LSTM-based Recurrent Neural Network Predicts Influenza-like-illness in Variable Climate Zones

LSTM-based Recurrent Neural Network Predicts Influenza-like-illness in Variable Climate Zones | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article LSTM-based Recurrent Neural Network Predicts Influenza-like-illness in Variable Climate Zones Alfred Amendolara, Christopher Gowans, Joshua Barton, Andrew Payne, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4896641/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background Influenza virus is responsible for a recurrent, yearly epidemic in most temperate regions of the world. Flu has been responsible for a high disease burden in recent years, despite the confounding presence of SARS-CoV-2. However, the mechanisms behind seasonal variance in flu burden are not well understood. This study seeks to expand understanding of the impact of variable climate regions on seasonal flu trends. To that end, three climate regions have been selected. Each region represents a different ecological zone and provides different weather patterns. Methods A Long short-term memory (LSTM)-based recurrent neural network was used to predict influenza-like-illness trends for three separate locations: Hawaii, Vermont, and Nevada. Flu data were gathered from the Center for Disease Control as weekly influenza-like-illness (ILI) percentages. Weather data were collected from Visual Crossing and included temperature, wind speed, UV index, solar radiation, precipitation, and humidity. Data were prepared and the model was trained as described previously. Results All three regions showed strong seasonality of flu trends with Hawaii having the largest absolute ILI values. Temperature showed a moderate negative correlation with ILI in all three regions (Vermont = -54, Nevada = -0.56, Hawaii = -0.44). Humidity was moderately correlated in Nevada (0.47) and weakly correlated with ILI in Hawaii (0.22). Vermont ILI did not correlate with humidity. Precipitation and wind speed were weakly correlated in all three regions. Solar radiation and UV index showed moderate correlation in Vermont (-0.33, -0.36) and Nevada (-0.5263, -0.55), but only a weak correlation in Hawaii (-0.15, -0.18). When trained on the complete data sets, baseline model performances for all three datasets at + 1 week were equivalent. Models trained on one region and used to predict cross-regional data performed uniformly and equivalent to baseline. Conclusions Results indicate that climate variables were weak to moderate predictors in all regions. Initial modeling attempts revealed acceptable and uniform performance in all regions. When cross-regional predictions were made, performance remained uniform across all regions, implying that climate patterns may be more important than absolute climate values. Additionally, this data suggests that climate may not be as influential on flu trends as population-level human factors. modeling influenza LSTM neural network Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Background Influenza-like-illness (ILI) is a major concern for public health, significantly contributing to morbidity and mortality each year [ 1 ]. With the recent coronavirus disease 19 (COVID-19) pandemic, the burden and influence of the flu has become more complex and important to manage. While COVID-19 has tapered since its initial appearance, prior work has shown that COVID-19 and ILI share similar factors that contribute to the disease's spread, such as seasonality cycles [ 2 ]. So, while COVID-19 is likely here to stay, the flu has not disappeared and, despite decreased incidence during pandemic lockdowns, continues to cause significant disease burden by itself. By better understanding factors that contribute to its spread, public health professionals are better able to plan for, predict, and mitigate much of the impact of the flu virus. Temperature, humidity, UV index, and solar radiation are key climate factors correlated with measured ILI rates [ 2 ]. A region’s climate (whether it is more equatorial and tropical, temperate, or arid), exhibits distinct flu rates and patterns from other climate regions [ 3 ]. This study aims to target three specific and distinct climate regions with readily available climate history. Hawaii, Vermont, and Nevada were selected because of their respective tropical, temperate, and arid climates. Hawaii provides uniform tropical data sets, Vermont contributes a temperate 4-season climate, and Nevada, being 75% arid, provides a desert climate. By focusing on various regions with distinct climates, the model can target their specific weather patterns and assess underlying drivers. In addition to climate, many other factors contribute to the spread of influenza, including human travel patterns and air pollution [ 2 – 6 ]. This presents challenges when developing tools to predict flu trends, an important consideration when managing the sizable disease burden the flu causes each year [ 7 ]. One such tool that has seen use in recent years is Long Short-Term Memory (LSTM) based recurrent neural networks [ 8 ]. Long Short-Term Memory nodes seek to solve the disappearing or exploding gradient issue found commonly with neural networks, especially when dealing with highly dimensional time series data, through the inclusion of a constant error carousel as well as a “forget gate”. Both features allow the LSTM node to reset occasionally, while still retaining its time-dependent memory [ 9 , 10 ]. They are particularly adept at capturing patterns in time dependent sequential data such as stock price trends, weather, and disease incidence. This strength lends itself to disease forecasting, especially when considering covariate weather data. A number of flu modeling systems in recent years have used LSTM, on its own or in combination with other architectures to successfully predict flu patterns [ 8 , 11 – 15 ]. Building upon previous work, LSTM models were trained and tested on the three selected regions to determine which regional weather patterns and climate factors impacted flu and ILI dynamics. The results from these region-specific models can help public health specialists and local governments tailor interventions and allocate resources more effectively to limit transmission rates in their respective areas. This study contributes to the current understanding of how climate impacts the spread of ILI by utilizing LSTM modeling for both forecasting flu rates across various climates and weather patterns as well as for elucidating information about underlying climate drivers of seasonal flu patterns. Methods Data Collection and Preparation Flu data were gathered from the CDC as weekly unweighted influenza-like-illness (ILI) percent for Vermont, Nevada, and Hawaii [ 1 ]. These states were chosen to represent three distinct climate regions: temperate, arid, and tropical. Data spanned from approximately 2010 to 2023, depending on availability. Weather data were acquired from Visual Crossing and included temperature, wind speed, UV index, solar radiation, precipitation, and humidity [ 16 ]. These weather data sets were chosen based on previous reports, including our findings presented in Amendolara et al. 2023. Data were prepared by normalizing and generating a time series with a lag of 10 weeks. Approximately 10% of available data in each region was reserved for testing and validation, leaving 600 weeks of training data per region. Data were prepared using Excel and Python. Flu data for each region were visualized (Fig. 1 ). Additionally, a correlation matrix was generated to assess correlation between flu data and climate variables (Table 1 ). Model An LSTM-Based recurrent neural network was used to predict influenza-like-illness trends for three separate locations: Hawaii, Vermont, and Nevada. This general model architecture has been previously described and validated [ 17 ]. In brief, the model was built with a variable shaped, bidirectional 500 node input layer, two bidirectional 500 node LSTM hidden layers and a variable shaped dense output layer. The model was written in Python v3.9.13 using TensorFlow v2.10 and the Keras API v2.10. It was trained on a computer running Windows 11 with the following specifications: AMD Ryzen 9 5900X 12-Core Processor @ 3.70 GHz, 64BG 3600 MHz DDR4 RAM, Nvidia RTX 3060 Ti 8GB. Three versions of the model were trained - one for each region. Baseline model performance in each region was evaluated by calculating per-prediction error as well as mean squared error (MSE) for each same-region test set. A Kruskal-Wallis H-test was performed to ensure that error was not significantly different at baseline between the three models. In order to assess generalizability and differences in regional climate influence, trained models were used to predict 400–600 weeks (about 11 and a half years) of data from each of the other two regions i.e., the model trained on Vermont data was used to predict 600 weeks of data from Hawaii and 400 weeks of data from Nevada. Overall model performance was evaluated using MSE. All data and code used in this report are available via GitHub and Zenodo: https://zenodo.org/doi/ 10.5281/zenodo.13294740 . Results Seasonal Flu Trends are Similar Throughout Regions Influenza-like-illness patterns were largely similar between regions. Hawaii had one unusually high amplitude spike starting at approximately week 200, but the data from all three regions otherwise showed similar seasonality (Figure 1). COVID-19 pandemic restrictions are likely reflected started around the 500-week mark, where a departure from the regular seasonal pattern may be observed in all three regions. Weather data followed a similar pattern. Hawaii displayed less overall amplitude change in temperature, solar radiation, and UV index across the year. Absolute values of climate variables differed as well. However, all three regions still displayed underlying seasonal patterns. Plotted data from each region may be viewed in Additional File 1. Climate Variables Correlate with ILI Differently in Each Region Temperature showed a moderate negative correlation with ILI in all three regions (Vermont = -54, Nevada = -0.56, Hawaii = -0.44). Humidity was moderately correlated in Nevada (0.47) and weakly correlated with ILI in Hawaii (0.22). Vermont ILI did not correlate with humidity. Precipitation and wind speed were weakly correlated in all three regions. Solar radiation and UV index showed moderate correlation in Vermont (-0.33, -0.36) and Nevada (-0.5263, -0.55), but only weak correlation in Hawaii (-0.15, -0.18) (Table 1). Table 1. Correlation Matrices. Vermont % ILI Temp max Temp min Temp Humidity Precipitation Wind speed mean Solar radiation UV index % ILI 1.00 Temp Max -0.53 1.00 Temp Min -0.54 0.97 1.00 Temp -0.54 0.99 0.99 1.00 Humidity -0.01 0.04 0.16 0.09 1.00 Precipitation -0.13 0.23 0.29 0.26 0.37 1.00 Wind speed mean 0.19 -0.42 -0.41 -0.41 -0.41 -0.04 1.00 Solar radiation -0.34 0.79 0.71 0.77 -0.28 0.09 -0.23 1.00 UV index -0.36 0.81 0.73 0.78 -0.30 0.08 -0.22 0.97 1.00 Nevada % ILI Temp max Temp min Temp Humidity Precipitation Wind speed mean Solar radiation UV index % ILI 1.00 Temp Max -0.56 1.00 Temp Min -0.55 0.92 1.00 Temp -0.57 0.98 0.97 1.00 Humidity 0.47 -0.84 -0.69 -0.79 1.00 Precipitation 0.11 -0.32 -0.16 -0.25 0.41 1.00 Wind speed mean -0.09 -0.05 0.15 0.07 -0.10 0.30 1.00 Solar radiation -0.53 0.68 0.68 0.71 -0.67 -0.23 0.19 1.00 UV index -0.55 0.75 0.71 0.76 -0.74 -0.29 0.20 0.90 1 Hawaii % ILI Temp max Temp min Temp Humidity Precipitation Wind speed mean Solar radiation UV index % ILI 1.00 Temp Max -0.43 1.00 Temp Min -0.45 0.91 1.00 Temp -0.45 0.98 0.97 1.00 Humidity 0.22 -0.41 -0.32 -0.34 1.00 Precipitation 0.03 -0.24 -0.12 -0.17 0.48 1.00 Wind speed mean -0.24 0.29 0.45 0.33 -0.60 -0.19 1.00 Solar radiation -0.15 0.36 0.26 0.31 -0.38 -0.27 0.18 1.00 UV index -0.18 0.44 0.29 0.37 -0.46 -0.36 0.17 0.91 1.00 Baseline Performance is Similar Across Models When predicting ILI rates on same-region test data, all three models performed similarly at baseline. The Vermont model was able to achieve a MSE of 0.353, Hawaii a MSE of 0.099, and Nevada a MSE of 0.216. Error distribution looks broadly uniform between models and a Kruskal-Wallis H-test shows that they are not statistically different ( p- value = 0.0794) (Figure 2). Models Predict Cross-Regional Data Equally Well Testing on the larger, cross-region data set produced in general similar MSE compared to baseline performance. Hawaii- and Nevada-trained models did experience slightly decreased performance, though still within a reasonable range compared to baseline (Table 2). The best prediction performance was achieved by the Vermont-trained model predicting Hawaii data (MSE 0.179). This was also a significant performance improvement from the Vermont baseline performance (Baseline MSE = 0.353). In fact, both cross-regional predictions improved the Vermont model’s MSE (Table 2, Figure 3). Nevada-trained models performed similarly to baseline on both test sets, and performance was generally unremarkable. Visually though, the Nevada-trained model appears to provide the most consistent predictions with little overshooting or undershooting of true values (Figure 4). The worst prediction performance occurred when using the Hawaii-trained model to predict Nevada data (Table 2, Figure 5). However, this performance was still comparable to baseline performance. It is likely these performance differences are not meaningful. Despite the slight variation between model performance, overall, each model performed very similarly and well within the spread of