Department of Earth System Science zender@uci.edu
University of California Voice: (949) 824-2987
Irvine, CA  92697-3100 Fax: (949) 824-3256

Project Summary
This proposal outlines a program of scientific research and educational development focused on radiative forcing and climate. The scientific program consists of determining the role of three gaseous species in closing Earth’s energy budget and in radiatively forcing present and future climate. The gases are water vapor, collision complexes of oxygen ($O2\cdot X$), and nitrogen dioxide. Absorption by these gases has heretofore been oversimplified or neglected in large scale atmospheric models, so their radiative forcing and climate impact remain uncertain. By representing these missing absorption processes, this project will directly improve predictions of atmospheric heating, estimates of Earth’s energy budget, and, likely, the accuracy of climate predictions.

Oversimplifications in current representations of $H2O$, $O2\cdot X$, and $NO2$ in large scale models are known and will be reduced or eliminated. First, water vapor absorption of solar and terrestrial radiation in cloudy atmospheres is systematically underestimated and biased by neglecting the sub-gridscale partitioning of vapor between the saturated cloud and the sub-saturated clear sky. Second, collision complexes of oxygen absorb about 1 W m$-2$ globally annually averaged, but the climate response to this forcing is unknown. Third, neglected $NO2$ absorption will be examined using a combination of satellite and model data to constrain its distribution. These mechanisms are, in sum, expected to reduce the current discrepancy in the global annual mean atmospheric energy budget by 2–8 W m$-2$ (10–30% of the total discrepancy). Moreover, $H2O$ and $NO2$ forcings are likely to change with climate.

The educational component of the proposal is to found a project to coordinate the development, solicitation, standardization, and dissemination of Freely Available Community Texts (FACTs) suitable for education and teaching in the Earth system sciences. Each FACT will be a living monograph available via the World Wide Web to students and scientists anywhere to study, modify, and improve. The license ensures authors retain recognition, copyright, and review priveleges over modifications to their original material. The project will begin with two existing, pilot FACTs designed to educate students about radiative forcing and aerosols. We envision contributions of new material and FACTs from students and faculty within our department, but, more significantly, from the international geosciences community.

Contents

1 Results from Prior NSF Support

Dr. Zender has not previously had NSF support.

2 Introduction

This proposal outlines a five year program of research and educational development in Earth System Sciences. The first portion of this proposal describes a project to coordinate development and dissemination of a series of freely available educational geoscience monographs over the World Wide Web. Each monograph will be written, reviewed and maintained by talented students and educators with research interests in that field. We argue that this project will become a high-quality resource for geoscience education because, being free, it will enlist talents and contributions from the worldwide community of interested students and educators.

The second portion of this proposal outlines a scientific research program to improve our understanding of gaseous radiative forcing of climate, and of the climate response to this forcing. The research project employs a hierarchy of models to simulate radiative forcing by three important gaseous species, $H2O$, $O2\cdot X$, and $NO2$. This project is organized into sub-projects for each gas, and concludes with integrative studies involving all three gases. The studies contribute separately and in tandem to the overall goals of closing Earth’s radiative energy budget, understanding the response of the climate system to trace gas forcing, and improving predictions of atmospheric heating.

3 Freely Available Community Texts

3.1 Vision Statement

The amount of information a geoscientist must master continuously increases as journals and data archives proliferate and grow. Educators in the geosciences must ingest even more information in order to remain current in their teaching and research careers. Comprehensive textbooks become out of date soon after printing, and narrow, specialists texts are exorbitantly priced. Buying more textbooks to keep up with this information glut is a short term solution that only relatively priveleged students and researchers can afford. A better solution is to harness the networking power of the World Wide Web to coordinate the distributed development, maintainance, and distribution of “living texts” in the geosciences. The educational component of this proposal is to found a project to organize the creation of Freely Available Community Texts (FACTs) suitable for education and teaching in Earth System Sciences.

The FACT project is not an idealistic pipe dream, as shown by the rapid growth of freely distributed and community maintained software known as Free Software or Open Source software [Raymond, 1999]. Adopting the successful principles underlying the Free Software movement, FACT authors will retain copyright to their monographs, but will give the academic community the license to modify, extend, and update (possibly portions of) their texts in perpetuity. It is anticipated that graduate students, postdocs, and researchers will contribute significantly to FACTs from their theses, collections of homework problems, and self-developed course teaching materials. Moreover, all interested geoscientists with access to the Internet, regardless of nationality or income, can benefit from and contribute to the high-quality FACTs.

3.2 Mission

The FACT project is intended to standardize and disseminate our fundamental knowledge of Earth System Sciences in a flexible, adaptive, distributed framework which can evolve to fit the changing needs and technology of the geosciences community. FACTs will be created, reviewed, and continuously maintained and updated by members of the international academic community communicating with eachother through a well-organized project website. Each FACT will describe a specific subject area of the Earth system in detail, using a consistent nomenclature and style common to the series. For example, the FACTs I write (see § 3.3) are integrated with my research and teaching interests in climate and radiative forcing (and thus with my Department’s research mission). The primary author or maintainer of the FACT is responsible for reviewing and approving changes and updates.

FACTs are intended to cover fundamental and well-established principles of a given discipline, not to replace or be an alternative for tradiational, peer-reviewed scientific journals. Thus FACTs will not include speculative or unpublished theories. FACTs should start with the first priniciples of a given field. The continual improvement of FACTs over time will result in texts that are more current and up-to-date with recent advances in the field every year. The long term (5–10 year) goal is to build high quality state of the art monographs which are indispensable to researchers as well as students.

