Source: VIRGINIA POLYTECHNIC INSTITUTE submitted to NRP
NON-LINEAR CHARACTERISTICS OF AGRI-ENVIRONMENTAL DATA SERIES
Sponsoring Institution
National Institute of Food and Agriculture
Project Status
COMPLETE
Funding Source
Reporting Frequency
Annual
Accession No.
0213968
Grant No.
(N/A)
Cumulative Award Amt.
(N/A)
Proposal No.
(N/A)
Multistate No.
(N/A)
Project Start Date
Jul 1, 2008
Project End Date
Jun 30, 2013
Grant Year
(N/A)
Program Code
[(N/A)]- (N/A)
Recipient Organization
VIRGINIA POLYTECHNIC INSTITUTE
(N/A)
BLACKSBURG,VA 24061
Performing Department
Crop & Soil Environmental Sciences
Non Technical Summary
A critical mass of evidence is accumulating that the here-to-fore overwhelming reliance on the linear paradigm in science and engineering is no longer universally tenable. It is now increasingly acknowledged that the dynamic behavior of most natural complex systems is nonlinear. Natural complex system dynamics are characterized by non-linear, interlocked, and recursive relationships; they are thermodynamically far from equilibrium, open, and dissipative; respond non-proportionally to perturbations; maintain stable self-organized global patterns. The long-term global behavior of these systems is clearly regular and stable indicating some degree of determinism. On the other hand, their short term, local behavior may appear highly irregular and unpredictable; a classic example being regional climate vis-a-vis local weather. All these characteristics are shared by intrinsically nonlinear deterministic chaotic systems which under appropriate conditions will remain in a self-organized state with a mix of global order and local disorder. Consequently, non-linear analysis approaches would be more appropriate to gain realistic insight into the structure of such systems. Observation of natural complex system dynamic behavior results in environmental data series. The archives of such series are enormous and continue to grow both in size, scope, and detail. This project focuses on three specific agro-environmental subsets of these archives. We posit that the generating processes for these series are fundamentally nonlinear dynamic and deterministic chaotic. Our thesis is that deterministic non-linear dynamics theory can provide better understanding and information with respect to analysis of the selected environmental data series. We envisage that the findings of this research will contribute to the growing field of nonlinear science; specifically, its applications in the agricultural and life sciences. Also, that it will help to: a. improve our understanding of complex systems and thereby enable formulation of better descriptive and predictive models for such systems, and b. enable or open the door to better, more effective and efficient environmental monitoring, data analysis, experimental design, and management.
Animal Health Component
(N/A)
Research Effort Categories
Basic
(N/A)
Applied
(N/A)
Developmental
(N/A)
Classification

Knowledge Area (KA)Subject of Investigation (SOI)Field of Science (FOS)Percent
1020110201020%
1020110207020%
1020420201010%
1020420207010%
1320110207010%
1320110209020%
1320420207010%
Goals / Objectives
The goal is to advance the analysis of agri-environmental data series generated by public monitoring networks and stored in large digital databanks. The overall objectives of this project are: a. to contribute to the young but growing field of nonlinear analysis of spatio-temporal environmental data series in general and agri-environmental data series in particular b. to gain deeper quantitative insights into the complex natural systems that produce such series c. to thereby enable (or open the door to) better, more effective and efficient environmental monitoring, data analysis, experimental design, and management. Achieving these objectives will lead to better quantitative understanding and definition of the nature of the natural systems that generated the series. The techniques that are developed will be universally applicable and potentially very beneficial in terms of cost and time effectiveness to all types of environmental monitoring, data analysis, experimental design, and management. They will potentially permit extrapolation of the information from monitored to non-monitored areas and therefore minimize the cost and effectiveness of public agri-environmental monitoring programs.
Project Methods
Data series in three digital databanks will be used to develop and test the non-linear analytical. These are: 1. Air temperature series: Collected at 15 minute, hourly, and daily intervals at the 10 Agricultural Research and Extension Centers in Virginia have been archived in the MESONET databank. 2. Soil temperature series: Soil temperature data series at hourly intervals for many depth increments of the profile at the 6 monitoring sites in 3 mid-Atlantic states (Virginia, North Carolina, and Maryland) are archived in the USDA/NRCS Soil Climate Analysis Network (SCAN) databank. 3. Rainfall series: Decadal rainfall totals for synoptic stations with 25 or more years of record in Burkina Faso and Niger are archived by the US National Oceanic and Atmospheric Administration, National Climatic Data Center. The following procedures will be undertaken for each series 1. Assemble and visually examine the values for quality, consistency, missing data, and outliers and examine non-stationary characteristics i.e. trends and periodicities. 2. Calculate the auto-correlation function 3. Calculate the Hurst exponent 4. Use time delay phase space embedding to calculate the Lyapunov exponent and the correlation dimension and to construct recurrence plots (RPs) and (possibly) cross recurrence plots (CRPs) and joint recurrence plots (JRPs) 5. Perform recurrence quantification analysis (RQA) on the RPs and (if implemented) CRPs and JRPs As the findings and results of these analyses accumulate we would be able to bolster the hypothesis that non-linear approaches (vis-a-vis linear approaches) may reveal important characteristics and patterns of the underlying dynamics of the system producing agri-environmental data series. If indeed the results showed nonlinear dynamics, the characteristics and patterns of the recurrence plots can serve as a signature of the system dynamics at a given location. This will permit comparisons across locations at various scales and can therefore be used to extrapolate from monitored to non-monitored area.

