Global Warming Prediction Project

Monthly Predictions from September 2009 to August 2012

Fri, 15 Jan 2010 11:11:57 +0100

This is the first set of monthly predictions which is available on this site.

One set contains a prediction for the entire globe and predictions of both land air (lat) and sea surface temperature (sst) anomalies for nine latitudinal areas. Each area represents a 20° wide ring around the globe. Each single prediction is described by an individual prediction interval of high, most likely, and low prediction values. The results for each area is shown in a separate slide and every slide in a set shows the predictions of the coming 36 months along with the observed averaged values of the past 36 months and the corresponding predictions for this period generated at earlier times for reference. A linear trend based on observed temperatures and most likely predictions summarizes the development of the anomalies in the displayed period. (more...)

TRENDS
Global temperature anomalies stay unchanged at 0.4°C.
Largest expected warming at 70N-50N_sst with 0.0055°C per month (0.4°C in 6 years).
Largest expected cooling at 70N-50N_lat with -0.0069°C per month (-0.5°C in 6 years; mainly due to exceptionally high temperatures in Dec 06 and Jan 07).
Land air temperatures are in average slightly falling (6 out of the 9 regions show a falling trend, for only one area a warming is expected).
Sea surface temperatures are in average slightly growing (5 out of 9 regions show a growing and 2 a falling trend).
On the northern hemisphere temperatures are rather falling including the artic region with the one exception, 70N-50N_sst.
On the southern hemisphere temperatures in almost all regions are rising, especially in the oceans including the antarctic sea.

Show slides...

see also:
About the Data
About the Prediction Models

About Monthly Predictions

Fri, 15 Jan 2010 07:50:53 +0100

On this site we will publish sets of monthly predictions of global temperature anomalies on a regular basis.

One set contains a prediction for the entire globe and predictions of both land air (lat) and sea surface temperature (sst) anomalies for nine latitudinal areas. Each area represents a 20° wide ring around the globe. Each single prediction is described by an individual prediction interval of high, most likely, and low prediction values. The results for each area is shown in a separate slide and every slide in a set shows the predictions of the coming 36 months along with the observed averaged values of the past 36 months and the corresponding predictions for this period generated at earlier times for reference. A linear trend based on observed temperatures and most likely predictions summarizes the development of the anomalies in the displayed period.

The presented results are obtained from non-linear prediction models which are self-organizing by our advanced and proven knowledge extraction technologies exclusively using the information stored in the observed noisy temperature data of past periods (more...). All prediction models will be updated, too, as new data is available.

The data source for the models and the predictions is maintained and published by the Climate Research Unit, School of Environmental Sciences, University of East Anglia, UK.

see also:
About the Data
About the Prediction Models

Yearly Predictions till 2020

Thu, 14 Jan 2010 20:07:34 +0100

The slides below show yearly projections of air and sea surface temperature anomalies till 2020 for nine latitudinal areas of the globe. The prediction models were obtained from public data by non-experts in climatology using the KnowledgeMiner (yX) software.

Show slides...

see also:
About the Data

About the Prediction Models

Thu, 14 Jan 2010 19:29:11 +0100

Decision-making, whatever the field of human endeavour, requires formulation and a good understanding of what the problem is. To predict what may happen to a system under certain circumstances is often very difficult even for the simplest of systems, especially if they are not man-made. Humans have for centuries been seeking proxies for real processes. A substitute that can generate reliable information about a real system and its behaviour is called a model and they form the basis for any decision. It is worth building models to aid decision making, because models make it possible to:

identify the relationships between cause and effect (the subject of identification). This leads to a deeper understanding of the problem at hand by deriving an analytical relationship between them,
predict the respective objects can expect over a finite future time span (the subject of prediction), but also to experiment with models. Exactly the ability to make predictions about the future forms the core of intelligence at all. Our brain uses a memory-prediction model to make continuous prediction of future events in parallel across all our senses,
simulate the objects’ behaviour by experiment with models, and thus answer “what-if” questions (subject of simulation) essential to decision-making.

The world around us is getting more complex, more interdependent, more connected and global. We can observe it, but we cannot understand it because of its complexity and the myriad interactions that are impossible to know let alone foresee. Uncertainty and vagueness, coupled with rapid developments radically affect humanity. Though we observe these effects, we most often do not understand the consequences of any actions, the dynamics involved and the inter-dependencies of real-world systems in which system variables are dynamically related to many others, and where it is usually difficult to differentiate which are the causes and which are the effects.

We are facing these complex problems, which do need decision-making, but the means – the models – for understanding, predicting, simulating, and where possible controlling such systems are simply missing increasingly. This is more and more the situation with many real-world problems. To fill this essential gap, new and appropriate inductive self-organising modelling methods have been theoretically and practically developed as powerful tools in revealing the missing implicit relationships within complex systems.