To preserve the freedom of the community to modify the FACTs, and the rights of the authors to control their intellectual property, FACTs will be distributed under a license called the GNU Free Documentation License (FDL) (http://www.gnu.ai.mit.edu/copyleft/fdl.html). The FDL ensures authors retain recognition, copyright, and review priveleges over modifications to their original material. Once an author applies the FDL license, the FACT and all derived works of it are forever freely available to all under the same terms. This perpetuity ensures that FACTs are always and forever available via the World Wide Web to students and scientists anywhere to study, modify, and improve.

3.3 Prototypes and Curricular Integration

Two prototype FACTs, “Radiative Transfer in the Earth System” and “Particle Size Distributions: Theory and Application to Aerosols, Clouds, and Soils” have been developed to supplement the textbook in my new graduate course, ESS 111/211: “Radiative Processes and Remote Sensing”. These prototypes are available from my website at http://www.ess.uci.edu/~zender/rt and http://www.ess.uci.edu/~zender/psd, respectively. Examination of the Radiative Transfer FACT illustrates some of the promise and pitfalls of FACTs.

1. HTML format is for online browsing and the other formats are for printing (Postscript, PDF, DVI) or modification
2. FACTs can look and be as professional as any textbook
3. The Table of Contents, Reference section, and Index make FACTs easy to search
4. Hyperlinks and cross-references within and among FACTs is possible but has not yet been implemented

Examples of welcome contributions which interested students and researchers in the radiative transfer community could make to the radiative transfer FACT include (this task list should be maintained within the FACT itself)

1. Figures which illustrate the material in the text
2. More homework exercises (see the example on Page 18)
3. Standalone special interest boxes (e.g., “Geometric Devivation of Optical Depth” on pages 8–9)
4. Text for unfinished sections such as the optical properties on pages 30–31.

Each graduate student enrolled in my ESS 211 course will be asked to contribute an item from this to the Radiative Transfer FACT. The students will gain appreciation of the community value of distributed collaborative work, and will be credited for the authorship of their contribution in the FACT itself.

3.4 Scientific Quality

Two forms of “peer review” will help ensure FACTs remain high quality materials. First, a manuscript’s primary author may decline any revisions which do not meet his standards. Of course there is strong social pressure against submitting inferior contributions with one’s name attached. On the other hand, FACTs which are particularly well-written and widely used may eventually garner their authors (and their institutions) international recognition for their expertise and pedagogical skills. Second, FACTs are open texts so they are subject to continuous suggestions and refinements by new readers. The exposure will help authors identify any mistakes, omissions, or inadvertent plagiarism in the texts.

3.5 Existing Free Community Geosciences Educational Material

To our knowledge there are no other projects which create and make freely available geoscience texts that are open to community modification. The design of FACTs appears to be truly innovative, which makes foreseeing potential problems or learning from previous mistakes very difficult. We believe that texts which can be modified to fit changing needs of users will be of lasting intellectual value to the geoscience community, and that the community will recognize this and help to make FACTs a success.

3.6 Entraining Students and Educators

One of my priorities is to ensure that others contribute new FACTs which broaden the intellectual appeal of the project beyond my own research interests. I will actively solicit colleagues to consider contributing a FACT in their own niche. An informal poll of colleagues in my department and elsewhere showed all were interested on contributing material to this project. Many expressed hope that FACTs might become an integrative exercise for the department, allowing us to share our expertise and painstakingly developed lecture notes with eachother more easily.

One promising source of contributors are graduate students. While taking classes and preparing for comprehensive and thesis examinations, graduate students often have more time and motivation to read texts than more accomplished professionals in the field. Graduate students who find problems or gaps with these monographs are likelier to send contributions than others. As discussed in § 3.3 graduate students in my courses will be asked to make a contribution to FACTs.

3.7 FACTs Project Organization and Communication

All project bookkeeping and coordination will be performed in the open via mail lists and discussion forums. The FACTs website will initially be based at SourceForge (http://sourceforge.net), a widely used, pro-bono website for Open Source projects. SourceForge hosts my successful netCDF Operator (NCO) project (http://sourceforge.net/projects/nco) (and many much larger projects) and provides the essential tools for administering a large, distributed project like FACTs. The website will provide the following for each FACT: electronic forums for user/author discussion, announcement lists to notify users of new contributions, staging areas for contributions under review, download area.

3.8 Project Work Plan

During Year 1 the programmer/analyst will concentrate on generating document templates for authoring FACTs, and developing a working community website. PI Zender will integrate the prototype FACTs into his graduate-level course on Radiative Processes and will solicit colleagues to contribute new FACTs.

In Years 2 and 3 PI Zender will contribute new FACTs on particle wet and dry deposition processes. The programmer analyst will assist authors contributing FACTs in other areas of Earth Systemm Science, and will investigate the potential of new free document formats such as DocBook and MathML.

By Years 4 and 5 the FACT project should be firmly established. The programmer will work to increase hyperlinks and cross-references between existing FACTs. PI Zender will maintain and improve his four FACTs and will solicit new FACTs in areas where there are still large gaps in the Earth System Science curriculum.

3.9 Project Evaluation

The success or failure of FACTs should be measured by the number of quality contributions received and the number of people who read them. Assessing the first is relatively straightforward. The number of newly contributed manuscripts and pages of improvements to existing manuscripts can be tabulated annually. As with many websites, we will keep track of the number of visitors and manuscript downloads. We will not be aware of second-generation users, e.g., a student who receives a copy of the documentation (either paper or electronic) from a friend rather than directly from the FACTs website.

3.10 Significance to Professional Goals and Responsibilities

Because of its international scope and availability to students of all income levels, the FACT project may allow me and other geoscience educators to impact more students, and to a greater depth, than we could possibly hope to before the advent of the Internet. The integration of FACTs with my research interests and teaching responsibilities in the ESS department enhances its likelihood of success.

