Project description

Research Activities

Land use and land cover change and human-induced changes in atmospheric composition and climate are the driving issues of global change research. Because of many complex interactions between chemical, physical, ecological, and social systems, rational decision making and policy analyses for global and regional environmental change issues benefit from an integrated environmental assessment (IEA) approach. We are currently developing an integrated model of land use and land cover change for Amazonia (Fig. 1). This model will be incorporated into the Integrated Global System Model (IGSM, Prinn et al. 1996, 1997, Xiao et al. 1997a), so that we can address the coupling between land use/cover change, hydrology, biogeochemistry and changes in atmospheric chemistry, climate and socioeconomic factors (Fig. 2).

Remote sensing plays an essential role in providing input and validation data to the models and in the development and application of the model framework for integrated environmental monitoring, research and assessment. Our strategy for the application of remote sensing to regional and global IEA is to (1) expand the use of data from existing satellite platforms (e.g., Landsat TM, SPOT and AVHRR), and (2) develop the capacity to utilize data from the TRMM, EOS AM-1 and other EOS platforms when they are available. The results will be incorporated into an open GIS-model system which has the capability of (1) interactively linkage between GIS-managed data layers and models, (2) simultaneous display, interpretation and analysis of spatial input and output data of the models, (3) plug-in and play of the models provided by other research groups. The system will facilitate the data fusion from various satellite sensors, socioeconomic statistics, field survey information, site data and models.

Our approach includes two primary efforts. First, we will focus on the linkages between remote sensing and the components of IEA , e.g., land cover, fire, seasonal and interannual climate variations, biogeochemistry, atmospheric chemistry, climate, and socioeconomics (primarily human settlement and infrastructure). Second, we will apply remote sensing products to conduct error analyses of the IEA framework across various temporal and spatial scales. For AVHRR , TRMM and EOS AM-1, we will generally use satellite data that have been undergone radiometric calibration, atmospheric correction, and geometric registration; our usage of the new computing facility will focus on the analysis and application of remote sensing data to support research in regional and global IEA. In the following sections, we outline our research activities for integrating remote sensing into the IEA framework and associated resource needs. These research efforts in model development in UNH, MBL and MIT are funded by a number US governmental agencies (NASA, NSF, DOE, EPA) and an international consortium of industrial sponsors for the MIT IGSM.

Task 1. Land cover change

Satellite remote sensing provides critical observations of current and historical land cover changes. Landsat MSS, Landsat TM, SPOT and AVHRR have been widely used for land cover classification. The EOS AM-1 will provide novel, and more accurate information for land cover classification and change detection. The MODIS will cover the globe every 1 to 2 days and have 36 spectral channels with 250-m, 500-m or 1-km spatial resolution, thus more spectral information can be used for land cover classification, including directional surface reflectance, texture, vegetation index, acquisition geometry, land surface temperature, and snow/ice cover.

Recent efforts in continental and global scale land cover classification and change detection have relied primarily on the NDVI from the NOAA AVHRR (Loveland et al. 1991; Nemani and Running 1997; the NOAA/NASA AVHRR Pathfinder project). The standard MODIS land cover product from the EOS-DIS will include land cover types and land cover changes at 1-km resolution for the globe. There will be a total of 17 categories of land cover, following the IGBP global vegetation database (9 classes of natural vegetation, 3 class of developed lands, 2 classes of mosaic land and 3 classes of nonvegetated lands, i.e., snow/ice, bare soil/rocks and water). This level of land cover classification is not sufficient to support our biogeochemistry models. The TEM model currently differentiates between 18 upland ecosystem types and 13 floodplain and wetland ecosystem types (Melillo et al. 1993). The DNDC model uses information specific to cropland type (e.g., wheat, corn, rice), and distinguishes between grassland and pasture. Spatial extrapolation of the CH4 emission model also needs better spatial representation of wetland and seasonal inundated floodplain, which have a large uncertainty in their areas and spatial distribution in part due to rapid human land use change. Therefore, we will process 1-km multi-spectral and multi-temporal MODIS data to generate the needed classes of natural vegetated ecosystems and agricultural lands. The resultant 1-km global land cover product will be used (1) to support the development of the Dynamic Vegetation and Ecosystem Model (DVEM) and integrated modeling of land use and land cover change (Xiao et al. 1997c; Tian et al. 1995; Hall et al. 1995a,b); and (2) to derive 0.5ƒx 0.5ƒgridded mosaics to perform TEM and DNDC simulations. The processing of global 1-km data for land cover classification requires both enormous data storage and computing power as conventional multivariate supervised classifier such as maximum likelihood, unsupervised classifier (clustering algorithm and minimum distance classification), and the neural network approach are all CPU intensive. The availability of the new computing facility is essential for us overcome the constraints of available computers and mass storage in the existing laboratory.

The success of the above effort at the global scale require benchmark research on remote sensing and field observation at site and regional scales. Researchers at MBL and UNH are actively involved in the NSF Long Term Ecological Research sites (Harvard Forest site, Arctic tundra site, Hubbard Brook Forest site) and the NSF Land Margin Ecosystems Research sites (Plum Island Sound). We have also submitted a proposal to NASA LBA-ECOLOGY program. Therefore, we propose to use the new computing facility to develop a prototype for data fusion across various satellite sensors (including SPOT, Landsat TM, AVHRR, MODIS, MISR, active microwave sensors) at these LTER and LMER sites and the primary sites of the LBA project in Amazonia. We will explore spatial distribution of various stages of successional vegetation, and vegetation classification at community level which are important to assess biodiversity and enhance our understanding on the relationship between biodiversity and ecosystem functions. This effort will support the NASA Global Land Cover Test Site project. Although most of these sites are covered by one Landsat TM scene, the processing of image data over multi-years from various sensors still need moderate data storage and CPU time. It also requires substantial field work and use of the global position system (GPS).