baseline performance. Table 2. Comparison of Cross-Regional Performance. Presented as Mean Squared Error at +1-week prediction. Test Set (n = 600) Vermont Hawaii Nevada Training Set Vermont - 0.179 0.239 Hawaii 0.254 - 0.296 Nevada 0.258 0.244 - Discussion Climate and Flu Data Interestingly, ILI data are very similar between all three regions. This is somewhat counterintuitive, as previous research has shown that tropical climates tend to have flatter flu patterns and less seasonal variation [ 18 ]. While an arid region with higher temperature and UV index would be expected to have different ILI patterns due to those variables previously being shown to impact influenza trends, little difference between regions was observed. One limitation of this study is that there are almost certainly other factors impacting these state-level data though. Hawaii, for instance, may be impacted by travel as well as its relatively small population [ 19 ]. In all three states, data are collected from several discrete weather stations, yet climate factors are variable across such large expanses and average values are imperfect measures. These considerations may explain the weak to moderate correlations seen in the correlation matrix. Despite this, many climate variables serve as effective predictors, as we have shown previously [ 17 ]. This raises a potential question — are climate variables truly “drivers” of seasonal flu patterns or are they simply useful practical predictors of flu trends? Making this distinction could reveal other, hidden, variables that are true drivers of flu trends. Baseline Performance All three models performed similarly at baseline. There is some variability in MSE and error distribution, but this is to be expected as all three models were validated on slightly different test sets. This suggests that the model is robust to variations in training data regarding climate factors, and that the absolute amplitude of change in climate variables is significantly less important to model performance than the pattern of the change. While the pattern is expected to be similar at baseline for all time-series data, over 600 weeks of data is sufficient to decipher differences caused by input variables that differ between regions. For example, average temperature varies by 76 o Fahrenheit range in Vermont, but only by 15 o in Hawaii based on the Visual Crossing data. It is hard to imagine that this should have no discernable impact on baseline performance. We did show previously that shifting the phase of climate data will impact model performance [ 17 ]. This further supports that, at least for this approach, the seasonal trend of the climate data is significantly more important than the absolute values of the climate data. Cross-Regional Performance The most unexpected results came when using models trained on one region to predict flu data from another region. One would expect that the substantial variability in climate factors would impact performance. However, all three models showed comparable performance when applied to all regions. While the possibility exists that the model is simply robust enough to handle extreme variation in input data, this is unlikely to account for these results alone, suggesting that climate factors may not be as fundamental in driving flu transmission as previously assumed. This is particularly likely given the data preparation methods commonly used to prepare time series data. From a practical perspective, this shows the usefulness of time-series based approaches to flu modeling. A model trained on a relatively limited dataset may be applied almost anywhere else with similar results. However, these results also imply that climate factors may simply be well correlated with flu trends, rather than a true factor in seasonal flu variation. It is worth noting that the strongest correlations between flu and climate variables were seen in climate zones with strong seasonal variation i.e., Nevada and Vermont. One may then conclude that ILI is linked to season for reasons other than climate variable. This is further supported by similar ILI patterns across regions, despite the substantially different climate data. So, climate variables may serve as a predictor rather than a driver. The implications of this are somewhat difficult to interpret. Practically, it may not matter since they serve as a valuable data source for predictive models. Of course, these factors cannot be intentionally influenced by humans. On a biological level, there is evidence in lab environments that climate variables do impact flu transmission and spread [ 20 ]. But there exists a strong possibility that, due to the complicated nature of disease transmission and modern travel patterns, climate is less relevant in modern populations. That is to say that perhaps population-based factors are significantly more important than climate factors when considering yearly flu spread and burden. Notably, a recent paper showed that flu trends in Puerto Rico had synced with mainland United States trends, lending credence to this hypothesis [ 21 ]. This is further supported by the trends seen in the recent COVID-19 pandemic. As is readily apparent in the data presented here, flu burden dropped notably in the face of pandemic restrictions and precautions. Yet seasonal climate patterns remained unchanged. Thus, it stands to reason that the underlying pattern of climate data is more important for modeling and also may have more of an impact on real-world flu trends than the absolute numbers. Limitations This report does suffer from several limitations, which should be taken into account when interpreting results. Data were limited to flu and climate data that were available in an overlapping time period as well as recorded at the same time scale i.e., weekly. More data are usually preferable in the case of time series prediction, so including more than the currently available 12 years may improve performance, though perhaps not to a significant degree. This is not a major limitation but may be a consideration when evaluating model performance. Aside from limited data, which is an ongoing challenge when combining data from multiple separate sources, the regions (in this case individual states) are not perfect proxies for their particular climate type. Additionally, ILI rates are recorded statewide. Climate data on the other hand is more often collected from one or a handful of weather stations. This creates some inherent noise and imprecision in the data. Furthermore, states are not entirely uniform and may have variations in climate variables depending on location, elevation, and other environmental factors. This is a limitation of any model using data on this scale and is largely unavoidable. It does likely explain some of the minor inconsistencies in performance as well as the weaker correlation between certain variables and ILI. Finally, this report does not consider other potential variables that may impact flu spread. It appears, based on the results presented and existing literature, that there are other significant factors that drive flu trends including pollen, travel, indoor environmental variable, and other concurrent diseases [ 2 – 6 ]. Additional work, likely with a completely different modeling approach will be required to delineate the impact of population-level factors. Further Applications The results of this paper may be applied or extended in several ways. Practically, we have shown that this modeling approach is robust to variations in climate and flu data. This may allow a model that has been trained on data from a region with high quality, consistent flu reporting and reliable weather stations to be applied to a region without those resources. Thus, predictions could be made, and potential flu burden assessed, without the need to make a bespoke model. This could be especially useful in areas where flu data is limited or has only recently begun being tracked. Additionally, given that climate data may not be as strong of a driver as assumed, at least in the United States, research may be directed to better understanding, and modeling, the various population and environmental factors that could drive seasonal flu trends. This will likely result in more accurate, longer reaching predictive models. Conclusions Given the known limitations of the modeling approach presented here, it is difficult to draw strong conclusions about the underlying epidemiological processes of flu trends. However, our results may provide several insights for both practical modeling and biological understanding of flu patterns. First, time series-based models appear to be robust to data variation and strongly generalizable regardless of region. The use of time series models has increased in recent years, but limited work has been done comparing performance across purposely selected variable regions. Second, the climate variables presented here are useful predictors when building a model and should be considered when designing production models for flu prediction. Third, it may be that, while they are useful predictors, climate variables are not truly relevant drivers of flu trends in the real world. Further research is needed to delineate the impact of climate versus other population variables. But, given the uniform performance of this model despite variable input data, it is important to ask whether climate is as powerful a driver as assumed and, if it is not, what other factors impact yearly flu trends. This is particularly important when deciding on public health policy, predicting flu burden, and developing strategies to reduce said burden. Abbreviations CDC – Center for Disease Control and Prevention COVID-19 - Coronavirus Disease 2019 ILI – Influenza-Like-Illness LSTM – Long Short-Term Memory MSE – Mean Squared Error Declarations Ethics Approval and Consent to Participate Not applicable Consent for Publication Not Applicable Availability of Data and Material All data and code described in this paper are available via GitHub or Zenodo at https://zenodo.org/doi/10.5281/zenodo.13294740. Competing interests The authors declare that they have no competing interests Funding The authors have no funding to declare. Authors' contributions Conceptualization: A.A., A.P. and D.S.; Data curation: A.A., C.G. and J.B.; Formal analysis: A.A.; Investigation: A.A.; Methodology: A.A.; Project administration: A.A.; Software: A.A.; Supervision: A.P. and D.S.; Validation: A.A. and D.S.; Visualization: A.A.; Writing – original draft: A.A., C.G., J.B., A.P. and D.S.; Writing - review & editing: A.A., C.G., J.B., A.P. and D.S. References CDC FluView Interactive . 2024 [cited 2024; Available from: https://gis.cdc.gov/grasp/fluview/fluportaldashboard.html. Hoogeveen, M.J., Pollen likely seasonal factor in inhibiting flu-like epidemics. A Dutch study into the inverse relation between pollen counts, hay fever and flu-like incidence 2016–2019. Science of The Total Environment, 2020. 727 : p. 138543.DOI: https://doi.org/10.1016/j.scitotenv.2020.138543. Moriyama, M., W.J. Hugentobler, and A. Iwasaki, Seasonality of Respiratory Viral Infections. Annu Rev Virol, 2020. 7 (1): p. 83-101.DOI: 10.1146/annurev-virology-012420-022445. Liu, L., et al. LSTM Recurrent Neural Networks for Influenza Trends Prediction . in Bioinformatics Research and Applications . 2018. Cham: Springer International Publishing. Hoogeveen, M.J. and E.K. Hoogeveen, Comparable seasonal pattern for COVID-19 and flu-like illnesses. One Health, 2021. 13 : p. 100277.DOI: 10.1016/j.onehlt.2021.100277. Belderok, S.M., et al., Effect of travel on influenza epidemiology. Emerg Infect Dis, 2013. 19 (6): p. 925-31.DOI: 10.3201/eid1906.111864. Lakdawala, S.S. and K. Subbarao, The Ongoing Battle Against Influenza: The challenge of flu transmission. Nature Medicine, 2012. 18 (10): p. 1468-1470.DOI: 10.1038/nm.2953. Kara, A., Multi-step influenza outbreak forecasting using deep LSTM network and genetic algorithm. Expert Systems with Applications, 2021. 180 : p. 115153.DOI: https://doi.org/10.1016/j.eswa.2021.115153. Gers, F.A., J. Schmidhuber, and F. Cummins, Learning to Forget: Continual Prediction with LSTM. Neural Computation, 2000. 12 (10): p. 2451-2471.DOI: 10.1162/089976600300015015. Hochreiter, S. and J. Schmidhuber, Long short-term memory. Neural computation, 1997. 9 (8): p. 1735-1780. Suntronwong, N., et al., Climate factors influence seasonal influenza activity in Bangkok, Thailand. PLoS One, 2020. 15 (9): p. e0239729.DOI: 10.1371/journal.pone.0239729. Soebiyanto, R.P., et al., The Role of Temperature and Humidity on Seasonal Influenza in Tropical Areas: Guatemala, El Salvador and Panama, 2008–2013. PLOS ONE, 2014. 9 (6): p. e100659.DOI: 10.1371/journal.pone.0100659. Amin, S., et al., Detecting Dengue/Flu Infections Based on Tweets Using LSTM and Word Embedding. IEEE Access, 2020. 8 : p. 189054-189068.DOI: 10.1109/ACCESS.2020.3031174. Zhao, Z., et al., Study on the prediction effect of a combined model of SARIMA and LSTM based on SSA for influenza in Shanxi Province, China. BMC Infectious Diseases, 2023. 23 (1): p. 71.DOI: 10.1186/s12879-023-08025-1. Zhu, H., et al., Study on the influence of meteorological factors on influenza in different regions and predictions based on an LSTM algorithm. BMC Public Health, 2022. 22 (1): p. 2335.DOI: 10.1186/s12889-022-14299-y. Visual Crossing Weather (2010 - 2023) . 2024; Available from: https://www.visualcrossing.com/. Amendolara, A.B., et al., LSTM-based recurrent neural network provides effective short term flu forecasting. BMC Public Health, 2023. 