3.11 Prior Education, Outreach, and Service Accomplishments

1. Development of Widely Distributed Geophysical Software: I write and manage the netCDF Operators (NCO) toolset (NCO works with HDF4, too). NCO is an Open Source software project (http://nco.sourceforge.net) developing free tools to manipulate geophysical datasets. NCO is one of two components of the CCSM Component Model Processing Suite and is used daily by hundreds of geoscientists.
2. Undergraduate Teaching: I teach ESS 20E: “The Atmosphere”, an undergraduate survey course with approximately 150 students. All course material is placed on the web (http://eee.uci.edu/00s/42020), and the course includes web-based learning exercises. Students gave an overall grade of “B” to this course.
3. Mentoring of Undergraduates: At the University of Colorado I founded and directed the Astrophysical, Planetary, Atmospheric Sciences Departmental Help Center, a free tutoring center which assisted dozens of undergraduates in the physical sciences every semester. My role included securing funding, tutoring, and managing a group of about 10 paid graduate student tutors. This activity occurred under my direction from 1992–1995, and continued after my departure. Throughout this period I contributed to secondary school teaching by answering calls for to the Math and Science Teachers Hotline (1-800-866-MAST) organized by the University of Northern Colorado.

4 Enhanced Gaseous Absorption and Climate

4.1 Overview

Absorption of radiant energy is the fundamental mechanism which drives the climate system. Thus fully understanding the distribution of radiative absorption is essential to understanding climate. Satellite-borne instruments have measured the Earth’s global mean planetary albedo to be $30\pm 1$% [Li et al., 1997; Kiehl and Trenberth, 1997]. The partitioning of the $70\pm 1$% absorbed between the atmosphere and the surface is very poorly constrained because the spatio-temporal coverage of high quality surface observations is inadequate [Wild et al., 1995; Ohmura et al., 1998; Gilgen and Ohmura, 1999]. Until 1995 the consensus estimate of the partitioning, based on models, was that the surface absorbed about 50% of incoming solar radiation while the atmosphere absorbed the remaining 20%. Within the last six years, many independent observational studies have concluded that Earth’s atmosphere may absorb as much as 28% of incoming solar radiation [e.g., Cess et al., 1995; Wild et al., 1995; Li et al., 1995,  1997].

For our purposes this discrepancy between models and measurements of the atmospheric absorbed radiation component of Earth’s energy budget is called enhanced atmospheric absorption, i.e., absorption by processes not represented in large scale atmospheric models used for global studies and for climate prediction. Enhanced absorption accounts for up to 8% of global annual mean incoming solar radiation, or about 25 W m$-2$ [Li et al., 1997; Yu et al., 1999]. Model climates are very sensitive to globally enhanced atmospheric absorption of this magnitude [Kiehl et al., 1995; Collins, 2000] so reducing this discrepancy is crucial to improving prediction of climate and climate change. This proposal concerns improving representation of known gaseous absorption mechanisms, reducing the enhanced absorption discrepancy, and assessing the associated climate response.

Large scale atmospheric models are known to underestimate gaseous absorption by $H2O$, $O2\cdot X$, and $NO2$. This project will include studies of the following processes which are known to contribute to enhanced absorption: First, the sub-gridscale distribution of water vapor in clouds causes solar and terrestrial radiation absorption biases [Crisp, 1997; Fung and Ramaswamy, 1999]. Second, neglected $O2\cdot X$ solar absorption contributes about 1 W m$-2$ to the enhanced absorption [Pfeilsticker et al., 1997; Solomon et al., 1998; Zender, 1999a]. Third, neglected $NO2$ solar absorption causes an unknown, but potentially significant amount of enhanced absorption both in the stratosphere and troposphere [Zender et al., 1997; Solomon et al., 1999].

The proposal is organized as follows: Section 4.2 illustrates the spectral characteristics of all significant gaseous shortwave absorbers. Sections 4.3–4.5 describe the relevant shortcomings of current representations of absorption in large scale atmospheric models due to $H2O$, $O2\cdot X$, and $NO2$, respectively. Section 4.6 summarizes how climate might respond to enhanced absorption. Sections 4.7–4.10 describe our objectives and hypotheses, research plans, methods and procedures, and the expected significance of the results, respectively.

4.2 Shortwave Gaseous Absorption

Since the three absorption processes listed above involve solar absorption it is helpful to see the relative spectral structure of all gaseous solar absorbers. Figure 1 shows the modeled absorption optical depth $\tau abs\left(\lambda \right)$ of all significant gaseous solar absorbers for $200<\lambda <1400$ nm at noontime on a pristine clear sky day in Oklahoma.

Zender et al. [1997] describe the multistream narrow band (10 $\text{ cm}-1$) radiative transfer model, the experimental uncertainties, and all the input data in more detail. Structured or continuum absorption is present at nearly all wavelengths except in the neighborhood of 0.86 $\mu \text{ m}$. $O2\cdot X$ shares with $H2O$ the distinction of overlapping with all the significant solar absorbers. However, the strongest bands of $H2O$, $O2\cdot X$, and $NO2$ are largely independent of one another. Currently all large scale atmospheric models account for most $H2O$ (yellow) absorption, but none include absorption by $O2\cdot X$ (red and dark green) or $NO2$ (light green centered near 0.4 $\mu \text{ m}$).

4.3 Absorption by Water Vapor

Correct representation of water vapor absorption at all wavelengths is critical to climate studies because $H2O$ is both the dominant solar absorber and greenhouse gas in Earth’s atmosphere. Models predict that, globally averaged, gaseous $H2O$ absorbs about 72% and 55% of incident solar radiation in clear and cloudy atmospheres, respectively [Kiehl and Trenberth, 1997]. This absorption is maximized in humid regions above bright surfaces. Regions of persistent marine stratus such as offshore of California, Peru, and Namibia produce the greatest solar absorption by water vapor.