KnowledgeMiner has been active in the field of self-organizing knowledge extraction from data for many years. In this project, we use parallel implementations of our proven predictive modelling technologies to model one part of the earth climate system - the temperature distribution and development on land and sea - from public, observed, historic data, exclusively. They reliably self-organize models from high-dimensional, noisy data sets. This is essential for climate modeling since it allows considering system states for modeling and prediction which are way back in time (very long time lags [up to 816 months = 68 years, actually]; the memory of the system). This easily leads to 10,000+ input variables which need to be processed and which still should result in valid models.

Two types of models is applied in this project to calculate monthly predictions: non-linear, dynamic regression models and non-parametric models using the concept of similar patterns. Regression models are explicitely described by a set of regression equations self-organized from data (image above) while nonparametric models are more fuzzy described by a set of patterns which are similar to past periods in time. More about the modeling technologies can be found here.

see also:
About the Data

About the Data

Thu, 14 Jan 2010 18:28:29 +0100

The earth climate, seen from systems theory, is a complex system. Developing models for prediction for such complex systems by a certain theory is almost all very difficult, time-consuming and problematic, because the a priori knowledge about the system required to formulate the model is more or less incomplete. Therefore, based on the well accepted assumption that observational data of a system contains hidden information about the system‘s behavior modeling technologies based on knowledge extraction from noisy data has been a substantial alternative, and they are becoming increasingly important in many fields. (more...)

So data is a potential source of knowledge and the quality of the data (representativity, reliability, consistency) plays an important role. The basic raw data used in this project is taken from the site of the Climate Research Unit of University of East Anglia and is described this:

„Over land regions of the world over 3000 monthly station temperature time series are used. Coverage is denser over the more populated parts of the world, particularly, the United States, southern Canada, Europe and Japan. Coverage is sparsest over the interior of the South American and African continents and over the Antarctic. The number of available stations was small during the 1850s, but increases to over 3000 stations during the 1951-90 period. For marine regions sea surface temperature (SST) measurements taken on board merchant and some naval vessels are used. As the majority come from the voluntary observing fleet, coverage is reduced away from the main shipping lanes and is minimal over the Southern Oceans. Maps/tables giving the density of coverage through time are given for land regions by Jones and Moberg (2003) and for the oceans by Rayner et al. (2003). Both these sources also extensively discuss the issue of consistency and homogeneity of the measurements through time and the steps that have made to ensure all non-climatic inhomogeneities have been removed.“
source: Climate Research Unit of University of East Anglia

In climate modeling, it is common practice to use temperature anomalies instead of absolute temperature values. As the reference for anomalies calculation the average temperatures from 1961 to 1990 is used. This has several reasons and advantages:

„Stations on land are at different elevations, and different countries estimate average monthly temperatures using different methods and formulae. To avoid biases that could result from these problems, monthly average temperatures are reduced to anomalies from the period with best coverage (1961-90). For stations to be used, an estimate of the base period average must be calculated. Because many stations do not have complete records for the 1961-90 period several methods have been developed to estimate 1961-90 averages from neighbouring records or using other sources of data. Over the oceans, where observations are generally made from mobile platforms, it is impossible to assemble long series of actual temperatures for fixed points. However it is possible to interpolate historical data to create spatially complete reference climatologies (averages for 1961-90) so that individual observations can be compared with a local normal for the given day of the year.“
source: Climate Research Unit of University of East Anglia

see also:
About the Prediction Models

About KnowledgeMiner

Thu, 14 Jan 2010 17:26:14 +0100

KnowledgeMiner Software is a privately held company in the field of research, development, and application of unique self-organising, inductive, statistical learning modelling and knowledge discovery technologies for more than 14 years located in the U.S. and Germany. It has established research co-operations with recognised scientists and experts in adaptive learning, control systems, and knowledge mining from Germany, Ukraine, Czech Republic, China, USA, Greece, and Italy. The company is the developer of the KnowledgeMiner® software package, a distinguished commercial self-organising modeling and knowledge extraction tool. It implements a unique and innovative set of parallel algorithms for modelling, validation, and workflow processing of complex, high-dimensional systems to allow knowledge extraction from noisy data in a most objective, automated, and fast way. It has been applied in various fields like economics, ecology, medicine and sociology, biotechnology, and life sciences. Some of more recent projects include:

Cancer research,
Prediction of wastewater pre-precipitation and wastewater reuse,
Identifying walking gait abnormalities for persons wearing prosthetic legs,
Analyzing medical data obtained from observing eye movement of children (both healthy children and those who display reading abnormalities),
Modeling and prediction of regional economies and related economic problems,
Problematic pharmaceutical manufacturing processes,
Researching various aspects of language teaching and learning,
Discovering the relationship between altered multiple sclerosis ("MS") intensities in the caudate nucleus and patient disability in MS.

