Comment on 'Energy and air emission implications of a decentralized wastewater system'

The recent article by Shehabi et al (2012) compared a decentralized and a centralized system on the basis of energy use, greenhouse gas emissions and air pollutants, and argued that economies of scale lower the environmental impacts from a centralized system on a per-volume basis. The authors' analysis builds on a similar effort by Muga and Mihelcic (2008) that used a combination of economic, environmental and societal factors to evaluate the effectiveness of various wastewater treatment systems. We found the comparison between a small decentralized system (0.015 MGD serving 100 residents) and a large centralized system (66.5 MGD serving 500 000 residents) unconventional and inappropriate for at least a few reasons.

First, the centralized treatment plant chosen was too large not only in comparison to the decentralized plant, but also in comparison to other centralized plants. Muga and Mihelcic (2008) reviewed 16 255 wastewater treatment plants in the US and observed that over 80% of them have a capacity less than 5 MGD. In the state of New York, only 16 of the over 600 municipal treatment plants have capacities greater than 65 MGD. While these 16 plants constitute 62% of the installed capacity and serve 64% of the residents in the state, a disproportionate majority of plants are much smaller in size. 87% of the plants are smaller than 5 MGD, 69% are smaller than 1 MGD and 28% are smaller than 0.1 MGD¹. This trend may be evident in other states as well, in some cases more pronounced depending on proportion of rural/urban populations. Second, it would take 4600 small decentralized systems, such as the one described in Shehabi et al (2012), to make up a 66.5 MGD centralized treatment plant—an idea no city would entertain in its planning formulation. Similarly, a large centralized plant would be an inappropriate and highly expensive solution for a rural community. Large treatment plants achieve economies of scale in operational parameters such as energy and emissions, and are able to leverage rapid changes in technology, which smaller decentralized systems are unable to do. Indeed, size plays a role even within centralized treatment plants. The effluent violation rates for some parameters at small plants (0.01 MGD) are 10 times that of plants that are sized 100 MGD or larger (Weirich et al 2011). The energy use per volume is higher in centralized plants sized 0.1 MGD or smaller, as compared to those sized up to 1 MGD and 5 MGD (Muga and Mihelcic 2008).

Many communities are not large and dense enough to warrant large treatment plants. A comparison between the two methods of treatment is only fair if similar-sized plants are considered. The inclusion of alternative scenarios with a large plant for a rural area or many small plants for an urban area can greatly enhance the validity of the model presented by Shehabi et al (2012). Based on the data from New York, there are 34 plants that have capacities of 0.015 MGD or smaller serving 102 residents on average, nearly the same as that in case of the decentralized system described in Shehabi et al. Such a comparison is more likely to yield meaningful results and policy-worthy conclusions. While GHG emissions and particulate pollution are critical factors in evaluating the efficiency of wastewater treatment systems, very rarely do decisions about construction of new treatment plants get made on those factors. The lack of federal and state funds and an ageing infrastructure have forced many small communities to look toward decentralized solutions for managing wastewater. It is very likely that decentralized treatment systems will have a resurgence in small communities, this time in a planned and well-designed manner. Therefore, it is imperative that a fair comparison between the centralized and decentralized approaches be conducted to allow decision-makers and stakeholders to make informed choices.

We thank Jerry Shen for interpreting the statewide data for New York on municipal treatment plants.