23 (1): p. 1788.DOI: 10.1186/s12889-023-16720-6. Dave, K. and P.C. Lee, Global Geographical and Temporal Patterns of Seasonal Influenza and Associated Climatic Factors. Epidemiologic Reviews, 2019. 41 (1): p. 51-68.DOI: 10.1093/epirev/mxz008. Hawaii's Influenza Surveillance Program . 2024 August 1, 2024]; Available from: https://health.hawaii.gov/docd/about-us/programs/hawaiis-influenza-surveillance-program/#:~:text=The%20seasonal%20flu%20is%20a,we%20experience%20flu%20year%20round. Ference, R.S., J.A. Leonard, and H.D. Stupak, Physiologic Model for Seasonal Patterns in Flu Transmission. The Laryngoscope, 2020. 130 (2): p. 309-313.DOI: https://doi.org/10.1002/lary.27910. Paz–Bailey, G., et al., Recent influenza activity in tropical Puerto Rico has become synchronized with mainland US. Influenza and Other Respiratory Viruses, 2020. 14 (5).DOI: https://doi.org/10.1111/irv.12744. Additional Declarations No competing interests reported. Supplementary Files AdditionalFile1.pdf Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-4896641","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":351137463,"identity":"ea36f6fa-f58f-46a4-973c-bede00babfcb","order_by":0,"name":"Alfred Amendolara","email":"data:image/png;base64,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","orcid":"","institution":"Noorda College of Osteopathic Medicine","correspondingAuthor":true,"prefix":"","firstName":"Alfred","middleName":"","lastName":"Amendolara","suffix":""},{"id":351137464,"identity":"44a98eeb-da09-41d2-847f-dd1c7e5a8249","order_by":1,"name":"Christopher Gowans","email":"","orcid":"","institution":"Noorda College of Osteopathic Medicine","correspondingAuthor":false,"prefix":"","firstName":"Christopher","middleName":"","lastName":"Gowans","suffix":""},{"id":351137465,"identity":"49e05947-40c9-43af-be42-4c6720fb503f","order_by":2,"name":"Joshua Barton","email":"","orcid":"","institution":"Noorda College of Osteopathic Medicine","correspondingAuthor":false,"prefix":"","firstName":"Joshua","middleName":"","lastName":"Barton","suffix":""},{"id":351137466,"identity":"26b7e884-8285-4b9c-8bff-73d6db654731","order_by":3,"name":"Andrew Payne","email":"","orcid":"","institution":"Noorda College of Osteopathic Medicine","correspondingAuthor":false,"prefix":"","firstName":"Andrew","middleName":"","lastName":"Payne","suffix":""},{"id":351137467,"identity":"3aeb68d8-eace-4282-ada2-f54e17a82bc8","order_by":4,"name":"David Sant","email":"","orcid":"","institution":"Noorda College of Osteopathic Medicine","correspondingAuthor":false,"prefix":"","firstName":"David","middleName":"","lastName":"Sant","suffix":""}],"badges":[],"createdAt":"2024-08-11 21:23:14","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4896641/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4896641/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":64159836,"identity":"f12af160-1db1-45ff-9b6f-cba415ffaf03","added_by":"auto","created_at":"2024-09-09 07:08:10","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":60577,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eVermont, Nevada, and Hawaii Show Similar Seasonal Flu Trends. \u003c/strong\u003e(A) Overlayed data from all three states. (B) Separated data per state. Top: Vermont, Middle: Hawaii, Bottom: Nevada.\u003c/p\u003e","description":"","filename":"1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4896641/v1/1bf720e4d8657ddc944f391d.jpg"},{"id":64159391,"identity":"9824e3b3-ef39-455e-bc65-f98f01630a45","added_by":"auto","created_at":"2024-09-09 06:52:10","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":40749,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eBaseline Performance on Same-Region Test Data is Variable but Statistically Similar. \u003c/strong\u003e(A) Vermont-trained model predicting Vermont test data. (B) Hawaii-trained model predicting Hawaii test data. (C) Nevada-trained model predicting Nevada test data. Kruskal-Wallis H-test \u003cem\u003ep-\u003c/em\u003evalue = 0.0794. MSE = Mean Squared Error.\u003c/p\u003e","description":"","filename":"2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4896641/v1/55960f2817bf6da9c09a22a3.jpg"},{"id":64159393,"identity":"d80065f3-f9a2-408b-b50c-066aef1f19de","added_by":"auto","created_at":"2024-09-09 06:52:11","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":67358,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eLSTM model trained on data from Vermont is able to predict trends in Hawaii and Nevada\u003c/strong\u003e. (A) Training and validation curve. (B) 1+ week predictions made on seventy-four weeks of reserved Vermont training data (C) 1+ week predictions made on Hawaii data (D) 1+ week predictions made on Nevada data.\u003c/p\u003e","description":"","filename":"3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4896641/v1/cc6f98f4c8dc07bba0a34d0b.jpg"},{"id":64159392,"identity":"25a39b7c-cbcc-4d86-bea1-03b70638d777","added_by":"auto","created_at":"2024-09-09 06:52:11","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":69397,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eLSTM model trained on data from Hawaii is able to predict trends in Vermont and Nevada. \u003c/strong\u003e(A) Training and validation curve. (B) 1+ week predictions made on seventy-four weeks of reserved Hawaii training data (C) 1+ week predictions made on Vermont data (D) 1+ week predictions made on Nevada data.\u003c/p\u003e","description":"","filename":"4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4896641/v1/270c39d7648d6aa0b493e4f4.jpg"},{"id":64159388,"identity":"798474e1-07cc-49aa-80f1-d03742a9f0bb","added_by":"auto","created_at":"2024-09-09 06:52:10","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":64690,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eLSTM model trained on data from Nevada is able to predict trends in Hawaii and Nevada. \u003c/strong\u003e(A) Training and validation curve. (B) 1+ week predictions made on approximately one hundred weeks of reserved Nevada training data (C) 1+ week predictions made on Vermont data (D) 1+ week predictions made on Hawaii data.\u003c/p\u003e","description":"","filename":"5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4896641/v1/cac13a66877d9bc9db95e290.jpg"},{"id":82697232,"identity":"a7d95d6b-8c9a-4fc1-a812-37855ba92e8f","added_by":"auto","created_at":"2025-05-14 09:02:04","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1265348,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4896641/v1/b1585954-8d40-4311-a203-c93b9ad842b6.pdf"},{"id":64159390,"identity":"90a7b230-0c40-487f-84b0-a08e1d422ff0","added_by":"auto","created_at":"2024-09-09 06:52:10","extension":"pdf","order_by":9,"title":"","display":"","copyAsset":false,"role":"supplement","size":102694,"visible":true,"origin":"","legend":"","description":"","filename":"AdditionalFile1.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4896641/v1/9f3b8e9e8f6558a44627a580.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"LSTM-based Recurrent Neural Network Predicts Influenza-like-illness in Variable Climate Zones","fulltext":[{"header":"Background","content":"\u003cp\u003eInfluenza-like-illness (ILI) is a major concern for public health, significantly contributing to morbidity and mortality each year [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. With the recent coronavirus disease 19 (COVID-19) pandemic, the burden and influence of the flu has become more complex and important to manage. While COVID-19 has tapered since its initial appearance, prior work has shown that COVID-19 and ILI share similar factors that contribute to the disease's spread, such as seasonality cycles [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. So, while COVID-19 is likely here to stay, the flu has not disappeared and, despite decreased incidence during pandemic lockdowns, continues to cause significant disease burden by itself. By better understanding factors that contribute to its spread, public health professionals are better able to plan for, predict, and mitigate much of the impact of the flu virus.\u003c/p\u003e \u003cp\u003eTemperature, humidity, UV index, and solar radiation are key climate factors correlated with measured ILI rates [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. A region\u0026rsquo;s climate (whether it is more equatorial and tropical, temperate, or arid), exhibits distinct flu rates and patterns from other climate regions [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. This study aims to target three specific and distinct climate regions with readily available climate history. Hawaii, Vermont, and Nevada were selected because of their respective tropical, temperate, and arid climates. Hawaii provides uniform tropical data sets, Vermont contributes a temperate 4-season climate, and Nevada, being 75% arid, provides a desert climate. By focusing on various regions with distinct climates, the model can target their specific weather patterns and assess underlying drivers. In addition to climate, many other factors contribute to the spread of influenza, including human travel patterns and air pollution [\u003cspan additionalcitationids=\"CR3 CR4 CR5\" citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. This presents challenges when developing tools to predict flu trends, an important consideration when managing the sizable disease burden the flu causes each year [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eOne such tool that has seen use in recent years is Long Short-Term Memory (LSTM) based recurrent neural networks [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. Long Short-Term Memory nodes seek to solve the disappearing or exploding gradient issue found commonly with neural networks, especially when dealing with highly dimensional time series data, through the inclusion of a constant error carousel as well as a \u0026ldquo;forget gate\u0026rdquo;. Both features allow the LSTM node to reset occasionally, while still retaining its time-dependent memory [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. They are particularly adept at capturing patterns in time dependent sequential data such as stock price trends, weather, and disease incidence. This strength lends itself to disease forecasting, especially when considering covariate weather data. A number of flu modeling systems in recent years have used LSTM, on its own or in combination with other architectures to successfully predict flu patterns [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e, \u003cspan additionalcitationids=\"CR12 CR13 CR14\" citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eBuilding upon previous work, LSTM models were trained and tested on the three selected regions to determine which regional weather patterns and climate factors impacted flu and ILI dynamics. The results from these region-specific models can help public health specialists and local governments tailor interventions and allocate resources more effectively to limit transmission rates in their respective areas. This study contributes to the current understanding of how climate impacts the spread of ILI by utilizing LSTM modeling for both forecasting flu rates across various climates and weather patterns as well as for elucidating information about underlying climate drivers of seasonal flu patterns.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eData Collection and Preparation\u003c/h2\u003e \u003cp\u003eFlu data were gathered from the CDC as weekly unweighted influenza-like-illness (ILI) percent for Vermont, Nevada, and Hawaii [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. These states were chosen to represent three distinct climate regions: temperate, arid, and tropical. Data spanned from approximately 2010 to 2023, depending on availability. Weather data were acquired from Visual Crossing and included temperature, wind speed, UV index, solar radiation, precipitation, and humidity [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. These weather data sets were chosen based on previous reports, including our findings presented in Amendolara et al. 2023.\u003c/p\u003e \u003cp\u003eData were prepared by normalizing and generating a time series with a lag of 10 weeks. Approximately 10% of available data in each region was reserved for testing and validation, leaving 600 weeks of training data per region. Data were prepared using Excel and Python.\u003c/p\u003e \u003cp\u003eFlu data for each region were visualized (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Additionally, a correlation matrix was generated to assess correlation between flu data and climate variables (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003eModel\u003c/h2\u003e \u003cp\u003eAn LSTM-Based recurrent neural network was used to predict influenza-like-illness trends for three separate locations: Hawaii, Vermont, and Nevada. This general model architecture has been previously described and validated [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]. In brief, the model was built with a variable shaped, bidirectional 500 node input layer, two bidirectional 500 node LSTM hidden layers and a variable shaped dense output layer. The model was written in Python v3.9.13 using TensorFlow v2.10 and the Keras API v2.10. It was trained on a computer running Windows 11 with the following specifications: AMD Ryzen 9 5900X 12-Core Processor @ 3.70 GHz, 64BG 3600 MHz DDR4 RAM, Nvidia RTX 3060 Ti 8GB.\u003c/p\u003e \u003cp\u003eThree versions of the model were trained - one for each region. Baseline model performance in each region was evaluated by calculating per-prediction error as well as mean squared error (MSE) for each same-region test set. A Kruskal-Wallis H-test was performed to ensure that error was not significantly different at baseline between the three models.