Task 2. Fire

Fire occurrence represents a fundamental disturbance to land cover and soil nutrient status, and is associated with emissions of trace gases (e.g., CO2) and pollutant aerosols. In the process of shifting cultivation for cropland and pasture, fire is a major agent of land conversion because of its low cost. At present, fire monitoring and assessment have primarily relied on the use of NOAA AVHRR and GOES systems (Koffi et al. 1995; Kennedy et al. 1994; Kasischke et al. 1993), which have significant limitations due to their low saturation temperature (about 33 (C or 306 K). The MODIS sensor on the EOS AM-1 includes characteristics specifically for fire detection (channel saturates at 500 K) and burn scar detection, thus providing a unique capability for fire monitoring. The fire product will have global coverage at daytime and nighttime, and will include important ancillary information (e.g., fire class, smoldering and flaming ratio).

We will incorporate the MODIS fire product (e.g., fire frequency, intensity, and spatial extent) at 1-km resolution into our land use change model. The dynamic vegetation/ecosystem model (DVEM) also will use this information to indicate the timing and location of fire-driven secondary succession events. Biomass burning data, which provide information on carbon loss and particulate emissions from terrestrial ecosystems to the atmosphere, will be incorporated into atmospheric chemistry model. This effort will require manipulation of daily global 1-km images in order to incorporate the fire information into the models; an incremental improvement in the UNH lab would make this analysis possible.

Task 3. Climate variability

Seasonal and interannual climate variability (e.g., the ENSO cycle), significantly affects natural and managed terrestrial ecosystems. Temperature, precipitation and solar radiation are key abiotic driving variables to biogeography models and biogeochemistry models (VEMAP Member 1995). The TEM and DVEM use monthly climate fields, while the PnET (Aber et al. 1995) and DNDC models use daily climate fields. At present, spatial data sets of climate are mostly generated using various approaches for interpolating point data from weather stations. There is much uncertainty about the climate data quality and spatial representation, particularly cloudiness data, which leads to significant uncertainty in estimates of net primary production (NPP) of terrestrial ecosystems (Pan et al. 1996). For instance, TEM currently uses monthly cloudiness data to calculate photosynthetic active radiation (PAR) and net irradiance (NIRR). PAR is used to calculate gross primary production (GPP, Raich et al. 1991) and NIRR is used to calculate potential evapotranspiration (PET, Vorosmarty et al.1989). There is a large uncertainty about the estimates of cloudiness from weather stations because of spatial undersampling and non-standard instrumentation.

We will use data from the TRMM and MODIS to apply and generate the climate fields required by our biogeochemistry and biogeography models. The MODIS land surface temperature product includes daily and monthly values at 1-km to 5 km to 0.5ƒ spatial resolution for the global land surface. Data from the TRMM and MODIS can be used to generate monthly precipitation data at a much coarser, but still useful, spatial resolution (1ƒ(1ƒ). EOS AM-1 and TRMM will also be used to generate monthly NIRR and PAR at 1ƒ(1ƒ spatial resolution for the globe.

Task 4. Biogeochemistry

The regional and global extrapolation of TEM and DNDC will provide estimates of carbon and nutrient fluxes and pools of terrestrial ecosystems at regional to global scales over time. Satellite remote sensing plays an essential role in evaluating the model estimates of NPP across various temporal and spatial scales. Schimel et al. (1997) used the relationship between NDVI and model-estimated NPP for the conterminous US to help characterize model limitations at the process level.

To evaluate the spatial patterns and temporal dynamics of NPP predicted by TEM and DNDC, we will conduct two comparisons using remote sensing data. First, we will compare NPP estimates by TEM and DNDC with (1) NDVI data from AVHRR (e.g., the Global Land 1-km AVHRR product from April 1, 1992 to September 30, 1996) and (2) NDVI generated from MODIS which provides daily NDVI and Modified Vegetation Index (MVI) at 250 m to 1-km spatial resolution for the globe. We will employ spatial statistics and time-series analysis methods to compare the model NPP output with NDVI. Second, we will compare NPP estimates by TEM and DNDC with NPP estimates by the satellite-based production-efficiency approach (Prince and Goward 1995; Goward et al. 1994) which has focused on the empirical and theoretical relationship between annual NPP (ANPP) and annual integrated PAR (APAR. A simple production-efficiency model has the following form: ANPP = ( ( ( APAR and APAR = FPAR ( PAR, where FPAR is the fraction of PAR that is absorbed by the vegetation canopy. Empirical observations have supported the notion of a positive relationship between FPAR and NDVI (Potter et al. 1993; Ruimy et al.1994), at least within functional ecosystem types. The above comparison will delineate spatial and temporal regions of inconsistency between NPP predictions of TEM/DNDC and the remotely sensed proxy. Analyses of these inconsistencies will highlight possible areas of for further improvement in model calibration, process-level details, or input data.

The new facility will provide large data storage and powerful CPU for us to explore the linkage between remote sensing and biogeochemistry modeling. We have already begun to explore how to implement mass parallel computing of TEM model.

Task 5. Atmospheric chemistry

The atmospheric chemistry model in the IGSM (Fig. 2) currently includes 25 chemical species including CO2, CH4, N2O, CFCs, O3, H2O, NOx, HOx, SO2 and sulfate aerosols. The model simulates 53 gas-phase and aqueous phase chemical reactions, including O3-HOx-NOx-CO-CH4 reactions in the troposphere (Crutzen and Zimmermann 1991). Surface observational data of CFCs, N2O, CH4, CO, and CO2 from the ALE/GAGE/AGAGE (Prinn et al. 1990) and NOAA/CMDL (Novelli et al. 1992) networks have been used to validate the modeled results for current years. Coupled with the climate model in the IGSM, the atmospheric chemistry model has been used to simulate evolution and radiative forcing of chemical species in the atmosphere (Wang et al. 1995, 1997; Prinn et al. 1996, 1997).

The MOPITT measures pollution in the troposphere and will provide horizontal and vertical measurements of CO and CH4. It will provide global data for CO and CH4 at daily temporal resolution and at 22 km to about 100 km spatial resolution. Measurement of CH4 from the EOS AM-1 platform will be useful in resolving spatial and temporal variation in CH4. We will use these high spatial-temporal resolution data of CO and CH4 at appropriate level of aggregation to test the atmospheric chemistry models. We will also use these observation data to study tropospheric oxidizing capacity and evolution of OH radicals in the atmosphere.