Progress 07/01/08 to 06/30/13

Outputs
Target Audience: Target audiences were : (a) public agencies and private organizations providing emergency food assistance to drought-stricken countries of sub-saharan Africa, (b) scientists and policy makers in these countries, and (c) scientists worldwide trying to better understand the climate of the sub-saharan, semi-arid ecotone. This was achieved by (a) publication of results in anopen access, online journal, and (b) a public-access website (still under construction) with 1349 years of decadal rainfall records for 30 rainfall stations in Burkina Faso and Niger with coordinates ranging from 9.88 to 18.5 degree north latitude and -4.77 to 13.2 degree longitude. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided? We developed linkages withscientists at (a) the Natural Resources Conservation Service, Lincoln, Nebraska, (b) Dept., of Agricultural and Biosystems Engineering, Iowa State University, Ames, Iowa,and (c) the Institute for Water Saving Agriculture, North-West Agricultural and Forestry University, Yangling, China How have the results been disseminated to communities of interest? Results were disseminated via: (a) publication in open access, online journal, and (b) a public-access website (still under construction). What do you plan to do during the next reporting period to accomplish the goals? Nothing Reported

Impacts
What was accomplished under these goals? Rainfall over semi-arid sub-saharan West Africa did not return to its pre-1969 levels following the disastrous famine due to the unprecedented rainfall failure in consecutive years from 1969 through 1974 across the entire region. Consequently, many countries annually plan for delivering emergency humanitarian food aid (most provided by the US) to mitigate famine in this region. If failure of the rains at a given location can be reliably predicted ex-ante, it would help planners to respond in a timelier and more targeted manner. A good indicator would be a characteristic of the seasonal rainfall distribution that can be shown to be reasonably invariant over time and space. The question is: can we find such an indicator given the highlyvariable and unreliable inter-seasonal and intra-seasonal distribution of the mono-modal, monsoonal rainfall across the semi-arid ecotone of sub-Saharan of West Africa?We investigated whether such invariance existed for the seasonal median date (meaning the date when 50% of the seasonal total occurs). Weposted such invariancemay exist since the sun’s cyclic declination forces the advance and retreat of the Inter-tropical Front over West Africa. To test this, we examined the statistical properties of the seasonal median date for 1349 station-years of rainfall records for 30 point rainfallgauges in Burkina Faso and Niger with coordinates ranging from 9.88° to 18.5° north latitude and −4.77° to 13.2° longitude.Our resultsshowed the median date was quite narrowly distributed, and therefore can be used as a reasonable rainfall predictor. Our analysescan be extended for the entire network of point rainfall stations over semi-arid West Africa. The results can be used to generate near real-time GIS- based maps for the entire region pinpointinglikely areas of food security vulnerability as the current rainy season progresses. This would assist humanitarian and famine relief and assistance planners to respond in a more timely and targeted manner topotential failure of the rains in this region. Finally, the techniques that have been developed will be universally applicable and potentially very beneficial in terms of cost and time effectiveness to all types of environmental monitoring, data analysis, experimental design, and management in the United States and internationally.