\u003c/p\u003e \u003cp\u003eIn order to assess generalizability and differences in regional climate influence, trained models were used to predict 400\u0026ndash;600 weeks (about 11 and a half years) of data from each of the other two regions i.e., the model trained on Vermont data was used to predict 600 weeks of data from Hawaii and 400 weeks of data from Nevada. Overall model performance was evaluated using MSE.\u003c/p\u003e \u003cp\u003eAll data and code used in this report are available via GitHub and Zenodo: https://zenodo.org/doi/\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.5281/zenodo.13294740\u003c/span\u003e\u003cspan address=\"10.5281/zenodo.13294740\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cstrong\u003e\u003cem\u003eSeasonal Flu Trends are Similar Throughout Regions\u0026nbsp;\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eInfluenza-like-illness patterns were largely similar between regions. Hawaii had one unusually high amplitude spike starting at approximately week 200, but the data from all three regions otherwise showed similar seasonality (Figure 1). COVID-19 pandemic restrictions are likely reflected started around the 500-week mark, where a departure from the regular seasonal pattern may be observed in all three regions. Weather data followed a similar pattern. Hawaii displayed less overall amplitude change in temperature, solar radiation, and UV index across the year. Absolute values of climate variables differed as well. However, all three regions still displayed underlying seasonal patterns. Plotted data from each region may be viewed in Additional File 1.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eClimate Variables Correlate with ILI Differently in Each Region\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTemperature showed a moderate negative correlation with ILI in all three regions (Vermont = -54, Nevada = -0.56, Hawaii = -0.44). Humidity was moderately correlated in Nevada (0.47) and weakly correlated with ILI in Hawaii (0.22). Vermont ILI did not correlate with humidity. Precipitation and wind speed were weakly correlated in all three regions. Solar radiation and UV index showed moderate correlation in Vermont (-0.33, -0.36) and Nevada (-0.5263, -0.55), but only weak correlation in Hawaii (-0.15, -0.18) (Table 1).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 1. Correlation Matrices.\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003ctable style=\"width: 100%;border: none;border-collapse:collapse;\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"10\" style=\"width:100.0%;border:none;border-bottom:solid windowtext 1.0pt;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cstrong\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003eVermont\u003c/span\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width:18.22%;border:none;border-bottom:solid windowtext 1.0pt;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e\u0026nbsp;\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;border:none;border-bottom:solid windowtext 1.0pt;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e% ILI\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;border:none;border-bottom:solid windowtext 1.0pt;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003eTemp max\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.72%;border:none;border-bottom:solid windowtext 1.0pt;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003eTemp min\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.68%;border:none;border-bottom:solid windowtext 1.0pt;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003eTemp\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:10.58%;border:none;border-bottom:solid windowtext 1.0pt;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003eHumidity\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:12.5%;border:none;border-bottom:solid windowtext 1.0pt;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003ePrecipitation\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:11.54%;border:none;border-bottom:solid windowtext 1.0pt;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003eWind speed mean\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:9.64%;border:none;border-bottom:solid windowtext 1.0pt;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003eSolar radiation\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:10.56%;border:none;border-bottom:solid windowtext 1.0pt;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003eUV index\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 18.22%;padding: 0in 5.4pt;height: 0.1in;vertical-align: bottom;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e% ILI\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;background:#893BC3;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e1.00\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:6.72%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:6.68%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:10.58%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:12.5%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:11.54%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:9.64%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:10.56%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 18.22%;padding: 0in 5.4pt;height: 0.1in;vertical-align: bottom;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003eTemp Max\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;background:#63CFAA;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.53\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;background:#893BC3;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e1.00\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.72%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:6.68%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:10.58%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:12.5%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:11.54%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:9.64%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:10.56%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 18.22%;padding: 0in 5.4pt;height: 0.1in;vertical-align: bottom;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003eTemp Min\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;background:#61CFAB;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.54\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;background:#8D41BF;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.97\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.72%;background:#893BC3;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e1.00\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.68%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:10.58%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:12.5%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:11.54%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:9.64%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:10.56%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 18.22%;padding: 0in 5.4pt;height: 0.1in;vertical-align: bottom;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003eTemp\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;background:#61CFAB;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.54\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;background:#8A3DC2;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.99\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.72%;background:#8B3DC1;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.99\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.68%;background:#893BC3;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e1.00\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:10.58%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:12.5%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:11.54%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:9.64%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:10.56%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 18.22%;padding: 0in 5.4pt;height: 0.1in;vertical-align: bottom;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003eHumidity\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;background:#BFD570;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.01\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;background:#C7D56A;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.04\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.72%;background:#DDD75C;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.16\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.68%;background:#D0D665;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.09\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:10.58%;background:#893BC3;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e1.00\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:12.5%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:11.54%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:9.64%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:10.56%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 18.22%;padding: 0in 5.4pt;height: 0.1in;vertical-align: bottom;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003ePrecipitation\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;background:#AAD37D;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.13\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;background:#EAD854;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.23\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.72%;background:#E2CB5D;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.29\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.68%;background:#E7D358;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.26\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:10.58%;background:#D9BD67;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.37\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:12.5%;background:#893BC3;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e1.00\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:11.54%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:9.64%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:10.56%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 18.22%;padding: 0in 5.4pt;height: 0.1in;vertical-align: bottom;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003eWind speed mean\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;background:#E2D759;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.19\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;background:#76D09D;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.42\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.72%;background:#78D09C;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.41\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.68%;background:#78D09D;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.41\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:10.58%;background:#78D09D;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.41\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:12.5%;background:#BAD473;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.04\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:11.54%;background:#893BC3;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e1.00\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:9.64%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:10.56%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 18.22%;padding: 0in 5.4pt;height: 0.1in;vertical-align: bottom;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003eSolar radiation\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;background:#85D194;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.34\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;background:#A466A5;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.79\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.72%;background:#AE7699;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.71\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.68%;background:#A76BA1;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.77\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:10.58%;background:#8FD28E;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.28\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:12.5%;background:#D1D664;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.09\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:11.54%;background:#98D288;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.23\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:9.64%;background:#893BC3;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e1.00\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:10.56%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 18.22%;border-top: none;border-right: none;border-left: none;border-image: initial;border-bottom: 1pt solid windowtext;padding: 0in 5.4pt;height: 0.1in;vertical-align: bottom;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003eUV index\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;border:none;border-bottom:solid windowtext 1.0pt;background:#81D197;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.36\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;border:none;border-bottom:solid windowtext 1.0pt;background:#A262A7;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.81\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.72%;border:none;border-bottom:solid