first afternoon that provided an opportunity for each participant to describe his or her technical work or research, a walking tour of major attractions in Osaka, such as Osaka Castle, and a river cruise with a “Rakugo” storytelling guide. The symposium and organizing committee were co-chaired by Katharine Frase, vice president of industry solutions and emerging business at IBM, and Ichiro Kanaya, associate professor in the Gradu- ate School of Engineering at Osaka University. The next JAFOE symposium will be held in late fall 2012 in the United States. Funding for the symposium was provided by the Japan Science and Technology Agency, The Grainger Foundation, the Precise Measure- ment Technology Promotion Foun- dation, The Watanabe Memorial Foundation for the Advancement of JAFOE participants discuss their research during the poster session. Technology, and the National Sci- ence Foundation. On June 6–8, 2011, the tenth ment, the Smart Grid, bio-inspired The goals of FOE symposia, Japan-America Frontiers of Engi- materials, and robotics. Presenta- which bring together outstanding neering (JAFOE) Symposium tions on cutting-edge research in mid-career engineers (ages 30 to was held at the RIHGA Royal each subject area by two Japanese 45) from industry, academia, and Nakanoshima Center Building in and two Americans included: large- government to learn about develop- Osaka, Japan. The event, which was scale spatial simulation design and ments and approaches in a variety of organized by NAE and the Engineer- analysis; smart grid cyber-security; areas of research, are to support and ing Academy of Japan, was originally construction of advanced materials encourage interdisciplinary explora- scheduled to be held in Tsukuba. from injectable gels to nanoparticle tion and to facilitate contacts and However, after the March earth- arrays; and human-like assembly collaboration among the next gen- quake, tsunami, and Fukushima Dai- robots in factories. eration of engineering leaders. ichi nuclear disasters, the venue was On the first evening, Dr. Masato For more information about changed to the Kansai area, which Sagawa, president of Intermetal- a symposium series, visit www. was not subject to risks associated lics Co. Inc., gave an informative naefrontiers.org. To nominate an with the events in March. and welcoming dinner speech. Dr. outstanding engineer to participate Sixty up-and-coming engineer- Sagawa had led research on the in a future Frontiers meeting, con- ing leaders—30 from each coun- development of the neodymium tact Janet Hunziker at the NAE try—attended this very productive magnet, the strongest type of perma- Program Office at (202) 334-1571 meeting. The four sessions for this nent magnet. Other highlights of or [email protected]. year were on massive data manage- the symposium were a poster session The 50 BRIDGE

Second China-America Frontiers of Engineering Symposium

The second China-America and enabling new applications. The systems achieved by Carnegie Mel- Frontiers of Engineering (CAFOE) topics included (1) wireless circuit lon’s Tartan Racing, which won the Symposium, hosted by Qualcomm and silicon evolution and the latest DARPA Urban Challenge; (2) Edi- at its headquarters in San Diego, research on exploiting higher radio son2’s design and development of was held on March 28–30. The frequencies to achieve very high the Very Light Car, which gets more Chinese Academy of Engineering speed, point-to-point communica- than 100 miles per gallon equiva- was a partner with NAE in organiz- tions; (2) cognitive radio; (3) inte- lent (MPGe) and won the Automo- ing the event, which was supported grated systems design and recent tive X Prize; (3) a new approach to by The Grainger Foundation. NAE approaches to leveraging body-area designing forward collision-warning member Zhigang Suo, Allen E. and networks and processing sensor data systems that would improve vehi- Marilyn M. Puckett Professor of for preventive health applications; cle safety; and (4) a new routing Mechanics and Materials at Har- and (4) efforts to improve power protocol called Ubiquitous Query vard University, was U.S. co-chair. efficiency across networks to meet for Travel Information (UQTI) to Zhihua Zhong, president of Hunan users’ needs for higher data rates. facilitate access to information and University, was Chinese co-chair. Presentations in the third ses- improve mobility. Consistent with other bilat- sion, on bio-inspired engineer- A poster session on the first after- eral FOE symposia, this meeting ing, highlighted intricate natural noon was both an icebreaker and an brought together approximately 60 design principles and precision opportunity for participants to share engineers, ages 30 to 45, from U.S. engineering that are inspiring new information about their research and Chinese universities, compa- engineering applications from the and technical work. That evening, nies, and government laboratories molecular level to the tissue level. a dinner address was given by Irwin for a 2-1/2–day meeting to learn The first speaker explained the M. Jacobs, NAE chair and co- about leading-edge developments principles of molecular assembly founder and director of Qualcomm, in four fields of engineering: ocean and described how they are being on the history of Qualcomm, includ- engineering, wireless communica- used to assemble synthetic materi- ing the development of CDMA and tion, bio-inspired engineering, and als. This next talk was on molecu- its contributions to telecommunica- advanced vehicles and mobility. lar “reprogramming” approaches tions and areas for continuing inno- The ocean engineering session to regulating the behavior of stem vation. On the second afternoon, was focused on ocean resource cells. The third presentation was participants visited: the Qualcomm exploitation and hydrodynamics. on “lab-on-a-chip” micro-devices Museum which brought to life the Two of the talks were on the recent that integrate living cells as intelli- company’s history and impact; the Deepwater Horizon disaster, par- gent components. The final speaker Qualcomm Research Center, where ticularly technology drivers for described tissue-engineering meth- new technologies are being devel- deepwater drilling. The other two ods that mimic natural tissue and oped; and the manufacturing floor, speakers described their work on organ development processes. where prototypes for in-house use experimental and computational The emphasis in the last ses- are produced. hydrodynamics, which are essen- sion, on advanced vehicles and The next CAFOE Symposium tial tools for working in the harsh mobility, was on green and intel- will be held in China in March marine environment. ligent vehicle technologies being 2013. For more information about The four speakers in the ses- developed in response to shrinking a symposium series or to nominate sion on wireless communication supplies of fossil fuels and increas- an outstanding engineer to partici- highlighted recent advances and ingly complex and congested pate in a future Frontiers meeting, research approaches to improving urban driving environments. Top- contact Janet Hunziker at the NAE the capabilities and efficiency of ics included (1) technical break- Program Office at (202) 334-1571 wireless communication systems throughs for autonomous driving or by e-mail at [email protected]. FALL 2011 51