Publications

  • Type: Journal Articles Status: Published Year Published: 2013 Citation: N. Persaud. 2013. Invariance in the seasonal median dates for mono-modal monsoonal rainfall distribution over the semi-srid ecotone of sub-saharan West Africa. International Journal of Geosciences, 2013, 4, 1-7. (open access online at http://www.scirp.org/journal/ijg)


Progress 10/01/11 to 09/30/12

Outputs
OUTPUTS: We completed further analyses that confirmed our hypothesis that (a) the sun's declination can be directly related to the advance and retreat of the Inter-tropical Front over the food-insecure semi-arid ecotone of sub-saharan West Africa and therefore (b) the median date for mono-modal monsoonal rainfall over this region would exhibit invariant behanvior. We demonstrated strong invariance in the median date by examining its statistical properties for 1349 years of rainfall records for 30 rainfall stations in Burkina Faso and Niger with coordinates ranging from 9.88 to 18.5 degree north latitude and minus 4.77 to 13.2 degree longitude. The results showed that the median date can be used as a reasonable predictor that would enable humanitarian assistance planners to intervene in a timely manner to drought and crop failure in this region. PARTICIPANTS: Nothing significant to report during this reporting period. TARGET AUDIENCES: Emergency and humanitarian food aid agencies PROJECT MODIFICATIONS: Nothing significant to report during this reporting period.

Impacts
Rainfall over semi-arid sub-saharan West Africa did not return to its pre-1969 levels following the disastrous famine due to the unprecedented rainfall failure in consecutive years from 1969 through 1974 from across the entire region. Many countries annually plan for delivering emergency humanitarian food aid (most provided by the US) to mitigate famine in this region. If failure of the rains can be reliably predicted, it would help planners to respond in a more timely manner. Many past attempts, to reliably predict future rainfall based on analysis of statistical patterns of historical point rainfall data have not been remarkably successful. As a result, planners have increasingly relied on costly modeling using real time data from weather observation satellites to better understand and predict the meso-scale rainfall regime over this region vis-a-vis the in-place traditional observations from synoptic stations. Our results show that the median date can be used as a reasonably reliable predictor. The algorithms and analyses developed can be expanded to generate near real-time GIS- based maps that will enable humanitarian and assistance planners to pinpoint likely areas of food security vulnerability as the rainy season progresses. More broadly, This work is relevant to US agriculture because the findings contribute directly to the scientific body of knowledge on applications of nonlinear science in semi-arid agricultural environments enabling more effective determinants of climate change.

Publications

  • No publications reported this period


Progress 10/01/10 to 09/30/11

Outputs
OUTPUTS: We have completed 3 studies that analyzed long duration time series of available point rainfall data over Burkina Faso and Niger (between 9 to 23 degrees north latitude) in the sudano-sahelian zone of West Africa. We are summarizing our findings for publication. These studies are: (1) Latitudinal effect on statistical parameters of point rainfall distribution over the sudano-sahelian zone. (2) Quantifying intensification of drought across the sudano-sahelian zone based on harmonic analysis of point rainfall records. (3) Latitudinal effect on the mid-point of the rainy season over the sudano-sahelian zone. These findings contribute directly to the scientific body of knowledge on applications nonlinear science in semi-arid agricultural environments. The techniques developed are directly applicable to US semi-arid, dryland agri-climates West of the Mississippi enabling more effective determinants of climate change. PARTICIPANTS: We received assistance and input from the Scientific Director, Institut National de la Recherche Agronomique du Niger (INRAN), B.P. 429, Niamey, Niger TARGET AUDIENCES: Scientific Community PROJECT MODIFICATIONS: Nothing significant to report during this reporting period.

