Forecasting air quality time series using deep learning

This paper presents one of the first applications of deep learning (DL) techniques to predict air pollution time series. Air quality management relies extensively on time series data captured at air monitoring stations as the basis of identifying population exposure to airborne pollutants and determining compliance with local ambient air standards. In this paper, 8 hr averaged surface ozone (O₃) concentrations were predicted using deep learning consisting of a recurrent neural network (RNN) with long short-term memory (LSTM). Hourly air quality and meteorological data were used to train and forecast values up to 72 hours with low error rates. The LSTM was able to forecast the duration of continuous O₃ exceedances as well. Prior to training the network, the dataset was reviewed for missing data and outliers. Missing data were imputed using a novel technique that averaged gaps less than eight time steps with incremental steps based on first-order differences of neighboring time periods. Data were then used to train decision trees to evaluate input feature importance over different time prediction horizons. The number of features used to train the LSTM model was reduced from 25 features to 5 features, resulting in improved accuracy as measured by Mean Absolute Error (MAE). Parameter sensitivity analysis identified look-back nodes associated with the RNN proved to be a significant source of error if not aligned with the prediction horizon. Overall, MAE's less than 2 were calculated for predictions out to 72 hours.

Implications: Novel deep learning techniques were used to train an 8-hour averaged ozone forecast model. Missing data and outliers within the captured data set were replaced using a new imputation method that generated calculated values closer to the expected value based on the time and season. Decision trees were used to identify input variables with the greatest importance. The methods presented in this paper allow air managers to forecast long range air pollution concentration while only monitoring key parameters and without transforming the data set in its entirety, thus allowing real time inputs and continuous prediction.     

Incorporating geodiversity into conservation decisions

In a rapidly changing climate, conservation practitioners could better use geodiversity in a broad range of conservation decisions. We explored selected avenues through which this integration might improve decision making and organized them within the adaptive management cycle of assessment, planning, implementation, and monitoring. Geodiversity is seldom referenced in predominant environmental law and policy. With most natural resource agencies mandated to conserve certain categories of species, agency personnel are challenged to find ways to practically implement new directives aimed at coping with climate change while retaining their species‐centered mandate. Ecoregions and ecological classifications provide clear mechanisms to consider geodiversity in plans or decisions, the inclusion of which will help foster the resilience of conservation to climate change. Methods for biodiversity assessment, such as gap analysis, climate change vulnerability analysis, and ecological process modeling, can readily accommodate inclusion of a geophysical component. We adapted others’ approaches for characterizing landscapes along a continuum of climate change vulnerability for the biota they support from resistant, to resilient, to susceptible, and to sensitive and then summarized options for integrating geodiversity into planning in each landscape type. In landscapes that are relatively resistant to climate change, options exist to fully represent geodiversity while ensuring that dynamic ecological processes can change over time. In more susceptible landscapes, strategies aiming to maintain or restore ecosystem resilience and connectivity are paramount. Implementing actions on the ground requires understanding of geophysical constraints on species and an increasingly nimble approach to establishing management and restoration goals. Because decisions that are implemented today will be revisited and amended into the future, increasingly sophisticated forms of monitoring and adaptation will be required to ensure that conservation efforts fully consider the value of geodiversity for supporting biodiversity in the face of a changing climate.     

The promise of acceptance as an NGO security management approach

This paper explores three questions related to acceptance as a security management approach. Acceptance draws upon relationships with community members, authorities, belligerents and other stakeholders to provide consent for the presence and activities of a non‐governmental organisation (NGO), thereby reducing threats from these actors. Little is documented about how NGOs gain and maintain acceptance, how they assess and monitor the presence and degree of acceptance, or how they determine whether acceptance is effective in a particular context. Based on field research conducted in April 2011 in Kenya, South Sudan and Uganda, we address each of these three issues and argue that acceptance must be actively sought as both a programme and a security management strategy. In the paper we delineate elements common to all three contexts as well as missed opportunities, which identify areas that NGOs can and should address as part of an acceptance approach.     

URBANIZATION INFLUENCES ON AQUATIC COMMUNITIES IN NORTHEASTERN ILLINOIS STREAMS1

ABSTRACT: Biotic indices and sediment trace element concentrations for 43 streams in northeastern Illinois (Chicago area) from the 1980s and 1990s were examined along an agricultural to urban land cover gradient to explore the relations among biotic integrity, sediment chemistry, and urbanization. The Illinois fish Alternative Index of Biotic Integrity (AIBI) ranged from poor to excellent in agricultural/rural streams, but streams with more than 10 percent watershed urban land (about 500 people/mi²) had fair or poor index scores. A macroinvertebrate index (MBI) showed similar trends. A qualitative habitat index (PIBI) did not correlate to either urban indicator. The AIBI and MBI correlated with urban associated sediment trace element concentrations. Elevated copper concentrations in sediment occurred in streams with greater than 40 percent watershed urban land. The number of intolerant fish species and modified index of biotic integrity scores increased in some rural, urbanizing, and urban streams from the 1980s to 1990s, with the largest increases occurring in rural streams with loamy/sandy surficial deposits. However, smaller increases also occurred in urban streams with clayey surficial deposits and over 50 percent watershed urban land. These data illustrate the potentially complex spatial and temporal relations among biotic integrity, sediment chemistry, watershed urban land, population density, and regional and local geologic setting.