Water vapor accounts for about 55-60% of the total reduction in emission of terrestrial radiation to space (the “greenhouse effect”) in both clear and cloudy skies [Kiehl and Trenberth, 1997]. The strongest longwave forcing by $H2O$ occurs in the upper troposphere [Raval and Ramanathan, 1989; Kiehl and Briegleb, 1992; Lubin, 1994] where it plays a role in maintaining deep convection [Soden and Fu, 1995]. Upper tropospheric absolute humidity may increase in warmer climates due to the water-vapor feedback, although there are exceptions to this reasoning [Lindzen, 1990].

Line-by-line radiative transfer models obtain excellent agreement with the measured structure of water vapor monomer absorption [e.g., Mlawer et al., 1998; Daniel et al., 1999] so there is little systematic error in the tabulated line strengths of the $H2O$ monomer [Rothman et al., 1998]. Unfortunately large scale atmospheric models (henceforth general circulation models, GCMs, for concreteness) must make approximations to reduce the computational burden of predicting radiative fluxes. One such approximation made by all radiative parameterizations in GCMs (to our knowledge) is that the specific humidity within clouds equals the gridbox mean specific humidity rather than the saturated specific humidity. Thus GCMs neglect the sub-gridscale distribution of water vapor caused by clouds.

The sub-gridscale $H2O$ distribution caused by clouds is neglected for two reasons. First, the alternative requires multiple forward radiative transfer calculations. The ideal method for computing radiative fluxes for a column containing a single layer cloud would be to perform two radiative transfer calculations: one for the saturated, cloudy fraction of the gridcell, and one for the sub-saturated clear portion of the gridcell and weight the resulting fluxes by the respective areas of each. This time consuming procedure for one cloudy layer is the simplest example of the independent pixel approximation (IPA). However, this method becomes exorbitantly expensive for GCMs as the number of cloudy layers increases because a separate radiative calculation must be perform for each vertical overlap geometry matching the cloud overlap criteria (e.g., maximum overlap, random overlap) [e.g., Chou et al., 1998]. The second reason is that the effects of cloud scattering and absorption are much larger than the error caused by the approximation.

Crisp [1997] estimates that 20% of in-cloud absorption in a typical mid-latitude summer stratocumulus cloud is caused by vapor pressure difference between the cloud and its environment. This is consistent with Fung and Ramaswamy [1999] who studied the dependence of this enhanced water vapor absorption on cloud liquid water path and droplet effective radius. Some of this enhanced absorption is offset by reduced water vapor absorption in the now-drier environment (which has supplied the vapor required to saturate the clouds). However this offset is very small because much of the increased transmission through the sub-saturated environment is absorbed lower in the atmosphere.

Based on these studies and our own calculations, we expect that neglect of the sub-gridscale $H2O$ distribution contributes 1–2 W m$-2$ to the global and annual solar radiative budget discrepancy. Moreover, this enhanced absorption mechanism causes models to systematically underpredict in-cloud absorption and so contributes to cloudy sky solar absorption discrepancies observed at global scales [Cess et al., 1995; Wild et al., 1995; Collins, 1998]. Possible dynamical consequences of the bias are discussed in § 4.6.

Longwave absorption and emission are also affected by the sub-gridscale $H2O$ distribution in cloudy atmospheres. Water vapor transmission (and thus absorption) vary approximately exponentially with the pressure-weighted water vapor mass path. Thus the linear rearrangement required to segregate gridbox mean $H2O$ into saturated clouds and sub-saturated environments for radiative purposes will cause non-linear responses in the transmission and absorption, and hence longwave radiative fluxes. Because GCM clouds are not treated as saturated, they are less emissive at cloud top and cloud top cooling is artificially reduced, as is heating of the layer immediately above cloud top. This systematically reduces the outgoing longwave radiation in cloudy atmospheres. Our preliminary calculations show this artificial enhancement of the greenhouse effect is typically order 1 W m$-2$ for medium to high clouds. Dynamical implications of this bias are discussed in § 4.7.

4.3.1 Water Vapor Continuum Absorption

GCMs do not include contributions to solar absorption from the far wings of water vapor lines. This continuum absorption may account for up to 1 W m$-2$ [Crisp, 1997], and is adequately constrained by observational studies [Vogelmann et al., 1998]. We will extend water vapor vapor continuum effect into the near infrared using the most the most recently updated continuum of Clough et al. [1989].

4.3.2 Water Vapor Dimer Absorption

The water vapor dimer $\left(H2O\right)2$ may cause significant structured absorption at visible wavelengths [Chýlek and Geldart, 1997]. The absorption cross section of $\left(H2O\right)2$, $\alpha \left(H2O\right)2$, is very difficult to measure so researchers have instead resorted to ab initio models for determining $\alpha \left(H2O\right)2$ [Tso et al., 1998]. The predicted $\alpha \left(H2O\right)2$ are very uncertain in exact absorption band location, but less so in integrated absorption strength. Based on these values, Zender and Chýlek [1998] showed the global annual mean absorption from $\left(H2O\right)2$ was about 1 W m$-2$. However, high precision field measurements [Daniel et al., 1999, S. Clough, personal communication, 1998] cast doubt on the validity the Tso et al. [1998] $\alpha \left(H2O\right)2$. It would be premature to include $\left(H2O\right)2$ absorption in production climate prediction models until uncertainty in $\alpha \left(H2O\right)2$ is reduced and the $\alpha \left(H2O\right)2$ employed can be observationally validated. We remain open to quantifying $\left(H2O\right)2$ absorption should these advances occur.