Task 6. Physical Climate modeling

The spatial distributions and temporal dynamics of clouds and aerosols are important factors that must be considered by climate models. Aerosols from the eruption of Mt. Pinatubo (June 1991) reduced solar irradiance by 4% and decreased global mean temperatures by about 0.5ƒ C for about a year following the eruption (Blumthaler and Ambach 1994). The reduced-form climate model in the IGSM (Fig. 2) resolves two fundamentally different types of clouds: those associated with moist convection, and large-scale or supersaturated formations due to large-scale condensation (Solokov and Stone 1997). The model also takes into account changes in anthropogenic aerosol forcing (e.g., sulfate), which is mostly concentrated over land, and is particularly significant at the northern mid-latitudes (Prinn et al. 1997).

We will process the MODIS and MISR cloud and aerosol data to support the climate modeling. The MODIS will provide daily and monthly cloud data at 1 km to 0.5ƒ spatial resolution for the globe, and daily and monthly data of atmospheric aerosol optical depth (global) and aerosol size distribution (oceans) at 10 km to 0.5ƒ spatial resolution for land and oceans, separately. The cloud and aerosol data, when averaged over latitudinal zones, will be used for validation and development of the coupled atmospheric chemistry/climate model (Fig. 2).

Satellite observations also provide information about the biophysical status of the vegetation canopy, including the fraction of PAR (fPAR) absorbed by plant canopy, leaf area index (LAI), vegetation biomass density. The LAI is a key parameter in determining surface roughness length, total albedo, and precipitation interception rate. These biophysical parameters are essential in the coupling of land surface and atmosphere for regional climate models (Pielke et al. 1997) and atmospheric GCMs (Fennessy and Xue 1997). We will use both empirical algorithms and physically-based algorithms (e.g., canopy radiative transfer model; Braswell et al. 1996) to extract information of these biophysical variables. This satellite-derived information will also improve our ability to characterize the biogeochemical state of terrestrial ecosystems.

Task 7 . Socioeconomic Factors

Although most social and economic variables (e.g., income, prices) are not observable by remote sensing, satellite remote sensing can easily monitor the spatial distribution of human settlements (towns and cities) and infrastructure (roads, dams, reservoirs) over time. NOAA is using satellite data from the Defense Meteorological Satellite Program to generate global nighttime images of artificial lighting at 1-km resolution ( http://www.ngdc.noaa.gov for the USA, Central America, and South America). These nighttime images plus data from Landsat TM and SPOT, will provide useful geographical information on urbanization, which, particularly in developing countries, has resulted in significant losses of agricultural lands (Douglas 1994; Grubler 1994). Urbanization may make people more vulnerable to environmental change (e.g., air and water pollution effects on public health). Among various socioeconomic settings, sustainable human settlement (urban and rural) is a key issue related to land use change, air pollution, water resources, desertification, biodiversity, and climate change. We will use EOS AM-1 data to monitor human settlement and infrastructure, in support of land use change modeling, vulnerability and risk assessment of human societies to natural disasters, climate variability (e.g., flood, drought) and climate change.

One of the key factors in determining the success of IEA is to have data fusion between social-economic and remote sensing. We will pursue an integration of socioeconomic data with remote sensing data in an open GIS-model system. Socioeconomic data (e.g., population, price, income, trade) are mostly in political or administrative units (e.g., national, state, county) which generally have irregular shapes (i.e., polygons), while biophysical environment and land cover data (which can be derived from remote sensing) are mostly in a grid format. On the global scale, an effort has been devoted to use county-level population data to generate 5í (longitude) by 5í(latitude) resolution grid data of population (Tobler et al. 1995). At the global scale, we will collect country-level data to support the land use change model (Fig. 1) and the EPPA model (Fig. 2). We will also collect available town to county level socioeconomic data. Our data fusion effort at the global scale requires large data storage and powerful CPU to manipulate data.

Description of the research equipment and needs

In order to support the research and educational activities we have outlined in the Section B, we need to substantially increase our computing facility and instruments. We propose to purchase the following equipment and instruments for the Center of Excellence in Application of Remote Sensing to Regional and Global Integrated Environmental Assessments (Table C.1). The Laboratory of Remote Sensing at the Institute for the Study of Earth, Oceans and Space, University of New Hampshire, which focuses on mapping the current state and rates of change in tropical forests as part of the NASA Landsat Pathfinder Program, currently have the following large equipment and instruments:

• One SGI Challenge L Server: four 150 MHZ Processors (R4400 Processor Chip); 1.2 GBs per second sustained I/O Bandwidth; 512 MBs Main Memory; 12 GBs Internal Disk Space;
• Workstations: 3 Sun SPARC 1s with 8-bit graphics; 4 SGI R4000 Indigos; 2 SGI R4400 Indigo2s
• Data storage: HP 90 GB Optical Juke Box
• Output devices: Apple Laser Color Printer; HP 755 Color Plotter; 8 mm tape drives;

A more in-depth discussion about the equipment and instruments (Table C.1) for the Center of Excellence is presented below.

The new computational server will reduce the time required for processing and manipulating sensor data and the time required to obtain data modeling results. Recent tests at the Institute have shown a 5 to 1 reduction in the time required to process image data with the processor chip that is used on the proposed new server. While the old server's processors can be upgraded (however, not cost effectively), there are other parts of the server that can not be upgraded. For example, the new server's back plane or main computer bus where all the data travels has a significant increase in through put capacity (5 GBs/sec sustained). This, along with a new shared main memory architecture, adds to faster processor performance and thus, faster job turn around times. Image processing and data modeling that has taken a week in the past could be completed in a day or less. Processing speed or job turn around time becomes extremely important as the size and amount of sensor data increases and as the complexity and size of processing and modeling requirements also increase. Also, additional processing requirements will need to be considered as the role of the current remote sensing lab expands to include both increased remote sensing research involving multiple disciplines and projects as well as providing for academic education and training. Another important aspect in server architecture is scalability. The proposed new server can easily scale up to meet future computational needs as it can accommodate up to 128 processors (without a decrease in back plane speeds), accommodate up to 64 GBs of main memory and provide access of up to 864 GBs of internal disk space. Additionally, the server architecture used, symmetric multiprocessing (SMP) architecture, provides the capability to run a separate task on each processor or use several processors to run one task. Our effort in coupling between remote sensing data and models (e.g., TEM, DNDC, land use change model) will benefit substantially from its capability of mass parallel computing.