Penn Engineering and NAE Explore “Engineered Networks” at NAE Regional Meeting and Symposium

advances that have enabled a new generation of small aerial vehicles and explained the principles and methodologies used to create small, agile micro vehicles capable of nav- igating complex indoor environments. He also described the theory and framework for using cooperating vehicles to transport large payloads, perform some kinds of construction tasks, and so on. The presentation by Daniel E. Koditschek, Alfred Fitler Moore Professor and chair of the Depart- ment of Electrical and Systems Engineering at UPenn, “Interior Eduardo D. Glandt, dean, School of Engineering and NAE President Charles M. Vest addressing the audience. Applied Science, welcoming meeting participants. Networks of Control and Coor- dination,” included a review of a NAE and Penn Engineering and NAE members to the event. long-term research agenda seek- hosted a regional meeting and NAE President Charles M. Vest ing to confer dynamical dexterity symposium on April 26, 2011, at then delivered his opening remarks on increasingly utilitarian robots, the University of Pennsylvania and provided an overview of NAE with a focus on the key problem of (UPenn) to investigate the surge of activities. limb coordination. He described research interest in the science and the conceptual interplay between engineering of complex, networked Session I: Robotic Networks robotics and neuromechanics, as dynamic systems. Advances in net- In the first session, three speakers well as between applied mathemati- centric technology and autonomous detailed recent advances and future cal theory and empirical engineer- systems promise unprecedented lev- directions in collective robotics. ing design. els of performance, robustness, and The session was introduced by Kos- The third presentation was by efficiency. The symposium, Engi- tas Daniilidis, professor of computer keynote speaker Naomi Leon- neered Networks, focused on two and information science and direc- ard, Edwin S. Wilsey Professor areas in which network science and tor, General Robotics, Automation, of Mechanical and Aerospace networked systems have led to spec- Sensing and Perception (GRASP) Engineering at Princeton Univer- tacular advances: (1) the study of Laboratory at UPenn, who sum- sity. Her talk, “Collective Motion collective behavior in networked, marized the history of coordination and Ocean Sampling Networks,” multi-robot systems and (2) the among robotic vehicles and coop- focused on the design and control analysis of aggregation and strategic eration between manipulators, as of optimum trajectories for self- interaction in social and economic well as work inspired by analogous directed underwater gliders. She networks. swarms of organisms in biology. described the significant potential Eduardo D. Glandt, dean of The first talk, “Networks of Aer- social impact of her research, the the School of Engineering and ial Robots,” was by Vijay Kumar, underlying theory, and the daunt- Applied Science at UPenn, NAE UPS Professor of Mechanical Engi- ing technical problems she faces member, and chair of the sympo- neering and Applied Mechanics and in designing control strategies and sium, welcomed more than 100 deputy dean for education at Penn implementing mobile sensor net- students, faculty, industry partners, Engineering. He described recent works comprising an autonomous The 52 BRIDGE ocean observing and prediction sys- tions, he explained, are the founda- and Engineering at Stanford Uni- tem. She also reviewed the practical tion for a new network science. versity, turned his attention to experience her group had accrued The first speaker in this session, issues related to “Internet Monetiza- with a fleet of autonomous vehicles Ali Jadbabaie, Skirkanich Asso- tion.” Saberi explained how recent that tracked optimally planned ciate Professor of Innovation in advances in algorithmic game theory paths over a period of many days in the Department of Electrical and and mechanism design have led to support of studies of Monterey Bay Systems Engineering, focused on the development of novel algorithms by environmental scientists. information aggregation and social for computing equilibria in online learning. In his talk, “Aggregation in and sponsored search auctions. He Session II: Social and Complex Networks,” Dr. Jadbabaie then presented a novel approach to Economic Networks presented and then analyzed a non- algorithmic game theory and mecha- The second session was intro- Bayesian model of information nism design. duced by George Pappas, Joseph aggregation and social learning that Dean Glandt brought the sym- Moore Professor of Electrical and shows how individuals in a social posium to a close with some brief Systems Engineering and deputy network might reach consensus and remarks and invited all participants dean for research at Penn Engineer- rationally aggregate information at to continue the lively discourse at ing. He highlighted emerging intel- the same time. the reception. For more informa- lectual connections between the The final speaker—and second tion about Penn Engineering and fields of networked robots and the keynote speaker—Amin Saberi, the symposium, visit www.seas. networked economy. These connec- Department of Management Science upenn.edu.

New Study on Integrated STEM Education

In collaboration with the National acquire knowledge and skills in each and college-readiness skills, and Research Council (NRC) Board of the STEM disciplines. would increase the number of stu- on Science Education, NAE has One aspect of STEM education dents who might consider careers in launched a new study to explore that has received relatively little STEM-related fields. the potential and challenges of attention is how and to what degree The goal of the 36-month NAE/ integrated K–12 STEM (science, the four subject areas are, or might NRC study is to develop a research technology, engineering, math- be, integrated in primary and sec- agenda for determining the value— ematics) education. Over a period ondary curricula and the potential in terms of student achievement, of many years, considerable invest- impact of integration on learn- motivation, career aspirations, and ments have been made in research ing. Connections among STEM other factors—of integrated K–12 to improve K–12 STEM teaching subjects are commonplace in the STEM education in the United and learning. Most of this research world we live in but are largely States. Margaret Honey, president has focused on single subjects in the absent in K–12 classrooms. Advo- and CEO of the New York Hall STEM quartet, most often science cates of STEM integration argue of Science, is chair of the 15-per- or mathematics. This orientation that STEM subjects would be more son study committee. The project is important, not only because it is relevant to students and teach- is funded by the S.D. Bechtel, Jr. consistent with the way education ers, would enhance motivation for Foundation, the National Science is delivered in this country, but also learning, would improve student Foundation, the Samueli Founda- because of the complexity and lack achievement, would address calls tion, and PTC, Inc. of understanding of how students for students with better workplace FALL 2011 53