4.4 Absorption by Collision Complexes of Oxygen

Three recent high quality laboratory experiments directly measured the binary cross-sections for $O2\cdot X$ absorption, $\sigma bO2\cdot X$ [Greenblatt et al., 1990; Newnham and Ballard, 1998; Smith and Newnham, 2000]. In addition, measured $\sigma bO2\cdot X$ has been inferred and evaluated in two high-precision field studies [Solomon et al., 1998; Mlawer et al., 1998].

Zender [1999a] used the available measurements to study the regional, vertical, seasonal, and annual patterns of $O2\cdot X$ abundance and radiative forcing in a general circulation model (GCM). He showed that $O2\cdot X$ absorbs about 1 W m$-2$ globally annually averaged, with no significant difference between clear and cloudy skies. $O2\cdot X$ absorption has strong vertical and regional gradients. These are due to the quadratic dependence of the heating on concentrations of $O2$ and $N2$ (i.e., air density), and to the regional distributions of reflective surfaces (e.g., stratus clouds, snow/ice, and bright desert).

The results already obtained for $O2\cdot X$ forcing of climate illustrate some of the vertical, seasonal, and geographic characteristics we will examine for $H2O$ and $NO2$ heating biases. Figure 2a shows the simulated JJA average radiative forcing of $O2\cdot X$.

Surface albedo and cloud vertical location play a stronger role than anticipated in modulating enhanced absorption by $O2\cdot X$. These factors will also be very important for enhanced $H2O$ and $NO2$ absorption. The climate response to $O2\cdot X$ forcing remains unknown, although from the forcing pattern it appears likely that the strongest response will occur in summertime polar regions.

4.5 Absorption by Nitrogen Dioxide

Accurate, high spectral resolution $NO2$ cross-sections $\sigma NO2$ [Harder et al., 1997] allow $NO2$ absorption to be computed when its distribution is known. Long term measurements of stratospheric $NO2$ concentration are available from the SAGE measurements. However, there are no credible estimates of global forcing by $NO2$ yet because its tropospheric distribution is highly variable and uncertain [Emmons, 1997]. Significant sources of $NOx$ to the troposphere include pollution events, soils, and lightning [Seinfeld and Pandis, 1997; Price et al., 1997], all of which are sensitive to anthropogenic influences [Solomon et al., 1999].

Solomon et al. [1999] demonstrated that $NOx$ produced by lightning and pollution can significantly increase the atmospheric absorptance in cloudy as well as clear skies. In the presence of $O3$, $NOx$ quickly establishes a photochemical equilibrium between $NO$ and $NO2$. Absorption by $NO2$ from lightning and pollution caused peak instantaneous noontime radiative forcings of 5–30 W m$-2$ in Boulder, Colorado [Solomon et al., 1999]. The forcing of $NO2$ from lightning is strongly sensitive to the vertical distribution of the $NO2$ relative to cloud top.

4.6 Climate Sensitivity to Enhanced Absorption

The climate sensitivity to enhanced absorption will remain unknown until and unless causal mechanisms are identified. Some qualitative effects are known from physical reasoning and climate sensitivity experiments. Enhanced absorption introduces an artificial bias where models allow solar energy to be absorbed by the surface instead of in the atmosphere. This artificially heats the surface, destabilizes the boundary layer, and enhances sensible and latent heat transfer. Climate sensitivity studies to prescribed enhanced absorption in clouds show significant responses in the latent heat and wind fields, especially in the tropics [Kiehl et al., 1995]. Fully coupled simulations show enhanced absorption in clouds improves tropical sea surface temperature and latent heat predictions relative to tropical observations [Collins, 2000]. In our experiments of climate sensitivity to enhanced gaseous absorption by $H2O$, $O2\cdot X$, and $NO2$, we will look for similarities and differences with these previous studies.