windowtext 1.0pt;background:#AC739B;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.73\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.68%;border:none;border-bottom:solid windowtext 1.0pt;background:#A568A3;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.78\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:10.58%;border:none;border-bottom:solid windowtext 1.0pt;background:#8BD190;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.30\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:12.5%;border:none;border-bottom:solid windowtext 1.0pt;background:#D0D665;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.08\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:11.54%;border:none;border-bottom:solid windowtext 1.0pt;background:#99D287;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.22\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:9.64%;border:none;border-bottom:solid windowtext 1.0pt;background:#8D41BE;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.97\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:10.56%;border:none;border-bottom:solid windowtext 1.0pt;background:#893BC3;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e1.00\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 18.22%;padding: 0in 5.4pt;height: 0.1in;vertical-align: bottom;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 6.78%;padding: 0in 5.4pt;height: 0.1in;vertical-align: bottom;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 6.78%;padding: 0in 5.4pt;height: 0.1in;vertical-align: bottom;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 6.72%;padding: 0in 5.4pt;height: 0.1in;vertical-align: bottom;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 6.68%;padding: 0in 5.4pt;height: 0.1in;vertical-align: bottom;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 10.58%;padding: 0in 5.4pt;height: 0.1in;vertical-align: bottom;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 12.5%;padding: 0in 5.4pt;height: 0.1in;vertical-align: bottom;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 11.54%;padding: 0in 5.4pt;height: 0.1in;vertical-align: bottom;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 9.64%;padding: 0in 5.4pt;height: 0.1in;vertical-align: bottom;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 10.56%;padding: 0in 5.4pt;height: 0.1in;vertical-align: bottom;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"10\" style=\"width:100.0%;border:none;border-bottom:solid windowtext 1.0pt;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cstrong\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003eNevada\u0026nbsp;\u003c/span\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width:18.22%;border:none;border-bottom:solid windowtext 1.0pt;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e\u0026nbsp;\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;border:none;border-bottom:solid windowtext 1.0pt;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e% ILI\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;border:none;border-bottom:solid windowtext 1.0pt;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003eTemp max\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.72%;border:none;border-bottom:solid windowtext 1.0pt;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003eTemp min\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.68%;border:none;border-bottom:solid windowtext 1.0pt;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003eTemp\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:10.58%;border:none;border-bottom:solid windowtext 1.0pt;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003eHumidity\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:12.5%;border:none;border-bottom:solid windowtext 1.0pt;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003ePrecipitation\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:11.54%;border:none;border-bottom:solid windowtext 1.0pt;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003eWind speed mean\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:9.64%;border:none;border-bottom:solid windowtext 1.0pt;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003eSolar radiation\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:10.56%;border:none;border-bottom:solid windowtext 1.0pt;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003eUV index\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 18.22%;padding: 0in 5.4pt;height: 0.1in;vertical-align: bottom;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e% ILI\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;background:#893BC3;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e1.00\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:6.72%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:6.68%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:10.58%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:12.5%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:11.54%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:9.64%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:10.56%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 18.22%;padding: 0in 5.4pt;height: 0.1in;vertical-align: bottom;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003eTemp Max\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;background:#60CFAB;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.56\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;background:#893BC3;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e1.00\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.72%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:6.68%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:10.58%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:12.5%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:11.54%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:9.64%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:10.56%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 18.22%;padding: 0in 5.4pt;height: 0.1in;vertical-align: bottom;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003eTemp Min\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;background:#62CFAA;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.55\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;background:#934BB8;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.92\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.72%;background:#893BC3;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e1.00\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.68%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:10.58%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:12.5%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:11.54%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:9.64%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:10.56%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 18.22%;padding: 0in 5.4pt;height: 0.1in;vertical-align: bottom;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003eTemp\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;background:#5FCFAC;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.57\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;background:#8C3FC0;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.98\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.72%;background:#8D41BF;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.97\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.68%;background:#893BC3;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e1.00\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:10.58%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:12.5%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:11.54%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:9.64%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:10.56%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 18.22%;padding: 0in 5.4pt;height: 0.1in;vertical-align: bottom;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003eHumidity\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;background:#C9A27A;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.47\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;background:#2ECBCB;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.84\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.72%;background:#49CDBA;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.69\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.68%;background:#35CCC7;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.79\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:10.58%;background:#893BC3;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e1.00\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:12.5%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:11.54%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:9.64%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:10.56%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 18.22%;padding: 0in 5.4pt;height: 0.1in;vertical-align: bottom;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003ePrecipitation\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;background:#D9D65F;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.11\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;background:#8BD191;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.32\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.72%;background:#A8D37E;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.16\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.68%;background:#97D288;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.25\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:10.58%;background:#D0AE71;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.41\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:12.5%;background:#893BC3;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e1.00\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:11.54%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:9.64%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:10.56%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 18.22%;padding: 0in 5.4pt;height: 0.1in;vertical-align: bottom;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003eWind speed mean\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;background:#B5D476;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.09\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;background:#BED570;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.05\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.72%;background:#E1D75A;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.15\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.68%;background:#D2D663;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.07\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:10.58%;background:#B3D477;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.10\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:12.5%;background:#DEC462;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.30\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:11.54%;background:#893BC3;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e1.00\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:9.64%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:10.56%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 18.22%;padding: 0in 5.4pt;height: 0.1in;vertical-align: bottom;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003eSolar radiation\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;background:#66CFA8;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.53\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;background:#B07997;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.68\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.72%;background:#B17B96;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.68\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.68%;background:#AC749B;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.71\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:10.58%;background:#4BCDB9;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.67\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:12.5%;background:#9CD286;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.23\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:11.54%;background:#E9D755;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.19\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:9.64%;background:#893BC3;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e1.00\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:10.56%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 18.22%;border-top: none;border-right: none;border-left: none;border-image: initial;border-bottom: 1pt solid windowtext;padding: 0in 5.4pt;height: 0.1in;vertical-align: bottom;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003eUV index\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;border:none;border-bottom:solid windowtext 1.0pt;background:#61CFAB;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.55\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;border:none;border-bottom:solid windowtext 1.0pt;background:#A86CA0;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.75\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.72%;border:none;border-bottom:solid windowtext 1.0pt;background:#AC739B;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.71\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.68%;border:none;border-bottom:solid