EngineerGirl! Essay Contest Winners

Second place winners were: Sophia Giovanis, Grade 2, Long Lake, Minnesota; Haylie Thomas, Grade 8, Littleton, Colorado; Leigh- ton Suen, Grade 12, Staten Island, New York. Third-place winners included: Michelle Zhu, Grade 3, Cupertino, California; Olivia Rey nolds, Grade 7, Baltimore, Mary- land; and Megan Pence, Grade 12, Melissa Meng Ming Li Wu Samantha Harris Ponte Vedra, Florida. Winners of Honorable Mention included: The National Academy of First place for grades 3 through 5 Anna Eichelberger, Grade 4, Long Engineering EngineerGirl! (www. was awarded to Melissa Meng, a Lake, Minnesota; Doina Ghegeliu, engineergirl.org) website announced third-grader from Gilbert Link- Grade 7, Astoria, New York; Landis the winners of its 2011 Essay Con- ous Elementary School in Blacks- Bing, Grade 8, Land O’Lakes, Flor- test, “Engineering and Human burg, Virginia, for her description ida; Naina Iyengar, Grade 11, Mon- Service—Relief from a Disas- of the capsule designed to rescue mouth Junction, New Jersey; and ter.” This year, students in grades the Chilean miners trapped under- Zachary Tucker, Grade 12, Lima, 3 through 12 were asked to describe ground last year. In her essay, she Ohio. All of the winning essays an item used for disaster relief and explains the symbolism of the cap- have been posted on the Engineer- explain the engineering behind its sule’s design, as well as its technical Girl! website. design. Older students were also specifications. EngineerGirl!, NAE’s innova- challenged to suggest potential For grades 6 through 8, the first- tive website, is a general reference changes that might be made to the place winner was Ming Li Wu, a point for young women consider- item for use in a different disaster sixth grader from North Alabama ing careers in engineering, a field in relief scenario. Friends School in Huntsville, which they have been, and continue This year NAE received more Alabama. She described a solar- to be, underrepresented. Engineer than 960 entries from students all powered refrigerator created by Girl! provides career guidance for over the country. The essays were students for an engineering compe- students and parents, links to other judged based on creativity and tition in a story-like narrative that sites, games, and interesting facts originality, design potential and showed imagination and creativity. about engineering and the history feasibility, and communication. The winner of first prize for grades of women in engineering. Portions Contestants were encouraged to 9 through 12 was Samantha Har- of the site have been translated into find unique items for disaster relief ris, a 10th-grader from Charleston Spanish to meet the needs of the and to thoroughly explain their Catholic High School in Charles- Latino community. design and, for the older contes- ton, West Virginia. The subject The 2011 EngineerGirl! Essay tants, to think carefully about how of her essay was special chambers Contest was made possible by the to improve upon them. in underground mines that provide generous sponsorship of Energy Prizes were awarded to winners in emergency shelter for miners. Har- Solutions and Bechtel. For more three categories: elementary school ris’ essay provides an in-depth look information about EngineerGirl!, (grades 3 through 5), middle school at advances in engineering technol- see www.engineergirl.org or e-mail (grades 6 through 8), and high ogy that have made it possible to [email protected]. A companion school (grades 9 through 12). Prizes create underground chambers with website, www.EngineerYourLife.org, ranged from $500 for first place to controlled levels of oxygen and car- is geared for academically prepared certificates for honorable mention. bon monoxide. high school girls. The 54 BRIDGE

Projects by the Center for the Advancement of Scholarship on Engineering Education (CASEE)

The third Frontiers of Engineer- host a workshop for engineering 8 community colleges and 17 four- ing Education (FOEE) Sympo- deans to identify impediments to year engineering colleges took place sium, sponsored by the O’Donnell adapting these models at their insti- during the spring 2011 semester. In Foundation, will be held in Novem- tutions and discuss ways of over- June 2011, NAE hosted a policy ber. All attendees are early-career coming them. summit, with representatives of engineering faculty who have all participating institutions and introduced innovative educational CASEE recently developed a individuals from federal agencies projects in their classrooms. The website, Principal Investigators and education-related associations. attendees are selected from indi- Garnering Useful Instruction on Discussions centered on the survey viduals nominated by engineering Developing [Project] Effectiveness data per se, policy implications for deans and NAE members based on (PI GUIDE), for online mentor- student transitions, and the impacts the novelty and potential impact ing in educational project manage- of data collection on institutional of their educational innovations. ment and change leadership (http:// resources. Data analysis is ongoing, The 2011 symposium is organized govpiguide.org). The site includes and we hope to broaden the institu- around three dimensions of peda- video scenarios drawn from actual tional sample in future projects. gogy: (1) expertise in teaching situations and discussions by expe- students cutting-edge engineering rienced PIs about how to man- Finally, CASEE provides research- knowledge and skills; (2) project- age the situations they depict. In to-practice summaries for promoting based learning; and (3) active and addition, the site supports a peer- gender equity and improving teach- self-directed learning. mentoring network for novice ing based on research on engineer- PIs. In the future, the site will ing education, social science, and CASEE recently received a grant also include videos of PIs from educational psychology (www.nae. from Advanced Micro Devices Inc. Advanced Technological Education edu/casee and www.nae.edu/casee- (AMD) to create a Guide to Infus- (ATE), Transforming Undergradu- equity). CASEE also recently devel- ing Real-World Experiences into ate Education in STEM (TUES), oped a website that provides short Engineering Education. For this Research on Gender in Science and inspirational videos that can be used project, CASEE will ask engineer- Engineering (GSE), and the Robert as recruitment tools by institutions ing deans and faculty to submit J. Noyce Scholarship Programs. or individuals. This website high- descriptions of successful models of lights individuals with engineering real-world engineering education CASEE and the American Soci- degrees who have interesting jobs to a committee of industry and aca- ety for Engineering Education and provides links to websites that demic experts, who will select the (ASEE) conducted a pilot sur- describe engineering and pathways most exemplary models to highlight vey of engineering and engineer- to earning engineering degrees in the printed guide. In addition ing technology students in 2- and (http://engineeringaworldofdifference. to publishing the guide, NAE will 4-year institutions. The survey of org or http://eemawod.org). FALL 2011 55

Presentations at the Directors’ Guild of America

NAE worked with the Science entertainment industry hoping to and think about things like chem- and Entertainment Exchange and be inspired and to get story ideas istry and robots in ways they never IEEE to organize three presenta- attended the event. have before. I know that these sorts tions at the Directors’ Guild of The session was moderated by Jon of fertile conversations have directly America on June 9. Sponsored by Spaihts, a science-fiction screen- influenced my own storytelling.” IEEE, the event brought together writer. Spaihts noted, “There is a At the conclusion of the event, NAE member Frances Arnold of two-sided exchange going on here. Frances Arnold added, “Science is a Caltech, Maja Mataric of USC, Obviously, the more that film limitless source of ideas, and engi- and Randii Wessen of NASA JPL makers like myself learn from sci- neering is both cool and fun. Film- to discuss cutting-edge research in entists, the more stories we find. I makers like those here tonight need directed evolution, robotics, and hope this auditorium is filled with to spread the word to young people space exploration, respectively. storytellers whose imaginations are that engineering gives you the tools Approximately 80 members of the magnified by what they hear tonight to change the world.”