4.7 Objectives and Hypotheses

1. Objective: Integrate all significant gaseous absorption from $H2O$, $O2\cdot X$, and $NO2$ into our understanding and representation of the solar energy budget.
   Hypothesis: Enhanced solar absorption by $H2O$, $O2\cdot X$, and $NO2$ is 3–8 W m$-2$ 
   Forcing by $O2\cdot X$ is about 0.75–1.2 W m$-2$ [Zender et al., 1997] so the large uncertainty in the total range is due to the unpredictable magnitudes of absorption due to $H2O$ and $NO2$. Accounting for saturated $H2O$ in clouds increases total solar absorption predictably for a given column [Crisp, 1997; Fung and Ramaswamy, 1999]. However, this effect must be integrated over the global vertical distribution of clouds and other radiative boundary conditions. The extent to which total water vapor absorption is vertically redistributed, rather than increased, is unpredictable.
2. Objective: Quantify the role of clouds in increasing or diminishing enhanced absorption by $H2O$, $O2\cdot X$, $NO2$
   Hypothesis: Saturated in-cloud $H2O$ and $NO2$ will preferentially increase cloudy sky absorption relative to clear sky absorption
   Zender [1999a] showed that, globally averaged, $O2\cdot X$ enhances absorption equally in clear and in cloudy skies. Two key mechanisms in the proposed study, however, concern absorption by trace gases in (and near) clouds. Saturated water vapor in clouds, in conjunction with $NO2$ from thunderstorms, are both likely to enhance cloudy sky absorption relative to clear sky absorption.
3. Objective: Compare and contrast the regional, seasonal, and vertical characteristics of enhanced absorption by $H2O$, $O2\cdot X$, and $NO2$
   Hypothesis: Enhanced solar heating by $H2O$ will peak in the tropics near 600 mb
   Like $O2\cdot X$ (Figure 2b), $H2O$ concentration decreases monotonically from the surface, but $H2O$ is concentrated in the tropics. Cloud screening ensures that virtually all of the enhanced $H2O$ absorption will occur at cloud top, so the peak $H2O$ absorption bias should be above the peak $O2\cdot X$ bias. The geographic distribution of tropospheric $NO2$ forcing will be quite distinct from $H2O$ and $O2\cdot X$. Tropospherice $NO2$ forcing should show large land-sea, urban-rural, and tropical-extratropical contrasts due to the distributions of lightning, soils, and pollution.
4. Objective: Investigate the dynamical response to absorption by the (currently neglected) saturated water vapor and $NO2$ within clouds
   Hypothesis #1: The longwave effects of in-cloud saturation suppress the vertical development of deep convection
   The enhanced concentration of water vapor within clouds increases their emissivity, causing cloud tops to cool to space more efficiently. Moreover, the reduced $H2O$ concentration in the surrounding sub-saturated environment is more transmissive to thermal radiation from below. In conjunction these effects enhance cooling at cloud top and reduce heating immediately above cloud top and so act to stabilize the vertical region near the cloud top, i.e., to suppress convection.
   Hypothesis #2: The solar effects of in-cloud saturation enhance the vertical development of convection
   Allowing the saturated in-cloud water vapor to absorb shortwave radiation will enhance solar heating in the upper layers of clouds. This will compensate, to an extent, the longwave effects of in-cloud vapor absorption described above. The degree of compensation will depend on cloud type, zenith angle, and vertical location. $NO2$ from lightning is also expected to heat cloud tops and enhance the vertical development of convection.
5. Objective: Quantify the climate response to $O2\cdot X$ heating
   Hypothesis: $O2\cdot X$ increases the temperature at the polar tropopause by $\sim 1$ K
   Single-column simulations show the radiative relaxation timescale at the polar tropopause is $\sim 50$ days. $O2\cdot X$ heating of the summertime polar tropopause is $\sim 0.01$ K d$-1$ (Figure 2b), which should increase the radiative equilibrium temperature by $\sim 1$ K.
6. Objective: Assess the role of $NO2$ absorption in warming the stratosphere
   Hypothesis: $NO2$ absorption warms the polar stratosphere
   Many GCMs have strong (up to 10–14 K) cold biases at and above the summertime polar tropopause [Hack et al., 1998]. $NO2$ heating will have a maximum in the summer polar stratosphere. Stratospheric air density is very low so even a modest forcing (0.5 W m$-2$) can change its equilibrium temperature.

4.8 Research Plans

All of our objectives (§4.7) depend on successfully importing new radiative processes and procedures for $H2O$, $O2\cdot X$, and $NO2$ into a 3-D atmospheric general circulation model (GCM). The first three of our objectives (§4.7) will be met by analyzing the results of diagnostic GCM forcing simulations of one or two years in length. The GCM will be driven by climatological sea surface temperatures so that the diagnostic forcing by enhanced gaseous absorption is climatologically representative. Objectives 4–6 will require long term (about 15 year) feedback experiments with AMIP SSTs as well as in a fully coupled Climate System Model. Long term integrations are essential to isolating the climate response signal from the noise of natural variability, especially in polar regions.

Finally we would like to assess the change in enhanced absorption of of $H2O$ and $NO2$ expected between now and the Year 2100. The influence of a warmer climate on $H2O$ and of anthropogenic and natural changes in $NOx$ sources may cause their absorption to change significantly.

4.9 Methods and Procedures

We will use the NCAR Community Atmosphere and Climate System Models (CAM and CCSM) [Kiehl et al., 1998; Boville and Gent, 1998] for climate forcing and sensitivity studies with enhance $H2O$, $O2\cdot X$, and $NO2$ absorption. To represent the radiative effects of the sub-gridscale distribution of $H2O$ in cloud atmospheres, a radiative transfer procedure must segregate the clear sky humidity field $qc$ from the in-cloud humidity field $qs$ such that total gridcell humidity $\bar{q}$ is conserved. We will modify a new generalized framework for treating cloud overlap developed by W. Collins of NCAR which makes this segregation as economical as possible. For a gridbox with horizontal cloud fraction $AH$, the subgridscale humidity fields will be related by

Since clouds are very near saturation with respect to the cloud temperature $T$ theoretically $qs=qs\left(T\right)$. In practice, however, GCMs allow $AH=1$ when gridcell relative humidity $\text{ RH}<100\%$ [e.g., Kiehl et al., 1998]. In other words, a horizontally overcast gridbox is not necessarily saturated through its entire volume. This small complication to (1) should be easily surmountable.

A sophisticated, high resolution (10 $\text{ cm}-1$) offline radiative transfer model [Zender et al., 1997] will be used to parameterize $O2\cdot X$ and $NO2$ absorption cross-sections for use in the GCM. In order to quantify the radiative forcing due to $NO2$ one needs to know its distribution in the troposphere. This appears to be the most difficult portion of the entire project. As a first step we will use the global monthly daytime mean $NO2$ concentration dataset predicted by the MOZART Chemical Transport Model (CTM) [Brasseur et al., 1998]. MOZART simulations account for emissions of $NOx$ by lightning and pollution. These simulations have been evaluated against the vertical and diurnal structure of $NOx$ in many regions [Emmons, 1997; Brasseur et al., 1998]. Although spatially complete, these monthly data will not resolve the diurnal signal of $NO2$ established by its photochemical balance with $NO$. A simple diurnal cycle can easily be imposed on the monthly data based on the sun angle.

There are significant problems with using $NO2$ distributions from a CTM climatology. Among them is that the vertical location of $NOx$ within a cloud is crucial to determining the radiative forcing of $NO2$ [Solomon et al., 1999]. Unfortunately the vertical structure of $NOx$ from lightning is not well understood [Price et al., 1997] so a CTM cannot be expected to represent it correctly. A vast improvment to using pure CTM model data would be to use a blend of available satellite, in situ, and model data. Groups are attempting to assemble such climatologies [Solomon, 1999] and we will obtain these improved datasets when available to refine our estimates of the global radiative impact of $NO2$ on climate.