Two desktop high performance graphics workstations would be invaluable for visual analysis and display of remote sensing data and for running 2 and 3D visual models. The current lab with its singular purpose does not have this capability. These high performance graphics workstations contain dedicated graphics subsystems such as a geometry engine processor rated at 960 MFLOPS, a rasterization engine with a 120M pixel per second fill rate and a 32 bit frame buffer specifically tuned for handling 3D images and texture caching memory. These workstations will support two separate display screens that can be used independently or simultaneously. The two screens can be used to display one image between the two screens such as a geographic terrain image or, for example, continuously display a complex visual model on one screen while interacting with the modeling program on the other screen. This type of technology will greatly enhance the understanding of and research performed using remote sensing data as well as for providing the necessary tools to educate and train new personnel.

At a minimum, 6 data processing and instructional workstations would be required to support both active research and educational activities. These workstations would be used by the researchers and support staff for day to day sensor data processing and for scheduled training, seminars and academic instruction.

The 400 GB RAID disk storage system will meet the needs for having large amounts of research and remote sensing data readily accessible. This system will provide data in a fault tolerant way while providing sufficient disk storage space for current remote sensing data processing, image manipulation and global dataset modeling. This new expandable storage management system will alleviate limitations on research due to a lack of disk space that has been encountered in the past. The system planned is scalable and thus can grow as requirements do. The system will also provide additional security to the data between data backups to preclude loss of man hours and production time due to disk failures. The current lab does not have an adequate mass storage system and does not have a RAID system.

With the large amounts of remote sensing data the lab expects to receive, process and provide access to for research and educational needs, a system that can backup and restore these data from/to magnetic disks as well as provide uninterrupted data availability, reliability and archive support over the long term is critical. The system selected for this purpose does all of this plus its initial storage capacity is expandable from 1.36 TBs to 3.5 TBs which means that it also can scale up as required. This system will make it easy to archive and remove large data sets from the RAID system to make room for new data to be processed and as required these same data sets can easily be restored back to the RAID system for further processing. The current lab has a 5 year old HP juke box system for providing these functions. However, it only has a 90 GB capacity and would be more costly to upgrade than to purchase the new archiving system.

The lab currently has a small format color printer. However, it is not of the quality or capability of the Tektronix Phaser 480X color printer we proposed to purchase. The Tektronix 480X would provide the lab with a large format, 11.9" x 17.2", color printer capable of producing photographic-quality color prints using continuous-tone 300 dpi dye sublimation. Color prints of this size and quality would greatly enhance the ability to accurately display details of the results from processing remote sensing image data and the results from running visualization models for both research and education activities.

The new color plotter has better color resolution, 600 dpi vs. 300 dpi, than the plotter currently in use by the lab. The cost of producing plots is greatly decreased, $0.40 (40 cents) per square foot vs $1.20 per square foot and the maximum plot length is extended from 9 feet to 150 feet.

Global positioning system receivers comprise a key remote sensing instrument used in field work. This system allows investigators to locate positions on the earth with respect to remotely sensed data, in the field work to collect and validate training site locations and to perform post-processing accuracy assessments. The proposed configuration includes a radio transmitter, which will enable investigators to utilize broadcast signals to perform real-time differential corrections on field data. The two GPS receiver system is requested to support differential corrections of GPS coordinates in locations where broadcast signal are unavailable.

The Aerial Videography provides an important source of remotely acquired data at training sites to assist in the processing and interpretation of satellite imagery. As traditionally acquired, there is no spatial reference on the video frame. The proposed system of hardware and software will allow researchers to collect a GPS coordinate in conjunction with the video, and based on the SMPTE timecode equipment, transfer the GPS coordinate to the video frame. This, in turn, will facilitate the utilization and interpretation of the videography to support a variety of image processing applications.

Two technical personnel would be required to support the lab. These individuals would provide basic administration of all equipment and software as well as provide GIS application support. Additionally, these individuals would be responsible for the management (disk location, data storage and archiving) of all the research data in the remote sensing lab. This will be a non trivial task in view of the amount of data that is expected to be received and processed. These individuals would also assist with training of new personnel, students and personnel attending outreach seminars on remote sensing.

The project timeline is listed as the following (Appendix D).

Project Timeline for Acquisition

The dates given below allows for unexpected delays in ordering, product availability and shipping. The intent is to have all equipment on order within 60 days of the NASA award and installation, testing and final acceptance completed as soon as possible after each piece of equipment arrives.

Impact of project

This joint UNH, MBL and MIT project is likely to have significant impacts on regional and global IEA, global change science and policy, education and training. The following is just some of a long list of the potential impacts and implications of the project.

1. Contribution of the new equipment.

   Land use and land cover change is at the heart of the International Human Dimensions Programme on Global Environmental Change (IHDP). The new equipment will provide capacity for us to extract the most current information on land use and land cover change at regional and global scales, which is essential for our effort towards integrated modeling of land use and land cover change (see Fig. 1). Currently, our effort is constrained by computational resources, data storage, and data management.

   Availability of the new equipment and software will allow us to incorporate remote sensing with the Integrated Global System Model (IGSM, see Fig. 2) in a dynamic mode, so that we can use remote sensing data to initialize the IGSM state variables. We will also be able to validate the component models and to conduct error analysis of the IGSM. We are improving IGSM in a number of areas, for instance, analyzing the connections between global climate and local air pollution controls using a 3-dimensional atmospheric chemistry model, and incorporating global river networks and population into a land use change model. In cooperation with NCAR, MIT developed a new Model for Atmospheric Transport and Chemistry (MATCH) which is global, 3-dimensional, an driven by observed winds and predicted convective and turbulent subgrid-scale transport. The model has been tested using long and short lived tracers. This model has an existing chemistry subroutine and a version using the IGSM chemistry model is also under development. As part of the MIT Climate Modeling Initiative, MIT is also developing a new coupled ocean-atmosphere-land climate model based on the 3-D ocean model of Prof. Marshall, Wunsch and colleagues, and the latest version of the Max Planck Hmaburg ECHAM model for atmospheric circulation. Modifications to the latter model involving the convective and land processes subroutines are underway. The new facility in this proposal can process relevant global satellite observation data and will play a collaborative role in development and testing of the new MIT 3-D coupled ocean-atmosphere-land GCM. These 3-D atmospheric chemistry models and climate will eventually incorporated into the IGSM.