NAE-IOM “Go Viral” Challenge

A team of college students from community interaction. The devel- The first-place team received a Cooper Union for the Advance- opers of SleepBot are Edison Wang $3,000 prize and an opportunity ment of Science and Art, New York and Kevin Tulod, Cooper Union; to demonstrate SleepBot during University (NYU), and North- Jane Zhu, NYU; and William Qiao, the plenary session, on June 9, of western University won the top Northwestern. the Health Data Initiative Forum, prize in the National Academy of A team from Arizona State Uni- a gathering of health leaders, soft- Engineering–Institute of Medicine versity (ASU) claimed second prize ware engineers, and IT developers (NAE-IOM) “Go Viral to Improve with Freebee, an app that spreads working to accelerate the public use Health: IOM-NAE Health Data awareness on college campuses about of health data and spur innovation Collegiate Challenge.” health risks from alcoholism, smok- to improve individual and commu- SleepBot, an app that charts users’ ing, unsafe sex, drug use, and unsafe nity health. sleep habits and benchmarks them campus behavior. Freebee develop- The Freebee and IMPAct teams against potential threats associated ers are Jennifer Burkmier, Ramya received $2,000 and $1,000, respec- with sleep deprivation, was selected Baratam, Louis Tse, Chris Work- tively, and displayed their winning as the first-prize winner from sub- man, and Jane Lacson. Details about technologies in the exhibit hall at missions by 15 teams in response Freebee are available at http://www. the forum. Hosted by IOM and the to the challenge, which was issued youtube.com/watch?v=yd-qKqIjCT0. U.S. Department of Health and by IOM and NAE as part of the Another ASU team took third Human Services, the purpose of the Health 2.0 Developer Challenge. place with IMPAct, a Web-based, forum was to promote interaction The challenge to university under- interactive planner that enables among health and data experts and graduate and graduate students was individuals and families to keep explore topics related to applica- to work in interdisciplinary teams track of medical appointments. tions of health information. Sev- to develop a Web-based or mobile IMPAct team members are Tania eral companies and organizations product that tackles a health issue Lyon, Taylor Barker, Edgar Sanchez, announced new challenges and in a creative way and encourages Eric Kern, and Jennifer Jost. other initiatives during the event. The 56 BRIDGE

New Volume of Memorial Tributes Available

Volume 14 of Memorial Tributes, Victor Froiland Bachmann De Mello Thomas L. Martin Jr. a series honoring deceased mem- Michael L. Dertouzos David Middleton bers and foreign associates, is now Coleman Dupont Donaldson Joseph Miller available. Each volume is a collec- Jackson Leland Durkee William W. Moore tion of articles, mostly by friends or Gunnar Fant Morris Muskat business associates of the deceased, Irene K. Fischer Phillip S. Myers highlighting his or her contribu- Patrick F. Flynn Robert E. Newnham tions to engineering that have ben- John W. Fondahl James Y. Oldshue efited humankind. NAE members Gerard F. Fox Ralph B. Peck or foreign associates who wish to John L. Gidley Theodore H.H. Pian receive copies of Volume 14 or a John J. Gilman William Hayward Pickering previous volume should contact the Earl E. Gossard Nathan E. Promisel NAE Membership Office at (202) Serge Gratch Robert O. Reid 334-2198. Copies are available to William A. Griffith Allen F. Rhodes nonmembers from the National William T. Hamilton Jacob T. Schwartz Academies Press, (202) 334-3313. Howard L. Hartman William Rees Sears Tributes to the following indi- Martin C. Hemsworth Franklin F. Snyder viduals are included in Volume 14: Kenneth J. Ives George E. Solomon Joseph M. Juran Morgan Sparks Paul A. Beck Roger P. Kambour John E. Steiner Gary L. Borman Raphael Katzen Olin J. Stephens II Joseph E. Burke Ken Kennedy Thomas G. Stockham Jr. Spencer H. Bush Jack D. Kuehler Bruno Thürlimann L.G. (Gary) Byrd Ralph Landau Rong-yu Wan Benjamin A. Cosgrove Kurt H. Lange Charles M. Wolfe Alan G. Davenport Craig Marks A. Tobey Yu H. Ted Davis Albert R. Marschall

Calendar of Upcoming Events

September 19–21 U.S. Frontiers of Engineering October 16–17 NAE Annual Meeting December 2–3 Committee on Membership Symposium November 3–5 EU-U.S. Frontiers of Meeting Mountain View, California Engineering Symposium Irvine, California October 14–15 NAE Council Meeting Irvine, California All events are held in Washington, D.C., unless October 15 NAE Peer Committee November 8–9 NRC Governing Board otherwise noted. Meetings Meeting FALL 2011 57

NAE Annual Meeting, October 16–17, 2011

The 2011 NAE Annual Meeting honor hosted by the NAE Council. and Design.” In the afternoon, sec- will be held October 16–17 at the The induction ceremony for the tion meetings will be held at the JW Marriott Hotel and the Keck Class of 2011 will be held at noon Keck Center and the JW Marriott. Center of the National Academies on Sunday, October 16. An awards The Annual Meeting will conclude in Washington, D.C. Members of program will follow. that evening with an optional din- the NAE Class of 2011 will meet On Monday morning, October 17, ner dance at the JW Marriott. on Saturday, October 15, for an there will be a Business Session for The flyer for the NAE 2011 orientation. That evening the new members and foreign associates, fol- Annual Meeting is available on the members and foreign associates will lowed by the annual Forum “Making NAE website. Go to www.nae.edu attend a black tie dinner in their Things: 21st Century Manufacturing to register online.