Another area of success has been in the European Union, where KnowledgeMiner has been used to predict the environmental repercussions of toxicity residues from a list of 30,000 chemical compounds. KnowledgeMiner can model available data to check hazards like carcinogenicity, skin sensitization, developmental toxicity or acute toxicity of pesticides. This was done in two international research projects funded by the European Commission based solely on the compounds' chemical structures. The results of these projects fit the strict requirements set by the European legislative assessment concerning these chemical compounds, known as REACH.
This, when implemented by European regulators, ends up saving the lives of thousands of animals a year in Europe that previously were used as the frontline to test the toxicity of these compounds. We hope this approach will spread to other countries in the world to save the needless sacrifice of animals. The results of this project are available here:
http://www.caesar-project.eu

KnowledgeMiner is being used by NASA, Boeing, MIT, Columbia, University of Hamburg, Mobil Oil, Pfizer, Dean & Company, and many other corporations, universities, research institutions, projects and individuals around the world.

KnowledgeMiner is a registered trademark of KnowledgeMiner Software.
For more information and a press review copy of our software, please contact us.

info at knowledgeminer.com
http://www.knowledgeminer.com

About this Project

Thu, 14 Jan 2010 11:42:16 +0100

The objective of this project is doing monthly modeling and prediction of global temperature anomalies through self-organizing knowledge extraction using public data. It predicts temperatures of nine different regions 36 months ahead, which is a rather short-term forecast horizon for climate. In the future, the project may extend to quaterly and yearly projections to also cover medium-term goals. It is open to every interested person. The results will be published, updated, and archived on this site on a regular basis as new data are available, and former predictions and trends can be compared with actual data.

Although we use the terms Climate Change and Global Warming as well known and well established names for the problem, it is important to note that this project is impartial and has no hidden personal, financial, political or other interests. It is entirely independent, transparent, and open in results. Moreover, the independence and impartiality of this project is substantially driven and warranted by the way the prediction models are developed: autonomously by self-organizing knowledge extraction from observed data.

Why this Project?
Mathematical modelling is at the core of many decision support systems. Like many problems in economics, ecology, biology, biochemistry, sociology, and life sciences, the earth climate system is ill-defined and can be characterized by:

insufficient a priori information about the system for adequately describing the inherent system relationships,
possessing a large number of variables, many of which are unknown and/or cannot be measured,
noisy data available in very small to very large data sets,
vague and fuzzy objects whose variables has to be described adequately.

Common to all modelling problems this means to:

apply a systematic, holistic approach to modelling,
take into account the inadequacy of a priori information about the real-world system,
describe the vagueness and uncertainties of variables and, consequently, uncertainty of results and
handle very small to very large sets of noisy data.

For ill-defined systems the classical hard approach that is based on the assumption that the world can be understood objectively and that knowledge about the world can be validated through empirical means needs to be replaced by a soft systems paradigm which can better describe vagueness and imprecision. This approach is based on the observation that humans only have an incomplete and rather vague understanding of the nature of the world but nevertheless are able to solve unexpected problems in uncertain situations.

Systems can be modelled through deductive logical-mathematical methods (theory-driven approach) or by inductive modelling methods (data-driven approach). Deductive methods have been used to advantages in cases of well-understood problems and that obey well-known principles (microscopic approach). The spectacular results in aerospace are prime examples of this approach. Here, the theory of the object being modelled is well known and obeys known physical laws.

In contrast, inductive methods are used when macroscopic models (sometimes termed black box models) are the only alternative. These models are derived from real physical data and represent the relationships implicit within the system without knowledge of the physical processes or mechanisms involved.

There is in real world a vast treasure trove of data that is being continuously amassed that contains useful information about the behaviour of systems. This is priceless information, which only needs to be trawled and suitably mined so as to transform it into useful knowledge that will expose the causal relationships between the principal variables. Theory-driven approaches to modelling are unduly restrictive to this end because of insufficient a priori knowledge, complexity and the uncertainty of the objects, as well as the exploding time and computing demands.

The models, predictions, and results of this project are therefore entirely based on parallel implementations of data-driven, self-organizing, high-dimensional knowledge extraction technologies implemented in our KnowledgeMiner® (yX) software. We invite interested persons to participate in this ongoing project to get more knowledge about the development of the earth climate out of the huge amount of observed data available also at NASA.

We hope this project will provide a useful, transparent, and objective contribution to short- to medium-term climate modeling and prediction efforts.

Frank Lemke
KnowledgeMiner Software

contact

Acknowledgement
A special thanks to Anna Stathaki, Robert E. King, Volodymir Stepashko, and Pavel Kordik for their thoughts, ideas, and contribution to this project.