2. Outreach potential of the new facility

   The new facility will enhance the existing outreach programs and provide new opportunities for research, education and training. Here we discuss some of the opportunities.

   Data sharing: The NASA EOS-IDS at UNH and Hughes Applied Information Systems are collaborative partners in a prototype project that will provide full GIS capability over the internet and is fully interoperable with NASAís EOS Data and Information Core System (ECS). The initial prototype is called the UNH EOS Explorer. The Explorer is a web-based GIS (http://www.unh-ecs.sr.unh.edu). Specific components of the systems are a Java client and server, a Spatial Data Engine (SDE/ESRI) client and server and an Oracle database. The data we will generate using the proposed new computing facility will be incorporated into the Explorer, and thus be shared with other researchers, education institutions and individuals.

   Research support. UNH, MIT and MBL have a long tradition of leadership in interdisciplinary activities requiring broad participation by government, industry and non-profit organization worldwide. the EOS Institute at UNH recently signed a Memorandum of Understanding with NOAA to establish the Cooperative Institute for Coastal and Estuarine Environmental Technology (CICEET). The mission of CICEET is to identify, monitor, and reduce the impacts of contamination of coastal waters. Through CICEET, Institute investigators will be developing and analyzing data using Geographic Information System (GIS) and image processing tools. Equally importantly, they will be working with coastal managers and decision makers to understand the tools, the data, and the effective utilization of both. On yet another project, researchers at the Institute are also presently seeking funding from NASA to broaden the use of remotely sensed data to the state and local communities. The initial, 50-state seminar program will evolve into a series of hands-on training sessions, aimed at educating data managers and policy makers regarding available satellite imagery and the broad spectrum of applications enabled by access to and appropriate utilization of that imagery. For climate modeling, researchers at MIT work closely with NASAís GISS, the Max Planck Institute in Hamburg, Germany and ETH in Switzerland. UNH and MIT are homes to the IGBP program offices for GAIM and IGAC, respectively. UNH will provide adequate support for outside researchers to use the new equipment, so that it is utilized to its maximum capacity.

   Education and training. UNH, MIT and MBL have a long tradition of education and training for outstanding undergraduate, graduate and practicing professions. The new facility will provide the state-of-the-art research tools and information to students.

   UNH are developing curriculum enhancement in the areas of satellite remote sensing and biology for the Global Learning and Observation to Benefit the Environment (GLOBE), which joins K-12 students, educators and scientists in studying the globe environment. UNH organizes workshops that provide teacher training in GLOBE protocols.

   MIT Joint Program on the Science and Policy of Global Change has established a semi-annual MIT Global Change Forum to communicate results. The Forum meetings bring together high-level, international groups of climate scientists, economists, policy analyst, industry experts and policy makers for discussion of the science and policy aspects of global change. They also bring researchers into regular contact with the policy debate, so they can frame their work agenda with knowledge of the evolving needs. The Forums offer an opportunity for those taking different roles in the policy process to participate in a joint assessment and debate for current research results and policy proposals, and to learn from each other.

   MBL is active in postdoctoral training and is currently developing a undergraduate Environmental Study course under the support of the Andrew W. Mellon Fundation and other sponsors. About 15 to 20 students will come from various colleges and universities in the U.S. and spend a semester at Woods Hole. The courses, laboratory and field experiments will provide students with hand-on experience and the-state-of the-art knowledge in ecosystem studies.

3. Potential impacts of project on the Nationís environmental monitoring and research networks and programs.

   U.S. Global Change Research Program. As pointed out in the report ìOur Change Planet: The FY 1998 U.S. Global Change Research Programî, which is a Supplement to the Presidentís Fiscal Year 1998 Budget, the global change research community has recognized the need for improving models and other representations of the complex feedbacks among human and natural systems over time across various spatial scales. MIT, MBL and UNH are leading the effort toward to integrated assessment of global change at the global scale. The IGSM is designed to address both policy issues and some outstanding questions in global change science. Requests are increasing for us to provide input to studies of the economic impact and likely efficacy of other schemes under consideration within the Ad Hoc Group on the Berlin Mandate and elsewhere. Our work in improving IGSM is likely to have far-reaching potential implications in the international negotiations under the Framework Convention on Climate Change (FCCC).

   IGBP, IHDP and IPCC. The researchers in UNH, MBL and MIT are actively involved in various programs of IGBP, IHDP and IPCC climate change assessment. The results from TEM is included in the IPCC Climate Change 1995 Assessment Report (Melillo et al. 1996). With the help of the new computing facility, researchers will substantially increase our understanding of the earth system and will make significant contribution to the IGBP, IHDP and IPCC.

   NSF Long-Term Ecological Research (LTER) program. Researchers from UNH and MBL are actively conducting long-term ecological researches at the three of 18 LTER sites in the USA: Harvard Forest site in Massachusetts, Hubbard Brook Experimental Forest site in New Hampshire, and Arctic tundra site in Alaska. Field work and laboratory experiment have accumulated abundant data on ecological processes in various types of ecosystems. Synthesis and scaling-up of the plot-level research to landscape and region requires large-scale information of land cover, land use, disturbance and abiotic environment. The new computing facility will make such landscape and regional information to be generated in a timely fashion, and will make real-time integration of field work and remote sensing for environmental monitoring possible. The prototype we will develop in these three LTER sites can serve for other LTER sites.

   NSF Land Margin Ecosystems Research (LMER) program. Researchers in MBL are leading the effort in quantifying present function and future changes of coastal environment, where terrestrial ecosystems interact with marine ecosystems. The LMER currently has four sites: Columbia river, Chesapeake bay, Georgia river and Plum Island Sound. Generally, coastal areas are most populated areas. Rapid land use change in coastal regions affects significantly the structure and function of terrestrial ecosystems and marine ecosystems. The new computing facility will allow the researchers to detect land use change in a timely fashion, and will make real-time integration of field work and remote sensing for coastal environmental monitoring possible. The prototype we will develop at the Plum Island Sound site can also serve for other LMER sites.