windowtext 1.0pt;background:#A66AA2;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.76\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:10.58%;border:none;border-bottom:solid windowtext 1.0pt;background:#3ECCC1;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.74\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:12.5%;border:none;border-bottom:solid windowtext 1.0pt;background:#90D28D;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.29\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:11.54%;border:none;border-bottom:solid windowtext 1.0pt;background:#EAD854;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.20\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:9.64%;border:none;border-bottom:solid windowtext 1.0pt;background:#954FB5;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.90\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:10.56%;border:none;border-bottom:solid windowtext 1.0pt;background:#893BC3;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e1\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 18.22%;padding: 0in 5.4pt;height: 0.1in;vertical-align: bottom;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 6.78%;padding: 0in 5.4pt;height: 0.1in;vertical-align: bottom;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 6.78%;padding: 0in 5.4pt;height: 0.1in;vertical-align: bottom;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 6.72%;padding: 0in 5.4pt;height: 0.1in;vertical-align: bottom;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 6.68%;padding: 0in 5.4pt;height: 0.1in;vertical-align: bottom;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 10.58%;padding: 0in 5.4pt;height: 0.1in;vertical-align: bottom;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 12.5%;padding: 0in 5.4pt;height: 0.1in;vertical-align: bottom;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 11.54%;padding: 0in 5.4pt;height: 0.1in;vertical-align: bottom;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 9.64%;padding: 0in 5.4pt;height: 0.1in;vertical-align: bottom;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 10.56%;padding: 0in 5.4pt;height: 0.1in;vertical-align: bottom;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"10\" style=\"width:100.0%;border:none;border-bottom:solid windowtext 1.0pt;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cstrong\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003eHawaii\u003c/span\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width:18.22%;border:none;border-bottom:solid windowtext 1.0pt;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e\u0026nbsp;\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;border:none;border-bottom:solid windowtext 1.0pt;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e% ILI\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;border:none;border-bottom:solid windowtext 1.0pt;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003eTemp max\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.72%;border:none;border-bottom:solid windowtext 1.0pt;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003eTemp min\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.68%;border:none;border-bottom:solid windowtext 1.0pt;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003eTemp\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:10.58%;border:none;border-bottom:solid windowtext 1.0pt;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003eHumidity\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:12.5%;border:none;border-bottom:solid windowtext 1.0pt;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003ePrecipitation\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:11.54%;border:none;border-bottom:solid windowtext 1.0pt;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003eWind speed mean\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:9.64%;border:none;border-bottom:solid windowtext 1.0pt;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003eSolar radiation\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:10.56%;border:none;border-bottom:solid windowtext 1.0pt;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003eUV index\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 18.22%;padding: 0in 5.4pt;height: 0.1in;vertical-align: bottom;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e% ILI\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;background:#893BC3;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e1.00\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:6.72%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:6.68%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:10.58%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:12.5%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:11.54%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:9.64%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:10.56%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 18.22%;padding: 0in 5.4pt;height: 0.1in;vertical-align: bottom;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003eTemp Max\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;background:#72D0A0;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.43\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;background:#893BC3;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e1.00\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.72%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:6.68%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:10.58%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:12.5%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:11.54%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:9.64%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:10.56%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 18.22%;padding: 0in 5.4pt;height: 0.1in;vertical-align: bottom;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003eTemp Min\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;background:#6FD0A2;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.45\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;background:#964FB4;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.91\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.72%;background:#893BC3;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e1.00\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.68%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:10.58%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:12.5%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:11.54%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:9.64%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:10.56%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 18.22%;padding: 0in 5.4pt;height: 0.1in;vertical-align: bottom;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003eTemp\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;background:#6FD0A2;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.45\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;background:#8C40BF;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.98\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.72%;background:#8E43BD;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.97\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.68%;background:#893BC3;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e1.00\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:10.58%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:12.5%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:11.54%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:9.64%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:10.56%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 18.22%;padding: 0in 5.4pt;height: 0.1in;vertical-align: bottom;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003eHumidity\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;background:#E4D758;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.22\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;background:#76D09E;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.41\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.72%;background:#86D193;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.32\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.68%;background:#82D196;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.34\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:10.58%;background:#893BC3;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e1.00\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:12.5%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:11.54%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:9.64%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:10.56%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 18.22%;padding: 0in 5.4pt;height: 0.1in;vertical-align: bottom;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003ePrecipitation\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;background:#C3D56D;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.03\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;background:#94D28B;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.24\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.72%;background:#A8D37E;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.12\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.68%;background:#9FD383;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.17\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:10.58%;background:#CDA975;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.48\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:12.5%;background:#893BC3;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e1.00\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:11.54%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:9.64%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:10.56%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 18.22%;padding: 0in 5.4pt;height: 0.1in;vertical-align: bottom;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003eWind speed mean\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;background:#94D28A;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.24\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;background:#E6D159;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.29\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.72%;background:#D1AF71;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.45\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.68%;background:#E1C95E;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.33\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:10.58%;background:#55CEB3;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.60\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:12.5%;background:#9DD385;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.19\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:11.54%;background:#893BC3;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e1.00\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:9.64%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width:10.56%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 18.22%;padding: 0in 5.4pt;height: 0.1in;vertical-align: bottom;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003eSolar radiation\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;background:#A2D382;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.15\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;background:#DDC363;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.36\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.72%;background:#EAD854;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.26\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.68%;background:#E4CE5B;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.31\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:10.58%;background:#7BD09A;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.38\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:12.5%;background:#8DD28F;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.27\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:11.54%;background:#DCD75D;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.18\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:9.64%;background:#893BC3;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e1.00\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:10.56%;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 18.22%;border-top: none;border-right: none;border-left: none;border-image: initial;border-bottom: 1pt solid windowtext;padding: 0in 5.4pt;height: 0.1in;vertical-align: bottom;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;line-height:normal;'\u003e\u003cem\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003eUV index\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;border:none;border-bottom:solid windowtext 1.0pt;background:#9ED384;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.18\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.78%;border:none;border-bottom:solid windowtext 1.0pt;background:#D2B16F;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.44\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.72%;border:none;border-bottom:solid windowtext 1.0pt;background:#E7D258;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.29\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:6.68%;border:none;border-bottom:solid windowtext 1.0pt;background:#DCC164;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.37\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:10.58%;border:none;border-bottom:solid windowtext 1.0pt;background:#6ED0A3;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.46\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:12.5%;border:none;border-bottom:solid windowtext 1.0pt;background:#7FD198;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e-0.36\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:11.54%;border:none;border-bottom:solid windowtext 1.0pt;background:#DCD75D;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.17\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:9.64%;border:none;border-bottom:solid windowtext 1.0pt;background:#954EB5;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e0.91\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width:10.56%;border:none;border-bottom:solid windowtext 1.0pt;background:#893BC3;padding:0in 5.4pt 0in 5.4pt;height:.1in;\"\u003e\n \u003cp style='margin-top:0in;margin-right:0in;margin-bottom:0in;margin-left:0in;font-size:11.0pt;font-family:\"Calibri\",sans-serif;text-align:center;line-height:normal;'\u003e\u003cspan style='font-size:12px;font-family:\"Times New Roman\",serif;color:black;'\u003e1.00\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eBaseline Performance is Similar Across Models\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWhen predicting ILI rates on same-region test data, all three models performed similarly at baseline. The Vermont model was able to achieve a MSE of 0.353, Hawaii a MSE of 0.099, and Nevada a MSE of 0.216. Error distribution looks broadly uniform between models and a Kruskal-Wallis H-test shows that they are not statistically different (\u003cem\u003ep-\u003c/em\u003evalue = 0.0794) (Figure 2).