In Memoriam

PAUL M. ANDERSON, 85, presi- leadership in engineering advance- engineering practice, research and dent, Power Math Associates, died ment in the steel industry.” the engineering literature.” on April 26, 2011. Dr. Anderson was elected to NAE in 2009 “for HARRY E. BOVAY JR., 96, presi- DANIEL D. JOSEPH, 82, Regents contributions that have advanced dent, Mid-South Telecommunica- Professor Emeritus and Russell J. the analysis and control of electric tions Company, died on May 24, Penrose Professor Emeritus, Uni- power systems worldwide.” 2011. Mr. Bovay was elected to NAE versity of Minnesota, and Dis- in 1978 “for contributions to expan- tinguished Emeritus Professor of IRVING L. ASHKENAS, 94, sion of knowledge in the energy field Mechanical Engineering, Univer retired chairman of the board, Sys- including power generation and sity of California, Irvine, died on tems Technology Inc., died on April utilization, and leadership in petro- May 24, 2011. Dr. Daniel was 10, 2011. Mr. Ashkenas was elected chemical plant development.” elected to NAE in 1990 “for devel- to NAE in 1992 “for leadership in opment of ingenious analytical tools flying qualities theory and practice, WILLARD S. BOYLE, 86, retired and laboratory experiments used in and for contributions to flight con- executive director, Communica- the discovery and elucidation of trol systems and aerospace vehicle tions Research Division, AT&T novel fluid-mechanic phenomena.” system design.” Bell Laboratories, died on May 7, 2011. Dr. Boyle was elected to ROBERT G. KOUYOUMJIAN, ROBERT R. BEEBE, 83, inde- NAE in 1974 “for contributions 87, Professor Emeritus of Electrical pendent consultant, died on June to solid-state electronics including Engineering, Ohio State Univer- 11, 2011. Mr. Beebe was elected the solid-state laser and charge- sity, died on January 3, 2011. Dr. to NAE in 1990 “for notable con- coupled devices.” Kouyoumjian was elected to NAE tributions to the mining industry in in 1995 “for contributions to the the area of mineral processing and CYRIL M. HARRIS, 93, Charles development of the uniform geo- materials handling.” Batchelor Professor Emeritus of metric theory of diffraction and the Electrical Engineering and Professor analysis and design of antennas and DONALD J. BLICKWEDE, 90, Emeritus of Architecture, Columbia scatterers.” retired vice president, research, University, died on January 4, 2011. Bethlehem Steel Corporation, died Dr. Harris was elected to NAE in MAX V. MATHEWS, 84, Profes- on April 24, 2011. Dr. Blickwede 1975 “for contributions to the field sor (Research) of Music, Emeri- was elected to NAE in 1976 “for of acoustical engineering through tus, CCRMA/Music Department, The 58 BRIDGE

Stanford University, died on April Institute of Technology, died on JOHN A. TILLINGHAST, 84, 21, 2011. Dr. Mathews was elected June 3, 2011. Dr. Rohsenow was independent consultant, died on to NAE in 1979 “for contributions elected to NAE in 1975 “for con May 7, 2011. Mr. Tillinghast was to computer generation and analysis tributions to boiling and condens- elected to NAE in 1974 “for lead- of meaningful sounds.” ing liquid-heat transfer and the ership in the development and teaching of the concepts of heat utilization of advanced systems of UN-CHUL PAEK, 76, Professor and mass transfer.” generation and transmission.” Emeritus, Department of Infor- mation and Communications, JOHN H. SINFELT, 80, Emeritus J. ERNEST WILKINS JR., 87, Gwangju Institute of Science and Senior Scientific Advisor, Exxon Distinguished Professor of Applied Technology, died on May 3, 2011. Research and Engineering Com Mathematics and Mathematical Dr. Paek was elected to NAE in pany, died on May 28, 2011. Dr. Physics, Emeritus, Clark Atlanta 1998 “for the practical production Sinfelt was elected to NAE in 1975 University, died on May 1, 2011. of optical fibers.” “for contributions in catalysis by Dr. Wilkins was elected to NAE in metals and bifunctional catalysis, 1976 “for peaceful application of ROBERT J. PARKS, 89, retired and especially for the concept of atomic energy through contribu- deputy director, Jet Propulsion Lab- ‘polymetallic cluster’ catalysts.” tions to the design and development oratory, died on June 3, 2011. Mr. of nuclear reactors.” Parks was elected to NAE in 1973 ROBERT C. STEMPEL, 77, chair- “for contributions in radio-inertial man and CEO, Energy Conversion JACK KEIL WOLF, 76, Stephen guidance, communications meth- Devices Inc., died on May 7, 2011. O. Rice Professor, University of ods, systems engineering, and proj- Mr. Stempel was elected to NAE in California, San Diego, Center for ect management of spacecraft and 1990 “for outstanding contributions Magnetic Recording, died on May missiles.” to automotive emission control, 12, 2011. Dr. Wolf was elected to fuel economy, and safety engineer- NAE in 1993 “for contributions to WARREN M. ROHSENOW, 90, ing and for leading the integration information theory, communication Professor Emeritus, Massachusetts of such developments.” theory, magnetic recording, and engineering education.” FALL 2011 59

Publications of Interest

The following reports have been underrepresented in science and computing architectures. Despite published recently by the National engineering. The committee argues increasing diversity in computer Academy of Engineering or the that the federal government, indus- designs to optimize for power and National Research Council. Unless try, and post-secondary institutions throughput, the next generation otherwise noted, all publications are must work with K–12 schools and of discoveries is likely to require for sale (prepaid) from the National school systems to increase minor- advances in both hardware and Academies Press (NAP), 500 Fifth ity access to and demand for post- software. In addition, there is no Street, N.W., Lockbox 285, Wash- secondary STEM education and guarantee that parallel computing ington, DC 20055. For more infor- technical training. To that end, the will be as common and easy to use mation or to place an order, contact committee identifies best practices as sequential single-processor com- NAP online at http://www.nap.edu and provides a comprehensive road puter systems are now. Neverthe- or by phone at (888) 624-8373. map for increasing the involvement less, unless we aggressively follow (Note: Prices quoted are subject to of underrepresented minorities in the recommendations in this report, change without notice. Online orders STEM education and improving it will be “game over” for improve- receive a 20 percent discount. Please the quality of their education in ments in computing performance. add $4.50 for shipping and handling for general. Recommendations focus Unless parallel programming and the first book and $0.95 for each addi- on academic and social support, related software developments are tional book. Add applicable sales tax institutional roles, teacher prepa- widely adopted, the development of or GST if you live in CA, DC, FL, ration, affordability, and program new applications, which have always MD, MO, TX, or Canada.) development. driven the computer industry, will NAE members on the study stall, and many other parts of the Expanding Underrepresented Minor- committee were Wesley L. Har- economy are likely to follow. The ity Participation: America’s Science and ris, Charles Stark Draper Professor authoring committee of this report Technology Talent at the Crossroads. of Aeronautics and Astronautics describes the limits of single proces- For the United States to maintain and associate provost, Massachu- sors based on complementary metal its global leadership and competi- setts Institute of Technology; John oxide semiconductor (CMOS) tech- tiveness in science and technol- B. Slaughter, professor of educa- nology and the challenges inherent ogy, we must invest in research, tion and engineering, University in parallel computing and archi- encourage innovation, and educate of Southern California; and Rich- tecture, including increased power a strong, talented, diverse science ard A. Tapia, University Professor consumption and escalating require- and technology workforce. Accord- and Maxfield-Oshman Professor ments for dissipating heat. The ing to the authoring committee of of Engineering, Rice University. committee also provides a research, this report, the U.S. labor market is Paper, $40.00. practice, and educational agenda for projected to grow faster in science overcoming these challenges. and engineering than in any other The Future of Computing Performance: NAE members on the study sectors in the coming years, making Game Over or Next Level? With the committee were Samuel H. Fuller minority participation in science, end of dramatic exponential growth (chair), chief technology officer technology, engineering, and medi- in single-processor performance, and vice president of research and cine (STEM) on all educational the era of sequential computing development, Analog Devices Inc.; levels a national priority. The must give way to a new era of par- Robert P. Colwell, consultant, study committee analyzes the rate allelism. Because the change will R & E Colwell & Associates, of change and challenges to diver- mean overcoming significant sci- and retired fellow, Intel Corpora- sity, especially because minorities, entific and engineering challenges, tion; William J. Dally, Willard which are the fastest growing seg- this is an opportune time for inno- R. and Inez Kerr Bell Professor ment of the population, are severely vation in programming systems and of Computer Science, Stanford The 60 BRIDGE