Project and management plans

The project plan is (1) we expect to set up the equipment and instruments at the Center by June 30, 1998 (see Appendix D in the Section c); (2) we will use about 1 year to organize the existing remote sensing products (including data from AVHRR Pathfinder program and Landsat Pathfinder program) for supporting retrospective analysis of the IEA framework; (3) we will focus on processing, interpreting and analysis of data from the TRMM and EOS AM-1 as soon as the level 1 products of the EOS AM-1 are available.

Overall management will be the responsibility of the Principal Investigator. B. Moore, R. Prinn, H. Jacoby and J. Hobbie will coordinate and supervise the research efforts toward the linkage between remote sensing and IEA among the three institutions. The UNH will be responsible for the maintenance of the new computing facility. The usage of the equipment and instruments will be shared by researchers and students in the UNH, MBL and MIT. The UNH and MBL will primarily work on image processing for land cover classification and fire detection. The MBL and UNH will work on the linkage between remote sensing and biogeochemistry models and land use change model. The MIT and MBL will primarily work on processing remote sensing data for atmospheric chemistry and climate models, and human settlement and infrastructure. The assignment of Individual investigators is listed in Table e.1

Reference

Aber, J.D., S.V. Ollinger, C.A. Federer, P.B. Reich, M.L. Goulden, D.W. Kicklighter, J.M. Melillo, and R.G. Lathrop, Jr. 1995. Predicting the effects of climate change on water yield and forest production in the northeastern U.S. Climate Research 5: 207-222.

Braswell, B.H., D.S. Schimel, J.L. Privette, B. Moore III, W.J. Emery, E.W. Sulzman and A.T. Hudak. 1996. Extracting ecological and biophysical information from AVHRR optical measurements: an integrated algorithm based on inverse modeling. Journal of Geophysical Research 101(D18):23335-23348.

Blumthaler, M. and Ambach, W. (1994) Changes in solar radiation fluxes after the Pinatubo eruption. Tellus 46B:76-78.

Crutzen, P.J. and P.H. Zimmermann. 1991. The changing photochemistry of the troposphere. Tellus 43:136-151.

Douglas, I. 1994. Human settlements. In: W.B. Meyer and B.L. Turner II (eds.) Changes in Land Use and Land Cover: A Global Perspective. Cambridge University Press, New York, p149-169.

Federer, C.A., C. J. Vorosmarty, and B. Fekete. 1996. Intercomparison of methods for potential evapo-transpiration in regional or global water balance models. Water Resources Research, 32:2315-21.

Fennessy, M.J. and Y. Xue. 1997. Impact of USGS vegetation map on GCM simulations over the United States. Ecological Applications 7(1):22-33.

Goward, S. N., R. H. Waring, D.G. Dye and J. Yang. 1994. Ecological remote sensing at OTTER: satellite macroscale observations. Ecological Application 42:322-343.

Grubler, A. 1994. Technology. In: W.B. Meyer and B.L. Turner II (eds.) Changes in Land Use and Land Cover: A Global Perspective. Cambridge University Press, New York, p287-328.

Hall, A.S.C., H. Tian, Y. Qi, G. Pontius, J. Cornell and J. Uhlig. 1995a. Spatially explicit models of land use change and their application to the tropics. CDIAC, DOE Research Summary No. 31. February 1995.

Hall, A.S.C., H.Tian, Y. Qi. 1995b. Modeling spatial and temporal patterns of tropical land use change. Journal of Biogeography 22:753-757.

Jacoby, H. and R. Prinn, 1994. Uncertainty in climate change policy analysis. MIT Joint Program on the Science and Policy of Global Change, Report 1, 34 pages. Massachusetts Institute of Technology, Cambridge, MA.

Jacoby, H.D., R.S. Eckaus, A.D. Elleman, R.G. Prinn, D.M. Reiner and Z. Yang. 1996. QELRO impacts: domestic market, trade and distribution of burdens and climate change. MIT Joint Program on the Science and Policy of Global Change, Report 9. Massachusetts Institute of Technology, Cambridge, MA.

Kasischke, E.S., N.H.F. French, P. Harrell, N.L. Christensen, S.L. Ustin and D. Barry. 1993. Monitoring of wildfires in boreal forests using large area AVHRR NDVI composite image data. Remote Sensing Environment 45(1):61-71.

Kennedy, P.J., A.S. Belward and J.M. Gregoire. 1994. An improved approach to fire monitoring in West Africa using AVHRR data. International Journal of Remote Sensing 15(11): 2235-2255.

Kicklighter, D.W., M. Bruno, S. Dönges, G. Esser, M. Heimann, F. Joos, J. Kaduk, G.H. Kohlmaier, A.D. McGuire, J.M. Melillo, R. Meyer, B. Moore III, I.C. Prentice, W. Sauf, A. Schloss, S. Sitch, U. Wittenberg and G. Würth. A first order analysis of the potential of CO2 fertilization to affect the global carbon budget: An intercomparison study of four terrestrial biosphere models. (in preparation).

Koffi, B., J.M. Gregoire, G. Mahe and J.P. Lacaux. 1995. Remote sensing of bush fire dynamics in Central Africa from 1984-1988: Analysis in relation to regional vegetation and pluviometeric patterns. Atmosphere Research 39(1-3):179-200.

Li, C. and S. Frolking, 1997. Long-term simulations of soil organic matter in agricultural lands. Geoderma (in press).

Li, C., V. Narayanan, and R. Harriss, 1996, Model estimates of nitrous oxide emissions from agricultural lands in the United States. Global Biogeochemical Cycles 10:297-306.

Li, C., 1995, Impact of agricultural practices on soil C storage and N2O emissions in 6 states in the US. In: Advances in Soil Science (eds. Lai et al.), Soil Management and Greenhouse Effect, pp.101-112.

LI. C., S. Frolking and R. Harris. 1994a. Modeling nitrous oxide emissions from agriculture: A Florida case study. Chemosphere 28:1401-1415.

Li, C., S. Frolking and R. Harris. 1994b. Modeling carbon biogeochemistry in agricultural soils. Global Biogeochemical Cycles 8:237-254.

Li, C. S. Frolking and T. Frolking. 1992a. A model of nitrous oxide evolution from soil driven by rainfall events: 1. Model structure and sensitivity. Journal of Geophysical Research 97:9759-9776.