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eModels Predict Cross-Regional Data Equally Well\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTesting on the larger, cross-region data set produced in general similar MSE compared to baseline performance. Hawaii- and Nevada-trained models did experience slightly decreased performance, though still within a reasonable range compared to baseline (Table 2).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe best prediction performance was achieved by the Vermont-trained model predicting Hawaii data (MSE 0.179). This was also a significant performance improvement from the Vermont baseline performance (Baseline MSE = 0.353). In fact, both cross-regional predictions improved the Vermont model\u0026rsquo;s MSE (Table 2, Figure 3).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eNevada-trained models performed similarly to baseline on both test sets, and performance was generally unremarkable. Visually though, the Nevada-trained model appears to provide the most consistent predictions with little overshooting or undershooting of true values (Figure 4).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe worst prediction performance occurred when using the Hawaii-trained model to predict Nevada data (Table 2, Figure 5). However, this performance was still comparable to baseline performance. It is likely these performance differences are not meaningful.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eDespite the slight variation between model performance, overall, each model performed very similarly and well within the spread of baseline performance.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 2. Comparison of Cross-Regional Performance.\u0026nbsp;\u003c/strong\u003ePresented as Mean Squared Error at +1-week prediction. \u0026nbsp;\u0026nbsp;\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"415\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd width=\"28.19277108433735%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"18.795180722891565%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"53.01204819277108%\" colspan=\"3\"\u003e\n \u003cp\u003e\u003cstrong\u003eTest Set (n = 600)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"28.26086956521739%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"18.840579710144926%\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd width=\"18.840579710144926%\"\u003e\n \u003cp\u003e\u003cem\u003eVermont\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.666666666666668%\"\u003e\n \u003cp\u003e\u003cem\u003eHawaii\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.391304347826086%\"\u003e\n \u003cp\u003e\u003cem\u003eNevada\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"28.26086956521739%\" rowspan=\"3\"\u003e\n \u003cp\u003e\u003cstrong\u003eTraining Set\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"18.840579710144926%\"\u003e\n \u003cp\u003e\u003cem\u003eVermont\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"18.840579710144926%\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"16.666666666666668%\"\u003e\n \u003cp\u003e0.179\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.391304347826086%\"\u003e\n \u003cp\u003e0.239\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"26.262626262626263%\"\u003e\n \u003cp\u003e\u003cem\u003eHawaii\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"26.262626262626263%\"\u003e\n \u003cp\u003e0.254\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"23.232323232323232%\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"24.242424242424242%\"\u003e\n \u003cp\u003e0.296\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"26.262626262626263%\"\u003e\n \u003cp\u003e\u003cem\u003eNevada\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"26.262626262626263%\"\u003e\n \u003cp\u003e0.258\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"23.232323232323232%\"\u003e\n \u003cp\u003e0.244\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"24.242424242424242%\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e"},{"header":"Discussion","content":"\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eClimate and Flu Data\u003c/h2\u003e \u003cp\u003eInterestingly, ILI data are very similar between all three regions. This is somewhat counterintuitive, as previous research has shown that tropical climates tend to have flatter flu patterns and less seasonal variation [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. While an arid region with higher temperature and UV index would be expected to have different ILI patterns due to those variables previously being shown to impact influenza trends, little difference between regions was observed. One limitation of this study is that there are almost certainly other factors impacting these state-level data though. Hawaii, for instance, may be impacted by travel as well as its relatively small population [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]. In all three states, data are collected from several discrete weather stations, yet climate factors are variable across such large expanses and average values are imperfect measures. These considerations may explain the weak to moderate correlations seen in the correlation matrix. Despite this, many climate variables serve as effective predictors, as we have shown previously [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]. This raises a potential question \u0026mdash; are climate variables truly \u0026ldquo;drivers\u0026rdquo; of seasonal flu patterns or are they simply useful practical predictors of flu trends? Making this distinction could reveal other, hidden, variables that are true drivers of flu trends.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eBaseline Performance\u003c/h2\u003e \u003cp\u003eAll three models performed similarly at baseline. There is some variability in MSE and error distribution, but this is to be expected as all three models were validated on slightly different test sets. This suggests that the model is robust to variations in training data regarding climate factors, and that the absolute amplitude of change in climate variables is significantly less important to model performance than the pattern of the change. While the pattern is expected to be similar at baseline for all time-series data, over 600 weeks of data is sufficient to decipher differences caused by input variables that differ between regions. For example, average temperature varies by 76\u003csup\u003eo\u003c/sup\u003e Fahrenheit range in Vermont, but only by 15\u003csup\u003eo\u003c/sup\u003e in Hawaii based on the Visual Crossing data. It is hard to imagine that this should have no discernable impact on baseline performance. We did show previously that shifting the phase of climate data will impact model performance [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]. This further supports that, at least for this approach, the seasonal trend of the climate data is significantly more important than the absolute values of the climate data.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eCross-Regional Performance\u003c/h2\u003e \u003cp\u003eThe most unexpected results came when using models trained on one region to predict flu data from another region. One would expect that the substantial variability in climate factors would impact performance. However, all three models showed comparable performance when applied to all regions. While the possibility exists that the model is simply robust enough to handle extreme variation in input data, this is unlikely to account for these results alone, suggesting that climate factors may not be as fundamental in driving flu transmission as previously assumed. This is particularly likely given the data preparation methods commonly used to prepare time series data. From a practical perspective, this shows the usefulness of time-series based approaches to flu modeling. A model trained on a relatively limited dataset may be applied almost anywhere else with similar results.\u003c/p\u003e \u003cp\u003eHowever, these results also imply that climate factors may simply be well correlated with flu trends, rather than a true factor in seasonal flu variation. It is worth noting that the strongest correlations between flu and climate variables were seen in climate zones with strong seasonal variation i.e., Nevada and Vermont. One may then conclude that ILI is linked to season for reasons other than climate variable. This is further supported by similar ILI patterns across regions, despite the substantially different climate data. So, climate variables may serve as a \u003cem\u003epredictor\u003c/em\u003e rather than a \u003cem\u003edriver.\u003c/em\u003e The implications of this are somewhat difficult to interpret. Practically, it may not matter since they serve as a valuable data source for predictive models. Of course, these factors cannot be intentionally influenced by humans. On a biological level, there is evidence in lab environments that climate variables do impact flu transmission and spread [\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e]. But there exists a strong possibility that, due to the complicated nature of disease transmission and modern travel patterns, climate is less relevant in modern populations. That is to say that perhaps population-based factors are significantly more important than climate factors when considering yearly flu spread and burden. Notably, a recent paper showed that flu trends in Puerto Rico had synced with mainland United States trends, lending credence to this hypothesis [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e]. This is further supported by the trends seen in the recent COVID-19 pandemic. As is readily apparent in the data presented here, flu burden dropped notably in the face of pandemic restrictions and precautions. Yet seasonal climate patterns remained unchanged.\u003c/p\u003e \u003cp\u003eThus, it stands to reason that the underlying pattern of climate data is more important for modeling and also may have more of an impact on real-world flu trends than the absolute numbers.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eLimitations\u003c/h2\u003e \u003cp\u003eThis report does suffer from several limitations, which should be taken into account when interpreting results. Data were limited to flu and climate data that were available in an overlapping time period as well as recorded at the same time scale i.e., weekly. More data are usually preferable in the case of time series prediction, so including more than the currently available 12 years may improve performance, though perhaps not to a significant degree. This is not a major limitation but may be a consideration when evaluating model performance.\u003c/p\u003e \u003cp\u003eAside from limited data, which is an ongoing challenge when combining data from multiple separate sources, the regions (in this case individual states) are not perfect proxies for their particular climate type. Additionally, ILI rates are recorded statewide. Climate data on the other hand is more often collected from one or a handful of weather stations. This creates some inherent noise and imprecision in the data. Furthermore, states are not entirely uniform and may have variations in climate variables depending on location, elevation, and other environmental factors. This is a limitation of any model using data on this scale and is largely unavoidable. It does likely explain some of the minor inconsistencies in performance as well as the weaker correlation between certain variables and ILI.