Impacts
In their entirety, these studies supported our project hypothesis that the observed point rainfall distribution pattern over this zone results from the superimposition of shorter term, non-periodic stochastic influences on the more stable, periodic behavior of the Inter-tropical Convergence Zone over West Africa. These results would help policy makers better understand and respond to the long-term eco-geographical effects of climate change. Also, since the approaches enable better signalling of rainfall deficits, they could make US food assistance planning and decision making for this drought-prone region more timely and targeted. The findings contribute directly to the scientific body of knowledge on applications nonlinear science in semi-arid agricultural environments. The techniques developed are directly applicable to US semi-arid, dryland agri-climates West of the Mississippi enabling more effective determinants of climate change.

Publications

  • No publications reported this period


Progress 10/01/09 to 09/30/10

Outputs
OUTPUTS: We continued the rather difficult task to develop a non-linear spatio-temporal model to adequately describe the spatio-temporal latitudinal effect on point rainfall records from all staions in Niger and Burkina Faso in West Africa. We have compiled the results of calculations on (a) the annual and semi-annual harmonics from the rainfall series for all synoptic stations with 20 or more years of record from these two countries, and (b)the calculated mid-point dates of the monomodal rainy season for each year of record for the same synoptic stations. PARTICIPANTS: Nothing significant to report during this reporting period. TARGET AUDIENCES: Nothing significant to report during this reporting period. PROJECT MODIFICATIONS: Nothing significant to report during this reporting period.

Impacts
These results provide an approach to better quantify the reported intensification of drought across the sudano-sahelian zone based on the amplitude and phase shifts of the harmonic components. This would help policy makers better understand and respond to the long-term eco-geographical effects of climate change. The results would also make planning and decision making for food assistance more timely since the approach of the midpoint dates would signal that rainfall before would be equal to the rainfall after with a known degree of confidence.

Publications

  • No publications reported this period


Progress 10/01/08 to 09/30/09

Outputs
OUTPUTS: 1. We examined the non-linear patterns in the residuals obtained by removing the calculated annual and semi-annual harmonics from the rainfall series for all synoptic stations with 20 or more years of record in Burkina Faso and Niger, West Africa. The results showed that the latitudinal effect in space and time on point rainfall over these countries (that span the West African sudano-sahelian zone) can be described by the interplay of deterministic and stochastic factors. We are in the process of developing a non-linear spatio-temporal model to describe the latitudinal effect. 2. We examined the summary statistics and the parametric distributions for the calculated mid-point dates of the monomodal rainy season for each year of record for the same synoptic stations. The results showed that the variabilty of the mid-point dates increased with increasing latitude. On the other hand the occurrence of the midpoint date could be predicted with reasonable confidence. We are preparing three manuscripts that document these findings. PARTICIPANTS: Nothing significant to report during this reporting period. TARGET AUDIENCES: Nothing significant to report during this reporting period. PROJECT MODIFICATIONS: Nothing significant to report during this reporting period.

Impacts
These results provide an approach to better quantify the reported intensification of drought across the sudano-sahelian zone based on the amplitude and phase shifts of the harmonic components. This would help policy makers better understand and respond to the long-term eco-geographical effects of climate change. The results would also make planning and decision making for food assistance more timely since the approach of the midpoint dates would signal that rainfall before would be equal to the rainfall after with a known degree of confidence.

Publications

  • No publications reported this period


Progress 07/01/08 to 12/13/08

Outputs
OUTPUTS: In the 6 months since the inception of this project we have requested,assembled, and reformatted the digitized rainfall records for all synoptic stations in Burkina Faso and Niger, West Africa. These neighboring West African countries entirely straddle the soudano-sahelian zone. We have checked all the records received for errors and have calculated the mid-point dates of the rainy season for each year of record for all stations. We have calcuated the annual and semi-annual harmonic components of the annual rainfall distribution for each year for all stations with 20 or more years of record. PARTICIPANTS: Nothing significant to report during this reporting period. TARGET AUDIENCES: Nothing significant to report during this reporting period. PROJECT MODIFICATIONS: Nothing significant to report during this reporting period.

Impacts
We found: (a) A strong latitudinal effect on the occurrence and variabilty of the midpoint of the rainy season over the sudano-sahelian zone, and (b) Initial indications that we would be able to better quantify the reported intensification of drought across the sudano-sahelian zone based on the amplitude and phase shifts of the harmonic components we have thus far extracted.

Publications

  • No publications reported this period