University; Daniel W. Dobber- time conditions in impacted areas, equipment and infrastructure now, puhl, vice-president, Apple; Mark and inform decisions by respond- which can “lock us in” to commit- A. Horowitz, chair, Department ers. This workshop was one ses- ments to continuing greenhouse of Electrical Engineering, Stanford sion in a World Bank conference, gas emissions for decades, should be University; and David B. Kirk, fel- “Understanding Risk: Innovation in revisited. Finally, although it may low, NVIDIA. Paper, $36.00. Disaster Risk Assessment.” Work- be possible to scale back or reverse shop participants emphasized three responses to climate change, it is Blue Water Navy Vietnam Veterans messages: (1) the need to integrate difficult or impossible to “undo” and Agent Orange Exposure. The U.S. bottom-up communications from climate change itself. Therefore, Department of Veterans Affairs citizens to keep emergency respond- although efforts by local, state, and (VA) has established that Viet- ers and managers informed of private-sector actors are important, nam veterans who develop diseases changing conditions; (2) the need they are likely to be less successful associated with exposure to Agent to prepare people for disaster and than efforts based on strong fed- Orange are automatically eligible emergency situations by anticipat- eral policies establishing coher- for disability benefits—with the ing emotional reactions, develop- ent national goals and incentives exception of veterans who served ing and practicing emergency plans, and promoting U.S. engagement on deep-sea vessels. These “Blue and improving communications and in responses on an international Water Navy” veterans must prove preparedness; and (3) understand- level. The inherent complexities they were exposed to Agent Orange ing how virtual and personal social and uncertainties of climate change before they can claim benefits. At networks contribute to resilience can best be met by: iterative risk the request of the VA, the Institute and how access to technological risk management and efforts to signifi- of Medicine conducted a study on assessments can increase resilience. cantly reduce greenhouse gas emis- whether Blue Water Navy veterans NAE member Gerald E. Gallo sions; preparations for adapting to were subject to exposures to Agent way Jr., Glenn L. Martin Institute impacts; investments in scientific Orange similar to the exposures of Professor of Engineering, Univer- research, technology development, other Vietnam veterans. sity of Maryland, College Park, and information systems; and col- NAE member Menachem Eli was a member of the roundtable. laboration by scientific and techni- melech, Roberto C. Goizueta Pro- Free PDF. cal experts and other stakeholders fessor, Environmental Engineering involved in climate choices. Program, Yale University, was a America’s Climate Choices. The study NAE members on the study member of the study committee. committee argues that environmen- committee were Albert Carnesale Paper, $37.50. tal, economic, and humanitarian (chair), Chancellor Emeritus and risks posed by climate change indi- professor, University of California, How Communities Can Use Risk Assess- cate a pressing need for taking action Los Angeles, and Charles O. Hol- ment Results: Making Ends Meet: A now to limit the magnitude of cli- liday Jr., retired chairman of the Summary of the June 3, 2010 Workshop mate change and prepare for adapt- board and CEO, DuPont. Paper, of the Disasters Roundtable. During ing to its impacts. Despite some $29.95. and after a disaster, text messages, uncertainty about future risk, act- Tweets, Smartphone apps, and social ing now will reduce whatever risks Assessment of Fuel Economy Technolo- networks, along with 24-hour cable there are and reduce the need for gies for Light Duty Vehicles. Vari- news and other media, can deliver larger, more rapid, and potentially ous combinations of commercially relevant information to emergency more expensive actions to reduce available technologies could greatly responders, decision makers, and risks later. Reducing vulnerabilities reduce fuel consumption in passen- the general public. Participants in to the impacts of climate change ger cars, sport-utility vehicles, mini- the workshop summarized in this requires mostly common sense vans, and other light-duty vehicles volume identified ways to use these investments that would also protect without compromising vehicle per- technologies to communicate risks against natural climate variations formance or safety. This National associated with an emergency or and extreme events. In addition, Research Council report provides disaster, identify and assess real- crucial decisions about investing in estimates of potential fuel savings FALL 2011 61