Li, C. S. Frolking and T. Frolking. 1992b. A model of nitrous oxide evolution from soil driven by rainfall events: 2. Model applications. Journal of Geophysical Research 97:9777-9783.

Liu, Y. 1996. Modeling the emission of nitrous oxide (N2O) and methane (CH4) from the terrestrial biosphere to the atmosphere. MIT Joint Program on the Science and Policy of Global Change, Report 10, 219 pages. Massachusetts Institute of Technology, Cambridge, MA, August 1996. Ph.D. Dissertation.

Loveland, T.R., J.W. Merchant, D.O. Ohlen and J.F. Brown. 1991. Development of a land-cover characteristics database for the conterminous U. S. Photogrammetric Engineering and Remote Sensing 57:1453-1463.

McGuire, A.D., J.M. Melillo, D.W. Kicklighter, Y. Pan, X. Xiao, J. Helfrich, B. Moore III, C.J. Vorosmarty, and A.L. Schloss. 1997. Equilibrium responses of global net primary production and carbon storage to doubled atmospheric carbon dioxide: Sensitivity to changes in vegetation nitrogen concentration. Global Biogeochemical Cycles 11: 173-189.

McGuire, A.D., D.W. Kicklighter, and J.M. Melillo. 1996. Global climate change and carbon cycling in grasslands and conifer forests. pp. 389-411. IN: Global Change: Effects on Coniferous Forests and Grasslands, SCOPE Volume 56. Edited by A.I. Breymeyer, D.O. Hall, J.M. Melillo and G.I. ‰gren. John Wiley and Sons, Chichester, United Kingdom.

McGuire, A.D., J.M. Melillo, D.W. Kicklighter, and L.A. Joyce. 1995. Equilibrium responses of soil carbon to climate change: Empirical and process-based estimates. Journal of Biogeography 22: 785-796.

McGuire, A.D., L.A. Joyce, D.W. Kicklighter, J.M. Melillo, G. Esser, and C.J. Vorosmarty. 1993. Productivity response of climax temperate forests to elevated temperature and carbon dioxide: a North American comparison between two global models. Climatic Change 24: 287-310.

McGuire, A.D., J.M. Melillo, L.A. Joyce, D.W. Kicklighter, A.L. Grace, B. Moore III, and C.J. Vorosmarty. 1992. Interactions between carbon and nitrogen dynamics in estimating net primary productivity for potential vegetation in North America. Global Biogeochemical Cycles 6: 101-124.

McGuire, A.D., D.W. Kicklighter, J.M. Melillo and J. Helfrich. 1997. Simulation of historical patterns of net ecosystem production in high latitude ecosystems. Journal of Geophysical Research (in preparation).

Melillo, J.M., I.C Prentice, G.D. Farquhar, E.D. Schulze and O.E. Sala. 1996a. Terrestrial biotic responses to environmental change and feedbacks to climate. In: J.T. Houghton, L.G. Meira Filho, B.A. Callander, N. Harris, A. Kattenberg and K. Maskell (eds.) Climate Change 1995: The Science of Climate Change. Cambridge University Press. P444-481.

Melillo, J.M., R.A. Houghton, D.W. Kicklighter, and A.D. McGuire. 1996b. Tropical deforestation and the global carbon budget. Annual Review of Energy and the Environment 21: 293-310.

Melillo, J.M. 1995a. Human influences on the global nitrogen budget and their implications for the global carbon budget. pp. 117-133. IN: Toward Global Planning of Sustainable Use of the Earth: Development of Global Eco-Engineering, edited by S. Murai and M. Kimura. Elsevier, Amsterdam, The Netherlands

Melillo, J.M., D.W. Kicklighter, A.D. McGuire, W.T. Peterjohn, and K.M. Newkirk. 1995b. Global change and its effects on soil organic carbon stocks. pp. 175-189. IN: Role of Nonliving Organic Matter in the Earthís Carbon Cycle, edited by R.G. Zepp and Ch. Sonntag. John Wiley and Sons, Ltd.

Melillo, J.M. 1994. Modeling land-atmosphere interactions: a short review. pp. 387-409 IN: Changes in Land Use and Land Cover: A Global Perspective, edited by W.B. Meyer and B.L. Turner II. Cambridge University Press, Cambridge.

Melillo, J.M., A.D. McGuire, D.W. Kicklighter, B. Moore III, C.J. Vorosmarty, and A.L. Schloss. 1993. Global climate change and terrestrial net primary production. Nature 363: 234-240.

Nemani, R. and S. Running. 1997. Land cover characterization using multitemporal red, near-IR and thermal-IR data from NOAA/AVHRR. Ecological Applications 7(1):79-90.

Novelli, P.C., L.P. Steele, and P.P. Tans, 1992: Mixing ratio of carbon monoxide in the troposphere. J. Geophys. Res., 97(D18), 20731-20750.

Pan, Y., A.D. McGuire, D.W. Kicklighter, and J.M. Melillo. 1996. The importance of climate and soils on estimates of net primary production: A sensitivity analysis with the Terrestrial Ecosystem Model. Global Change Biology 2: 5-23.

Pielke, R. A., T.J. Lee, J.H. Copeland, J.L. Eastman, C.L. Ziegler and C.A. Finley. 1997. Use of USGS-provided data to improve weather and climate simulations. Ecological Applications 7(1):3-21.

Potter, C.S., J.T. Randerson, C.B. Field, P.A. Matson, P.M. Vitousek, H.A. Mooney, and S.A. Klooster. 1993. Terrestrial ecosystem production: a process model based on global satellite and surface data. Global Biogeochemical Cycles 7:811-841.

Prince, S. D. and S. N Goward. 1995. Global primary production: a remote sensing approach. Journal of Biogeography 22:815-835.

Prinn, R., H. Jacoby, A. Sokolov, C. Wang, X. Xiao, Z. Yang, R. Eckaus, P. Stone, D. Ellerman, J. Melillo, J. Fitzmaurice, D. Kicklighter, G. Holian and Y. Liu. 1997. Integrated global system model for climate policy assessment: Feedbacks and sensitivity studies. Submitted to Climatic Change.