4.9.1 Model Evaluation

Improvements to Earth’s radiative budget will be evaluated against a combination of in situ [Gilgen and Ohmura, 1999] and satellite observations. The Earth Radiation Budget Experiment (ERBE) [Ramanathan et al., 1989] currently provides the gold standard in long term top of atmosphere (TOA) solar and longwave radiative fluxes. We note that reduction of model biases relative to observations may not occur without first re-tuning some free parameters of the GCM. During the course of this research, data from the EOS CERES experiment [Wielicki et al., 1996] will improve and extend the ERBE record, offering vertical information on heating as well.

4.10 Expected Significance of Results

Taken individually, these enhanced absorption processes by $H2O$, $O2\cdot X$, and $NO2$ may appear to be minor. However, we expect the global annual mean absorption by these processes to total 3–8 W m$-2$. Thus we expect to reduce the current discrepancy in the atmospheric energy budget of 20–25 W m$-2$ [Yu et al., 1999] by about 15–30%. If we are quick then our CCSM/CAM climate simulations will be the first in the world to realistically account for this large a fraction of the enhanced absorption discrepancy. We hope to maintain CAM’s leading edge position by soon adding aerosol absorption as well [Collins et al., 2002].

4.11 Synergies with Existing Research Efforts

W. Collins (NCAR) is working on improved parameterizations of cloud overlap effects and of water vapor continuum absorption. The segregation of water vapor absorption into clear and cloudy sky components is a natural extension of his generalized framework for treating cloud overlap.

S. Solomon (NOAA) and colleagues are currently assembling a four-dimensional $NO2$ climatology from satellite (SAGE, GOME) and in situ observations [Solomon, 1999]. Their $NO2$ climatology will be more accurate than the MOZART CTM model predictions of $NO2$ we will start with. We will adopt such an empirical $NO2$ climatology as soon as possible.

4.12 Benefits to Community

The climate-sensitivity portion of this research will be performed in, and contributed to the NCAR CCSM. My previous research on $O2\cdot X$ forcing [Zender, 1999a] is now being incorporated into the Community Atmosphere Model (CAM). Everyone who uses CCSM simulations will ultimately benefit from the improvements contributed by this project.

4.13 Significance to Professional Goals and Responsibilities

This research plan firmly advances my personal career goal to help close Earth’s energy budget and to reduce the associated uncertainty in climate predictions. This proposal on gaseous absorption nicely balances my research on mineral dust aerosol radiative forcing, since aerosols and trace gases are the only significant known absorption mechanisms not accounted for in climate predictions.

My goal as a faculty member in ESS is to create a research group that regularly contributes expertise in radiative forcing issues of interest to national and international climate modeling and assessment programs.

4.14 Work Plan for Research

The research will take place in phases which are interleaved to synergize with other research projects. The tentative goals for each year are

• Assess climate response to $O2\cdot X$ forcing. Study $NO2$ forcing based on MOZART climatology.
• Study climate response to $NO2$ forcing. Assess forcing sensitivity to more realistic empirical $NO2$ climatology. Study forcing due to in-cloud water vapor saturation.
• Study climate response to in-cloud water vapor forcing. Examine solar forcing due $H2O$ continuum absorption in near-infrared. Add $\left(H2O\right)2$ processes if $\alpha \left(H2O\right)2$ are available.
• Evaluate climate sensitivity to total enhanced $H2O$, $O2\cdot X$, and $NO2$ absorption. Evaluate other gaseous absorption processes for possible refinements.
• Finish evaluations of climate sensitivity to total enhanced $H2O$, $O2\cdot X$, and $NO2$ absorption. Evaluate change in forcings due to Year 2100 conditions.

4.15 Prior Research Accomplishments

1. Radiative Effects of Tropical Cirrus Anvil on Climate: We first documented the effect of ice crystal size and habit on anvil formation and radiative properties [Zender and Kiehl, 1994]. An important conclusion from this study was that shortwave radiative properties of tropical cirrus anvils are very sensitive to the presence of small ($3<L<20$ $\mu \text{ m}$) ice crystals which account for less than $2\%$ of cloud mass. On a global scale, these anvil physics contribute to teleconnections between tropical anvil heating and the extratropical circulation [Zender and Kiehl, 1997].
2. Observational Detection of Enhanced Shortwave Absorption in Clouds: The discrepancy between models and observations of Earth’s atmospheric shortwave energy budget is up to 25 W m$-2$, globally annually averaged [Cess et al., 1995; Li et al., 1997; Yu et al., 1999]. The ARESE experiment measured the solar energy budget in clear and cloudy conditions using stacked aircraft observations. Detailed spectral analysis of the results shows significant evidence of enhanced shortwave absorption in cloudy atmospheres [Zender et al., 1997].
3. Radiative Forcing by Oxygen Collision Complexes: Scrutiny of the magnitude and causes of enhanced shortwave absorption has led to many interesting discoveries, including the recognition that absorption by collision complexes of oxygen is significant globally (about 1 W m$-2$). Zender [1999a] produced the first climatology of collision complexes and their absorption and illustrated many unique features of collision complex absorption. This study refuted the notion that $O2\cdot X$ contributes to enhanced cloudy absorption (globally).
4. Mineral Dust Aerosol Transport and Radiative Forcing: Globally mineral dust accounts for more aerosol mass and surface area. We have developed a mineral dust aerosol model [Zender, 1999b] and included it in a global aerosol assimilation model with an integrated aerosol suite (sulfate, dust, sea-salt, carbon) [Collins et al., 2001,  2002]. This assimilation scheme helped INDOEX flight operations forecast targets for aerosol missions. We are now applying this assimilation model to help constrain global aerosol radiative forcing.
5. Radiative Forcing and Absorption by Boundary Layer Aerosol in the US: An important task in understanding Earth’s atmospheric energy budget is to constrain models with long term observations of aerosol radiative forcing. In collaboration with a group making long term station observations of boundary layer aerosol, we have produced the first estimate of the direct radiative forcing by tropospheric aerosols from in situ observations in the southeastern US on an annual timescale [Yu et al., 2001].