and costs to consumers for avail- census. The panel suggests that The committee also estimates the able technologies for three types USCB take an assertive, proactive resources necessary to provide sta- of engines: spark-ignition gasoline approach to planning for 2020 and ble, long-term CI for research in engines; compression-ignition die- makes three core recommendations. combustion. sel engines; and hybrid engines. First, the panel identifies four broad NAE members on the study com- Because energy savings are directly topic areas for research early in the mittee were Chung K. Law, Robert related to the amount of fuel used, decade. Second, USCB should H. Goddard Professor of Mechani- the focus is on fuel consumption adopt an aggressive, assertive pos- cal and Aerospace Engineering, (i.e., the amount of fuel consumed ture toward research in these prior- Princeton University, and Mark S. in a given driving distance). Sav- ity areas. Third, USCB should set Lundstrom, Don and Carol Scifres ings can be measured by money bold goals to underscore the need Distinguished Professor of Electrical saved on purchases of fuel and for reengineering and building a and Computer Engineering, Purdue decreases in emissions of carbon commitment to change. University. Paper, $30.25. dioxide. By contrast, fuel econo- NAE members on the study com- my, the usual information given on mittee were Thomas M. Cook Assessment of Approaches for Using vehicle stickers, is a measure of how (chair), former president, T.C.I., Process Safety Metrics at the Blue far a vehicle can travel on one gal- and Arthur M. Geoffrion, James Grass and Pueblo Chemical Agent lon of fuel. The study committee A. Collins Chair in Management Destruction Pilot Plants. The U.S. concludes that vehicle stickers that Emeritus, UCLA Anderson School Department of Defense, through provide information about both of Management. Paper, $21.00. the Assembled Chemical Weap- fuel consumption and fuel economy ons Alternatives Program, is con- would be most helpful to consumers. Transforming Combustion Research structing two full-scale pilot plants NAE members on the study through Cyberinfrastructure. Even at the Pueblo Chemical Depot in committee were Trevor O. Jones in the face of climate change and Colorado and the Blue Grass Army (chair), chairman and chief execu- the increasing availability of alter- Depot in Kentucky for the purpose tive officer, ElectroSonics Medi- native energy sources, fossil fuels of destroying the last two remaining cal Inc.; Thomas W. Asmus, will continue to be used for many inventories of chemical weapons in retired senior research executive, decades. However, they are likely the U.S. stockpile. Together, these DaimlerChrysler Corp.; Rodica to become increasingly expensive two storage sites account for about A. Baranescu, manager, Fuels and as pressure to minimize the by- 10 percent of the original U.S. Lubricants Engine Group, Navistar products of combustion (pollutants) chemical agent stockpile, which Inc.; Linos J. Jacovides, president, also increases. The Multi-Agency is being destroyed in accordance Paphos Consulting, and retired Coordinating Committee on Com- with the international Chemical director, Delphi Research Labs; bustion Research requested that the Weapons Convention Treaty. The John G. Kassakian; professor of National Research Council con- facilities at the Pueblo and Blue electrical engineering and computer duct a study of the structure and Grass sites will use neutralization science, Massachusetts Institute of use of a cyber infrastructure (CI) technologies, which will require Technology; and Robert F. Sawyer, for research on improved combus- specially designed equipment, to Professor Emeritus, Department of tion systems. The committee was destroy chemical agents in rock- Mechanical Engineering, Univer- asked to explore the potential ben- ets, projectiles, and mortar rounds. sity of California. Paper, $60.00. efits of CI for research and future The Program Manager for Assem- applications; evaluate accessibility, bled Chemical Weapons Alterna- Change and the 2020 Census: Not sustainability, and economic mod- tives, which is responsible for safe Whether But How. In this interim els; identify CI necessary for edu- operation of the facilities, asked report, a panel of experts, formed cation in combustion science and the National Research Council by the National Research Council engineering; and identify human, to conduct a study on the efficacy at the request of the U.S. Census cultural, institutional, and policy of using process safety metrics at Bureau (USCB), suggests general challenges and describe how they these two sites. Process safety is research priorities for the 2020 are being addressed in other fields. a framework of design principles, The 62 BRIDGE engineering, and operating practices NAE member A. Thomas Young, During the subsequent investiga- for managing the integrity of operat- retired vice president, Lockheed tion, the FBI worked with other ing systems, processes, and person- Martin Corporation, was a mem- federal agencies to coordinate and nel handling hazardous substances. ber of the study committee. Paper, conduct scientific analyses of spore The authoring committee discusses $55.00. powders from the letters, environ- the pros and cons of using leading mental samples, clinical samples, and lagging process safety metrics to Federal Funding of Transportation and samples collected from labora- provide feedback on the effective- Improvements in BRAC Cases. In a study tories that were possible sources of ness of controls for mitigating risks by the Transportation Research the spores. The agency also called and minimizing the consequences Board on improving Defense Base in external experts, including some of incidents and offers recommen- Closure and Realignment Com- who had previously developed tests dations for facilitating the devel- mission (BRAC) decisions, the to differentiate among strains of opment and application of process study committee concludes that B. anthracis. In 2008, seven years safety metrics at both sites. traffic delays resulting from current into the investigation, the FBI NAE member Mauricio Futran, BRAC decisions and short time- asked the National Research Coun- consultant, Westfield, New Jersey, lines for implementing those deci- cil to conduct an independent was a member of the study commit- sions impose substantial costs on review of the scientific approaches tee. Paper, $21.00. surrounding communities and may used during the investigation to even be harmful to the military. determine (1) if they met appropri- New Worlds, New Horizons in Astron- The report offers recommendations ate standards for scientific reliabil- omy and Astrophysics: Panel Reports. for mitigating the adverse effects of ity and for use in forensic validation Every 10 years the National BRAC decisions for the near, short, and (2) whether the FBI reached Research Council releases a sur- and long term. Among its recom- appropriate scientific conclusions. vey of astronomy and astrophysics, mendations, the committee calls NAE members on the study com- outlines priorities for the coming on Congress to consider a special mittee were Alice P. Gast (chair), decade, and recommends overall appropriation or the allocation of president, Lehigh University, and priorities for the field as a whole. uncommitted Stimulus funds to David R. Walt, Robinson Profes- This volume is a collection of panel address severe transportation prob- sor of Chemistry, Tufts University. reports on key sub-areas: cosmology lems caused by increases in military Paper, $49.00. and fundamental physics; galaxies traffic. The purpose of these funds across cosmic time; the galactic would be to initiate projects as soon National Security Implications of Cli- neighborhood; stars and stellar as possible to reduce congestion mate Change for U.S. Naval Forces. In evolution; planetary systems and within three years. response to a request from the Chief star formation; electromagnetic NAE member Thomas B. Deen, of Naval Operations, the National observations from space; optical retired executive director, Trans- Research Council appointed a and infrared astronomy from the portation Research Board, National committee, under the auspices of ground; particle astrophysics and Research Council, was a member of the Naval Studies Board, to study gravitation; and radio, millimeter, the study committee. Free PDF. the national security implications and submillimeter astronomy from of climate change for U.S. naval the ground. This companion vol- Review of the Scientific Approaches forces. The committee concluded ume to New Worlds, New Horizons: Used During the FBI’s Investigation that if even the most moderate cur- A Decadal Survey of Astronomy of the Anthrax Letters. Less than a rent trends in climate continue, they and Astrophysics will be useful to month after September 11, 2001, will present new security challenges managers of research programs in letters containing anthrax spores for the U.S. Navy, Marine Corps, astronomy and astrophysics, con- (Bacillus anthracis, or B. anthracis) and Coast Guard. Although the gressional committees with juris- were sent through the U.S. mail. timing, degree, and consequences of diction over agencies that support Between October 4 and Novem- future climate change remain uncer- this research, the scientific com- ber 20, 2001, 22 individuals devel- tain, changes are already under way munity, and the public. oped anthrax; 5 cases were fatal. in regions around the world, such FALL 2011 63

