gfdl's home page > about us > science of GFDL > science and nonscience concerning human-caused climate warming > why climate data are imperfect and why they are crucial anyway
why climate data are imperfect and why they are crucial anyway
The availability of climate data in many forms is crucial in the quest to understand, simulate, and predict the climate system and how it might change in the future. Such data provide the basics for our characterizations of the time-averaged climate states of various statistics of temperature, pressure, wind, water amounts, cloudiness, and precipitation as a function of geographical location, time, and altitude. Most importantly, such data provide invaluable information on the natural variability of climate, ranging from seasons to decades.
These data sets have empowered important direct insights on how the climate system works. For example, the observed average daily and seasonal ranges of mean temperature provide valuable evaluations of our theoretical understanding of how the climate changes in response to changed radiative circumstances (e. g. day to night, summer to winter). On longer timescales, indirect inferences (or proxy measures) provide valuable information on how ice ages and warm epochs appear to depend sensitively on subtle changes to the heating of Earth due to seemingly small variations in the precession of Earth's orientation toward the sun and in Earth's elliptical orbit around the sun. Interestingly, the onset of ice ages and their terminations appear to respond more sensitively to these small solar heating changes than are calculated by our current climate models. For example, the ice core records show that atmospheric CO2 lowers as the climate cools, a positive feedback effect that we do not expect to be relevant over the next century. However, such observations of prehistoric climates are ambiguous enough that they do not justify any confident conclusions that our current climate models may be underestimating the century-scale global temperature increase due to added greenhouse gases.
For the atmosphere, there are thousands of places on earth that collect information daily for the primary purpose of weather forecasting. Fortunately, all the information collected for weather purposes are also central to the needs to characterize longer-term climate. Unfortunately, many kinds of key atmospheric information are not readily available from the weather networks. These include vertical velocity, radiative heating/ cooling, cloud characteristics, evaporation, and properties of critical trace species such as particles containing sulfate and carbon.
For the land surface, many local sites provide information on snow, water storage, runoff, and soil moisture. Unfortunately, the spatial coverage is far from adequate, and most stations provide little information on the state of the vegetative cover and its role in governing surface water budgets and reflectivity of solar radiation.
For the world ocean, the data coverage is spotty and episodic relative to the need to characterize the state of the ocean and its role in climate variability and climate change. For example, we are still waiting to see the first instantaneous "weather map" of the internal ocean's waves, jets, and vortices, a privilege that is taken for granted by atmospheric scientists. Fortunately, the ocean's surface is partly accessible to measurements from earth-orbiting satellites. This allows remote measurements of ocean surface temperatures, sea state, and ocean height, a measure of integrated density over a fairly deep layer that allows some inferences about ocean currents.
For all parts of the climate system, the ability to characterize long-term trends of key climate variables is minimally adequate at best and nonexistent at worst. Few climate measurement systems currently in place are configured to address what I call the climate monitoring requirement.
Climate monitoring is defined here as the systematic, long-term collection of key climate measurements, with careful attention paid to maintenance of calibration and continuity of records for very long time intervals, and with a strong focus on interpretation of the data gathered. Very few current climate measurement systems satisfy these stringent requirements. This mainly is because of the fact that almost all climate-relevant measurements are gathered for shorter-term purposes such as weather forecasting, and for efforts to understand specific processes such as clouds or El Niño.
So, why should we care about this climate monitoring deficiency? Who actually has a stake in improved climate monitoring? Climate data scientists do because their goal is to use the data to learn about how climate and climate change actually work. Climate theorists and modelers do because the current anthropogenic greenhouse warming projections are theoretically based, as manifested in the mathematical climate models (making climate change projections without attempting to evaluate them against the evolving real world is counter to the ethic of science). Policymakers do because they are already in the process of making policy (or nonpolicy) in the face of an imperfectly understood, but potentially very serious, global environmental threat. Policymakers, like scientists, always need to evaluate their conclusions against new information.
In spite of the compelling needs for improved climate monitoring, not much is now being done nationally or internationally about the current monitoring deficiencies. Even worse, many critical capabilities are deteriorating in the United States and elsewhere because of budgetary pressures. Why is this so? This is a question that continues to baffle me. I suspect the answer lies mainly in the unwillingness of top officials to make firm commitments to a problem that requires sustained focus for many decades.10 Also, the problem suffers from its apparent lack of glamour. "What? No immediate payoff?" It is also possible that some may not feel much need to get the right answer if their minds are already made up, a phenomenon not unheard of at both ends of the political spectrum.
This summary of some of the barriers to better climate monitoring reveals a serious challenge that is currently producing a net reduction in the global climate monitoring capability at the same time that international policy negotiators are taking the greenhouse warming problem seriously. Clearly, improved information is required to guide the dauntingly tortuous mitigation (or lack thereof) of greenhouse gas emissions over the next century. The emerging climate monitoring information can reveal that our greenhouse warming projections were either too high or too low. Given this information, future mitigation decisions can be strongly affected. Without this key information, we will be flying in the dark much longer.
10It is a personal privilege to acknowledge the pioneering efforts of Charles D Keeling to ensure the presence of today's impressive CO2 record (this volume). He has taught us that proper climate monitoring is difficult, and invaluable. Perhaps soon the world will begin to take his message seriously.
Next: Role of Controversy