\u003c/p\u003e \u003cp\u003eFinally, this report does not consider other potential variables that may impact flu spread. It appears, based on the results presented and existing literature, that there are other significant factors that drive flu trends including pollen, travel, indoor environmental variable, and other concurrent diseases [\u003cspan additionalcitationids=\"CR3 CR4 CR5\" citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. Additional work, likely with a completely different modeling approach will be required to delineate the impact of population-level factors.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eFurther Applications\u003c/h2\u003e \u003cp\u003eThe results of this paper may be applied or extended in several ways. Practically, we have shown that this modeling approach is robust to variations in climate and flu data. This may allow a model that has been trained on data from a region with high quality, consistent flu reporting and reliable weather stations to be applied to a region without those resources. Thus, predictions could be made, and potential flu burden assessed, without the need to make a bespoke model. This could be especially useful in areas where flu data is limited or has only recently begun being tracked. Additionally, given that climate data may not be as strong of a driver as assumed, at least in the United States, research may be directed to better understanding, and modeling, the various population and environmental factors that could drive seasonal flu trends. This will likely result in more accurate, longer reaching predictive models.\u003c/p\u003e \u003c/div\u003e"},{"header":"Conclusions","content":"\u003cp\u003eGiven the known limitations of the modeling approach presented here, it is difficult to draw strong conclusions about the underlying epidemiological processes of flu trends. However, our results may provide several insights for both practical modeling and biological understanding of flu patterns. First, time series-based models appear to be robust to data variation and strongly generalizable regardless of region. The use of time series models has increased in recent years, but limited work has been done comparing performance across purposely selected variable regions. Second, the climate variables presented here are useful predictors when building a model and should be considered when designing production models for flu prediction. Third, it may be that, while they are useful predictors, climate variables are not truly relevant drivers of flu trends in the real world. Further research is needed to delineate the impact of climate versus other population variables. But, given the uniform performance of this model despite variable input data, it is important to ask whether climate is as powerful a driver as assumed and, if it is not, what other factors impact yearly flu trends. This is particularly important when deciding on public health policy, predicting flu burden, and developing strategies to reduce said burden.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003eCDC \u0026ndash; Center for Disease Control and Prevention\u003c/p\u003e\n\u003cp\u003eCOVID-19 - Coronavirus Disease 2019\u003c/p\u003e\n\u003cp\u003eILI \u0026ndash; Influenza-Like-Illness\u003c/p\u003e\n\u003cp\u003eLSTM \u0026ndash; Long Short-Term Memory\u003c/p\u003e\n\u003cp\u003eMSE \u0026ndash; Mean Squared Error\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics Approval and Consent to Participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for Publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot Applicable\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of Data and Material\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll data and code described in this paper are available via GitHub or Zenodo at https://zenodo.org/doi/10.5281/zenodo.13294740.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no competing interests\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors have no funding to declare.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors\u0026apos; contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eConceptualization: A.A., A.P. and D.S.; Data curation: A.A., C.G. and J.B.; Formal analysis: A.A.; Investigation: A.A.; Methodology: A.A.; Project administration: A.A.; Software: A.A.; Supervision: A.P. and D.S.; Validation: A.A. and D.S.; Visualization: A.A.; Writing \u0026ndash; original draft: A.A., C.G., J.B., A.P. and D.S.; Writing - review \u0026amp; editing: A.A., C.G., J.B., A.P. and D.S.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003e\u003cem\u003eCDC FluView Interactive\u003c/em\u003e. 2024 [cited 2024; Available from: https://gis.cdc.gov/grasp/fluview/fluportaldashboard.html.\u003c/li\u003e\n\u003cli\u003eHoogeveen, M.J., \u003cem\u003ePollen likely seasonal factor in inhibiting flu-like epidemics. A Dutch study into the inverse relation between pollen counts, hay fever and flu-like incidence 2016\u0026ndash;2019.\u003c/em\u003e Science of The Total Environment, 2020. \u003cstrong\u003e727\u003c/strong\u003e: p. 138543.DOI: https://doi.org/10.1016/j.scitotenv.2020.138543.\u003c/li\u003e\n\u003cli\u003eMoriyama, M., W.J. Hugentobler, and A. Iwasaki, \u003cem\u003eSeasonality of Respiratory Viral Infections.\u003c/em\u003e Annu Rev Virol, 2020. \u003cstrong\u003e7\u003c/strong\u003e(1): p. 83-101.DOI: 10.1146/annurev-virology-012420-022445.\u003c/li\u003e\n\u003cli\u003eLiu, L., et al. \u003cem\u003eLSTM Recurrent Neural Networks for Influenza Trends Prediction\u003c/em\u003e. in \u003cem\u003eBioinformatics Research and Applications\u003c/em\u003e. 2018. Cham: Springer International Publishing.\u003c/li\u003e\n\u003cli\u003eHoogeveen, M.J. and E.K. Hoogeveen, \u003cem\u003eComparable seasonal pattern for COVID-19 and flu-like illnesses.\u003c/em\u003e One Health, 2021. \u003cstrong\u003e13\u003c/strong\u003e: p. 100277.DOI: 10.1016/j.onehlt.2021.100277.\u003c/li\u003e\n\u003cli\u003eBelderok, S.M., et al., \u003cem\u003eEffect of travel on influenza epidemiology.\u003c/em\u003e Emerg Infect Dis, 2013. \u003cstrong\u003e19\u003c/strong\u003e(6): p. 925-31.DOI: 10.3201/eid1906.111864.\u003c/li\u003e\n\u003cli\u003eLakdawala, S.S. and K. Subbarao, \u003cem\u003eThe Ongoing Battle Against Influenza: The challenge of flu transmission.\u003c/em\u003e Nature Medicine, 2012. \u003cstrong\u003e18\u003c/strong\u003e(10): p. 1468-1470.DOI: 10.1038/nm.2953.\u003c/li\u003e\n\u003cli\u003eKara, A., \u003cem\u003eMulti-step influenza outbreak forecasting using deep LSTM network and genetic algorithm.\u003c/em\u003e Expert Systems with Applications, 2021. \u003cstrong\u003e180\u003c/strong\u003e: p. 115153.DOI: https://doi.org/10.1016/j.eswa.2021.115153.\u003c/li\u003e\n\u003cli\u003eGers, F.A., J. Schmidhuber, and F. Cummins, \u003cem\u003eLearning to Forget: Continual Prediction with LSTM.\u003c/em\u003e Neural Computation, 2000. \u003cstrong\u003e12\u003c/strong\u003e(10): p. 2451-2471.DOI: 10.1162/089976600300015015.\u003c/li\u003e\n\u003cli\u003eHochreiter, S. and J. Schmidhuber, \u003cem\u003eLong short-term memory.\u003c/em\u003e Neural computation, 1997. \u003cstrong\u003e9\u003c/strong\u003e(8): p. 1735-1780.\u003c/li\u003e\n\u003cli\u003eSuntronwong, N., et al., \u003cem\u003eClimate factors influence seasonal influenza activity in Bangkok, Thailand.\u003c/em\u003e PLoS One, 2020. \u003cstrong\u003e15\u003c/strong\u003e(9): p. e0239729.DOI: 10.1371/journal.pone.0239729.\u003c/li\u003e\n\u003cli\u003eSoebiyanto, R.P., et al., \u003cem\u003eThe Role of Temperature and Humidity on Seasonal Influenza in Tropical Areas: Guatemala, El Salvador and Panama, 2008\u0026ndash;2013.\u003c/em\u003e PLOS ONE, 2014. \u003cstrong\u003e9\u003c/strong\u003e(6): p. e100659.DOI: 10.1371/journal.pone.0100659.\u003c/li\u003e\n\u003cli\u003eAmin, S., et al., \u003cem\u003eDetecting Dengue/Flu Infections Based on Tweets Using LSTM and Word Embedding.\u003c/em\u003e IEEE Access, 2020. \u003cstrong\u003e8\u003c/strong\u003e: p. 189054-189068.DOI: 10.1109/ACCESS.2020.3031174.\u003c/li\u003e\n\u003cli\u003eZhao, Z., et al., \u003cem\u003eStudy on the prediction effect of a combined model of SARIMA and LSTM based on SSA for influenza in Shanxi Province, China.\u003c/em\u003e BMC Infectious Diseases, 2023. \u003cstrong\u003e23\u003c/strong\u003e(1): p. 71.DOI: 10.1186/s12879-023-08025-1.\u003c/li\u003e\n\u003cli\u003eZhu, H., et al., \u003cem\u003eStudy on the influence of meteorological factors on influenza in different regions and predictions based on an LSTM algorithm.\u003c/em\u003e BMC Public Health, 2022. \u003cstrong\u003e22\u003c/strong\u003e(1): p. 2335.DOI: 10.1186/s12889-022-14299-y.\u003c/li\u003e\n\u003cli\u003e\u003cem\u003eVisual Crossing Weather (2010 - 2023)\u003c/em\u003e. 2024; Available from: https://www.visualcrossing.com/.\u003c/li\u003e\n\u003cli\u003eAmendolara, A.B., et al., \u003cem\u003eLSTM-based recurrent neural network provides effective short term flu forecasting.\u003c/em\u003e BMC Public Health, 2023. \u003cstrong\u003e23\u003c/strong\u003e(1): p. 1788.DOI: 10.1186/s12889-023-16720-6.\u003c/li\u003e\n\u003cli\u003eDave, K. and P.C. Lee, \u003cem\u003eGlobal Geographical and Temporal Patterns of Seasonal Influenza and Associated Climatic Factors.\u003c/em\u003e Epidemiologic Reviews, 2019. \u003cstrong\u003e41\u003c/strong\u003e(1): p. 51-68.DOI: 10.1093/epirev/mxz008.\u003c/li\u003e\n\u003cli\u003e\u003cem\u003eHawaii\u0026apos;s Influenza Surveillance Program\u003c/em\u003e. 2024 August 1, 2024]; Available from: https://health.hawaii.gov/docd/about-us/programs/hawaiis-influenza-surveillance-program/#:~:text=The%20seasonal%20flu%20is%20a,we%20experience%20flu%20year%20round.\u003c/li\u003e\n\u003cli\u003eFerence, R.S., J.A. Leonard, and H.D. Stupak, \u003cem\u003ePhysiologic Model for Seasonal Patterns in Flu Transmission.\u003c/em\u003e The Laryngoscope, 2020. \u003cstrong\u003e130\u003c/strong\u003e(2): p. 309-313.DOI: https://doi.org/10.1002/lary.27910.\u003c/li\u003e\n\u003cli\u003ePaz\u0026ndash;Bailey, G., et al., \u003cem\u003eRecent influenza activity in tropical Puerto Rico has become synchronized with mainland US.\u003c/em\u003e Influenza and Other Respiratory Viruses, 2020. \u003cstrong\u003e14\u003c/strong\u003e(5).DOI: https://doi.org/10.1111/irv.12744.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"modeling, influenza, LSTM, neural network","lastPublishedDoi":"10.21203/rs.3.rs-4896641/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4896641/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003eInfluenza virus is responsible for a recurrent, yearly epidemic in most temperate regions of the world. Flu has been responsible for a high disease burden in recent years, despite the confounding presence of SARS-CoV-2. However, the mechanisms behind seasonal variance in flu burden are not well understood. This study seeks to expand understanding of the impact of variable climate regions on seasonal flu trends. To that end, three climate regions have been selected. Each region represents a different ecological zone and provides different weather patterns.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eA Long short-term memory (LSTM)-based recurrent neural network was used to predict influenza-like-illness trends for three separate locations: Hawaii, Vermont, and Nevada. Flu data were gathered from the Center for Disease Control as weekly influenza-like-illness (ILI) percentages. Weather data were collected from Visual Crossing and included temperature, wind speed, UV index, solar radiation, precipitation, and humidity. Data were prepared and the model was trained as described previously.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eAll three regions showed strong seasonality of flu trends with Hawaii having the largest absolute ILI values. Temperature showed a moderate negative correlation with ILI in all three regions (Vermont = -54, Nevada = -0.56, Hawaii = -0.44). Humidity was moderately correlated in Nevada (0.47) and weakly correlated with ILI in Hawaii (0.22). Vermont ILI did not correlate with humidity. Precipitation and wind speed were weakly correlated in all three regions. Solar radiation and UV index showed moderate correlation in Vermont (-0.33, -0.36) and Nevada (-0.5263, -0.55), but only a weak correlation in Hawaii (-0.15, -0.18). When trained on the complete data sets, baseline model performances for all three datasets at +\u0026thinsp;1 week were equivalent. Models trained on one region and used to predict cross-regional data performed uniformly and equivalent to baseline.\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e \u003cp\u003eResults indicate that climate variables were weak to moderate predictors in all regions. Initial modeling attempts revealed acceptable and uniform performance in all regions. When cross-regional predictions were made, performance remained uniform across all regions, implying that climate patterns may be more important than absolute climate values. Additionally, this data suggests that climate may not be as influential on flu trends as population-level human factors.\u003c/p\u003e","manuscriptTitle":"LSTM-based Recurrent Neural Network Predicts Influenza-like-illness in Variable Climate Zones","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-09-09 06:52:05","doi":"10.21203/rs.3.rs-4896641/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"ee348a3f-49d0-4668-bf93-604601f39efc","owner":[],"postedDate":"September 9th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-05-14T08:53:53+00:00","versionOfRecord":[],"versionCreatedAt":"2024-09-09 06:52:05","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-4896641","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4896641","identity":"rs-4896641","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}