as the Arctic, that call for a naval processes. Consequently, FDA and (NAKFI) Conference on Imaging response. The report addresses both the U.S. Department of Health Science, researchers from academia, near- and long-term implications and Human Services asked the industry, and government formed for U.S. naval forces and provides National Research Council to 14 interdisciplinary teams to find a corresponding findings and recom- develop a conceptual model for common language and structure for mendations. Because the terms of evaluating products or product cat- developing new technologies, pro- reference for the study directed that egories and to provide information cessing and recovering images, min- it be based on scenarios developed on associated potential health con- ing imaging data, and visualizing by the Intergovernmental Panel on sequences. This report describes data effectively. The teams spent Climate Change and other peer- a proposed risk-characterization nine hours over a period of two days reviewed assessments, the commit- framework for evaluating, compar- exploring challenges at the inter- tee did not address the science of ing, and communicating the public- face of science, engineering, and climate change itself or challenge health consequences of decisions medicine. This volume provides the scenarios on which its findings about a wide variety of products. summaries written by each team and recommendations are based. The framework is intended to com- describing a challenge and outlining NAE members on the study complement, not replace, other risk- an approach to solving it. The sum- mittee were Frank L. Bowman based approaches that are already maries also include research areas (chair), president, Strategic Deci- in use or are under development that should be explored to clarify sions LLC, and U.S. Navy (retired); at FDA. The proposed framework the fundamental science behind the Arthur B. Baggeroer, Ford Profes- provides a common language for challenge, a plan for engineering sor of Engineering, Secretary of the describing potential public-health the application, an explanation of Navy/Chief of Naval Operations consequences of decisions, is appli- the reasoning behind the approach, Chair in Oceanographic Science, cable for all FDA centers, and and a description of the benefits to Departments of Mechanical Engi- draws extensively on the scientific society of solving the problem. neering and EECS, Massachusetts literature to define relevant health NAE members on the steering Institute of Technology; David J. dimensions for FDA decision mak- committee were Farouk El-Baz, Nash, chairman and CEO, Dave ing. In addition to conclusions and research professor and director, Nash & Associates International recommendations, the committee Center for Remote Sensing, Bos- LLC; and David A. Whelan, vice provides case studies of how the ton University, and Charles Elachi, president, Strategic Innovation framework can be used. director, Jet Propulsion Laboratory. Phantom Works, and chief scientist, NAE member John T. Watson, Paper, $34.75. Boeing Defense, Space, and Secu- University of California, San Diego, rity, The Boeing Company. Paper, was a member of the study commit- Forging the Future of Space Science: $43.00. tee. Paper. $45.75. The Next 50 Years. From September 2007 to June 2008, the Space Studies A Risk-Characterization Framework for Seeing the Future with Imaging Science: Board conducted a monthly seminar Decision-Making at the Food and Drug Interdisciplinary Research Team Summa- series, with each talk highlighting a Administration. Every day, the Food ries. Imaging science has the power different topic in space and Earth and Drug Administration (FDA), to illuminate regions as remote as science. This volume includes the which is responsible for ensuring distant galaxies and as close to home principal lectures from the series on the safety of food, drugs, and other as our own bodies. Although many subjects ranging from global climate products, must make decisions, disciplines that can benefit from change to the cosmic origins of life, often based on incomplete infor- imaging have common technical exploration of the Moon and Mars, mation, that have public-health problems, researchers often develop and scientific research to support consequences. FDA recognizes ad hoc methods for performing indi- human spaceflight. The prevailing that systematically collecting and vidual tasks rather than building messages throughout the seminars evaluating information on the frameworks that could address these are: how much we have accom- risks posed by regulated products problems. At the 2010 National plished over the past 50 years; how would improve its decision-making Academies Keck Futures Initiative profound our discoveries have been; The 64 BRIDGE how much contributions from the of lighter, more durable high- Protecting the Frontline in Biodefense space program have affected our temperature materials to reduce Research: The Special Immunizations daily lives; and yet how much engine weight without reducing Program. The U.S. Army’s Spe- remains to be done. thrust. Unfortunately, materials devel- cial Immunizations Program is an NAE members on the study board opment has been significantly cut important component of its over- were A. Thomas Young (vice chair), back since the early 1990s, when the all biosafety program for labora- Lockheed Martin Corporation United States led the world in propul- tory workers at risk of exposure to (retired); Daniel N. Baker, direc- sion technology. This study provides hazardous pathogens. The program tor, Laboratory for Atmospheric and an overview of current and planned provides immunizations for scien- Space Physics, University of Colo- efforts to meet U.S. military needs, tists, laboratory technicians, and rado at Boulder; Yvonne C. Brill, considers mechanisms for the timely other support staff who work with aerospace consultant, Skillman, insertion of materials in propulsion certain hazardous pathogens and New Jersey; Saroosh Sarooshian, systems and how these mechanisms toxins. Although the program was UCI Distinguished Professor and might be improved, and describes established to serve military per- director, Center for Hydrometeorol- general research and development sonnel, it was expanded through a ogy and Remote Sensing, University strategies for developing materials for cost-sharing agreement in 2004 to of California, Irvine; and Warren future military aerospace propulsion include other government and civil- M. Washington, senior scientist, systems. The conclusions and recom- ian workers, reflecting the expan- Climate Change Research Section, mendations point the way to improv- sion of biodefense research in recent Climate and Global Dynamics Divi- ing the efficiency, level of effort, and years. This report focuses on issues sion, National Center for Atmo- impact of materials development by related to the expansion of the pro- spheric Research. Paper, $39.25. the military. gram, such as regulatory frameworks NAE members on the study com- under which vaccines are adminis- Materials Needs and Research and mittee were Wesley L. Harris, tered, the addition of new vaccines, Development Strategy for Future Mili- Charles Stark Draper Professor of and factors that might influence the tary Aerospace Propulsion Systems. Aeronautics and Astronautics and development and manufacture of The ongoing development of mili- associate provost, Massachusetts vaccines for the program. tary aerospace platforms requires Institute of Technology, and Wil- NAE member Stephen W. Drew, continuous advances in technol- liam L. Johnson, Ruben and Donna Drew Solutions LLC, and Merck & ogy. However, significant advances Mettler Professor of Materials Sci- Co. Inc. (retired), was a member of in the performance and efficiency ence, Engineering and Applied the study committee. Paper, $43.50. of jet and rocket propulsion systems Science, California Institute of strongly depend on the development Technology. Paper, $47.00.

The Periodicals BRIDGE Postage Paid (USPS 551-240)

National Academy of Engineering 2101 Constitution Avenue, N.W. Washington, DC 20418