Prinn, R., H. Jacoby, A. Sokolov, C. Wang, X. Xiao, Z. Yang, R. Eckaus, P. Stone, D. Ellerman, J. Melillo, J. Fitzmaurice, D. Kicklighter, Y. Liu, and G. Holian. 1996. Integrated global system model for climate policy analysis: I. Model framework and sensitivity studies. MIT Joint Program on the Science and Policy of Global Change Report No. 7. Massachusetts Institute of Technology, Cambridge, Massachusetts. 76 p. ( see also http://web.mit.edu/globalchange/www/rpt7a.html)

Prinn, R.G., D.M. Cunnold, R. Rasmussen, P.G. Simmonds, F.N. Aleya, A. Crawford, P.J. Fraser, and R. Rosen, 1990: Atmospheric emissions and trends of nitrous oxide deduced from 10 years of ALE-GAGE data. J. Geophys. Res., 95(D11), 18369-18385.

Raich, J.W., E.B. Rastetter, J.M. Melillo, D.W. Kicklighter, P.A. Steudler, B.J. Peterson, A.L. Grace, B. Moore III, and C.J. Vorosmarty. 1991. Potential net primary productivity in South America: application of a global model. Ecological Applications 1: 399-429. 3: 241-265.

Ruimy, A. B. Saugier, and G. Dedieu. 1994. Methodology for the estimation of net primary production from remotely sensed data. Journal of Geophysical Research 99:5263-5283.

Schimel, D.S., VEMAP Participants, and B.H. Brasswell. 1997. Continental scale variability in ecosystem processes: Models, data and the role of disturbance. Ecological Monographs 67: 251-271.

Skole, D.L. 1994a. Data on global land-cover change: acquisition, assessment, and analysis. In: W.B. Meyer and B.L. Turner (eds.) Changes in Land Use and Land Cover: A Global Perspective. Cambridge University Press, New York. p437-471.

Skole, D.L., Chomentowski, W.H., Salas, W.A. and Nobre, A.D.. 1994b. Physical and human dimensions of deforestation in Amazonia. Bioscience 44,314-322.

Skole, D. L., and Tucker, C.J.. 1993. Tropical deforestation and habitat fragmentation in the Amazon: Satellite data from 1978 to 1988. Science 260,1905-1910

Sokolov, A. P. and P. H. Stone. 1995. Description and validation of the MIT version of the GISS 2-D Model. MIT Joint Program on the Science and Policy of Global Change Report 2. Massachusetts Institute of Technology, 46pp.

Sokolov, A. P. and P. H. Stone. 1997. A flexible climate model for use in integrated assessments. MIT Joint Program on the Science and Policy of Global Change Report 17. Massachusetts Institute of Technology, 46pp. also Climate Dynamics (in review).

Tian, H., J.M. Melillo, D.W. Kicklighter, A.D. McGuire, Y. Pan, X. Xiao, and J. Helfrich. 1997. Transient responses of terrestrial carbon flux and storage to changes in atmospheric CO2 and temperature from 1854 to 1990 in the conterminuous United States. Submitted to Ecological Applications.

Tian, H. 1995. Spatial and temporal patterns of carbon flux and land use/cover at scales from landscape to the biosphere: an empirical study with observation and modeling. Ph.D Dissertation, State University of New York, Syracuse, New York.

Tobler, W., U. Deichmann, J. Gottsegen and K. Maloy. 1995. The Global Demography Project. National Center for Geographical Information and Analysis Technical Report TR-96-6. Department of Geography, University of California, Santa Barbara, CA 93106.

VEMAP Members. 1995. Vegetation/ecosystem modeling and analysis project: Comparing biogeography and biogeochemistry models in a continental-scale study of terrestrial ecosystem responses to climate change and CO2 doubling. Global Biogeochemical Cycles 9:407-437.

Vorosmarty, C.J., C.J. Willmott, B.J. Choudhury, A.L. Schloss, T.K. Stearns, S.M. Robeson, and T.J. Dorman. 1996a. Analyzing the discharge regime of a large tropical river through remote sensing, ground-based climatic data, and modeling. Water Resources Research, 32: 3137-50.

Vorosmarty, C.J., B. Fekete, and B.A. Tucker. 1996b. River Discharge Database, Version 1.0 (RivDIS v1.0), Volumes 0 through 6. A contribution to IHP-V Theme1. Technical Documents in Hydrology Series. UNESCO, Paris.

Vorosmarty, C.J., B. Moore III, A.L. Grace, M.P.Gildea, J.M. Melillo, B.J. Peterson, E.B. Rastetter, and P.A. Steudler. 1989. Continental scale models of water balance and fluvial transport: an application to South America. Global Biogeochemical Cycles 3:241-265.

Wang, C., R.G. Prinn, A. Sokolov, P.H. Stone, Y. Liu and X. Xiao. 1995. A coupled atmospheric chemistry and climate model for chemically and radiatively important trace species. WMO-IGAC Conference on the Measurement and Assessment of Atmospheric Composition Change, 9-14 October 1995, Beijing; WMO GAW No. 107:182-184

Wang, C., R.G. Prinn, and A. Sokolov, 1997: A global interactive chemistry and climate model, manuscript in preparation.

Xiao, X., D.W. Kicklighter, J.M. Melillo, A.D. McGuire, P.H. Stone, and A.P. Sokolov. 1997a. Linking a global terrestrial biogeochemical model and a 2-dimensional climate model: implications for the carbon budget. Tellus 49B: 18-37.

Xiao, X., J.M. Melillo, D.W. Kicklighter, A.D. McGuire, R. Prinn, C. Wang, P.Stone and A. Skolov. 1997b. Transient climate change and net ecosystem production of the terrestrial biosphere. Global Biogeochemical Cycles (in review).

Xiao, X., J.M. Melillo, D.W. Kicklighter, A.D. McGuire, H. Tian, Y. Pan, C.J. V–r–smarty and Z. Yang. 1997c. Transient climate change and potential croplands of the world in the 21st century. Ambio (in review).

Yang, Z., R .S. Eckaus, A. D. Ellerman and H. D. Jacoby. 1996. The MIT Emissions Prediction and Policy Analysis (EPPA) Model, Massachusetts Institute of Technology Joint Program on the Science and Policy of Global Change, Report No. 6, Cambridge, MA, 49 pages. [online at http://web.mit.edu/globalchange/www/rpt6a.html]