5 References

6 Curriculum Vitae

Department of Earth System Science zender@uci.edu
University of California Voice: (949) 824-2987
Irvine, CA  92697-3100 Fax: (949) 824-3256

EDUCATION

0. (1996) Atmospheric Sciences, University of Colorado, Boulder. “Representation of tropical cirrus anvil in climate models”, Advisors: Jeffrey Kiehl and Gary Thomas
0. (1993) Atmospheric Sciences, University of Colorado, Boulder.
0. (1990) Physics, Harvard University

SPECIALTIES AND INTERESTS

Atmospheric Physics, Cloud and Aerosol Microphysics and Chemistry, Terrigenic Aerosol, Particle Composition and Optical Properties, Radiative Transfer and Radiative Forcing, Global Climate and Climate Change

PROFESSIONAL APPOINTMENTS

0. University of California at Irvine – Assistant Professor of Earth System Science
0. National Center for Atmospheric Research (NCAR), Boulder, CO – Affiliate Scientist of the Climate and Global Dynamics (CGD) Division
0. NCAR – Visiting Scientist in Atmospheric Chemistry and CGD Divisions
0. NCAR – Postdoctoral fellow in Advanced Study Program
0. University of Colorado at Boulder and NCAR CGD – Graduate research assistant
0. College of the Atlantic, Bar Harbor, ME – Visiting Faculty in Physical Sciences
0. Smithsonian Astrophysical Observatory, Cambridge, MA – Programmer, Technician

REFEREED PUBLICATIONS

Zender, C. S. and J. T. Kiehl, Radiative sensitivities of tropical anvils to small ice crystals, J. Geophys. Res., 99, 25869–25880, 1994.

Zender, C. S., B. Bush, S. K. Pope, A. Bucholtz, W. D. Collins, J. T. Kiehl, F. P. J. Valero, and J. Vitko Jr., Atmospheric absorption during the Atmospheric Radiation Measurement (ARM) Enhanced Shortwave Experiment (ARESE), J. Geophys. Res., 102, 29901–29915, 1997.

Zender, C. S. and J. T. Kiehl, Sensitivity of climate simulations to radiative effects of tropical anvil structure, J. Geophys. Res., 102, 23793–23803, 1997.

Cess, R. D., M. Zhang, F. P. J. Valero, S. K. Pope, A. Bucholtz, B. Bush, C. S. Zender, and J. Vitko Jr., Absorption of solar radiation by the cloudy atmosphere: Further interpretations of collocated aircraft measurements, J. Geophys. Res., 104, 2059–2066, 1999.

Zender, C. S., Global climatology of abundance and solar absorption of oxygen collision complexes, J. Geophys. Res., 104, 24471–24484, 1999.

Collins, W. D., P. J. Rasch, B. E. Eaton, B. Khattatov, J.-F. Lamarque, and C. S. Zender, Forecasting aerosols using a chemical transport model with assimilation of satellite aerosol retrievals: Methodology for INDOEX, In Press in J. Geophys. Res., 2000.

Yu, S., C. S. Zender, and V. K. Saxena, Direct radiative forcing and atmospheric absorption by boundary layer aerosol in the southeastern US: new observational estimates and model results, Submitted to Atmos. Environ., 2000.

FUNDING

• Co-I on NASA grant “Effects of land-use on climate and water resources: application of a land surface model for land-use management”, PI: G. B. Bonan, 1/1/2000–1/1/2003

SERVICE

• Peer-review for Geophys. Res. Lett., J. Geophys. Res., J. Atmos. Sci., Mon. Weather Rev., Nature, Q. J. R. Meteorol. Soc., Tellus, NSF, NASA, USGCRP
• Maintainer of NCAR CCM Column Radiation Model (http://www.cgd.ucar.edu/cms/crm). 1996–present.
• Author and administrator of NCO netCDF Operators (http://www.cgd.ucar.edu/cms/nco), a freely available geophysical data manipulation toolkit. 1995–present.
• Author and maintainer of Enhanced Absorption Bibliography (ftp://ftp.cgd.ucar.edu/pub/zender/arese/bib_aca.ps.gz). 1997–present.
• Contributor to the University of Northern Colorado Mathematics and Science Teachers Hotline (MAST) (800 866-MAST). 1995–present.
• University of Colorado, Boulder, CO – Founder and director of the APAS Help Center, a free tutoring center which assisted dozens of diverse undergraduates in the physical sciences every year. My role included securing funding, tutoring, and managing a group of about 10 paid graduate student tutors. 1992–1995.

HONORS

• Outstanding Student Presentation in Atmospheric Sciences Section, Fall AGU Meeting, San Francisco CA, 1995

COURSES TAUGHT

• Earth System Science 20E: The Atmosphere
• Earth System Science 111/211: Radiative Processes and Remote Sensing

COLLABORATORS

• C. A. Ammann (U. Massachusetts Amherst), G. B. Bonan (NCAR), G. P. Brasseur (MPI Hamburg), R. D. Cess (SUNY Stonybrook), P. Chýlek (Dalhousie), W. D. Collins (NCAR), J. T. Kiehl (NCAR), N. M. Mahowald (UC Santa Barbara), G. McFarquhar (NCAR) K. Oleson (NCAR), P. J. Rasch (NCAR), X. X. Tie (NCAR), F. P. J. Valero (Scripps), S. Yu (Duke)