March 27, 2024

Mr. Mike Rolband

Director of the Virginia Department of Environmental Quality

RE:  VRRM Version 4.1

NOTE: As a member of the SAG, I originally submitted these comments on August 17, 2023 as part of the informal comment period. However, I later became aware that the comments were never registered or considered by DEQ. Therefore, I am resubmitting, as I believe these comments are still relevant for VRRM Version 4.1.

I appreciate the opportunity to comment on VRRM Version 4.1, as well as supporting documents that have been provided.  It is an important step for DEQ and the Commonwealth to provide this updated analysis in addition to the extensive work being done on the manuals and regulations.  I have provided several comments below, accompanied by suggestions for addressing each comment category.

The VRRM is a Compliance Tool

As a member of the team that developed the original VRRM, the good work conducted for this update reminded me of conversations (and more than a few debates) that occcured at the time.  Many design charettes were held around the state to largely skeptical audiences of engineers and other technical professionals. Those of us involved were careful to describe the VRRM as a “compliance tool,” and not “a model.”

Both compliance tools and models – in their best sense -- are based on a good understanding of the state of the science. In addition, a compliance tool sends certain signals to the development community.  If a design engineer selects design approach X versus Y, then compiance is either easier or harder. These signals are very important to consider when developing such a tool in order to achieve desired outcomes: less runoff, more treatment, less soil compaction, more vegetation, etc. and, importantly, to avoid unintended outcomes.  In addition, a compliance tool should be crafted to be usable and understandable for every regulated development/redevelopment site across the Commonwealth.

Those of us in the Bay Watershed now live in a compliance world governed by the CBWM, so the line between a model and compliance tool have been blurred.  However, the VRRM will still function in essence as a compliance tool. If adding impervious cover vis-à-vis turf leads to a better compliance outcome for the designer, then that is what will happen at site after site.  While it is accurate that recent data indicate that direct phosphorus wash-off loads from impervious are lower than documented in older studies (e.g. NURP), adding more impervious will simply exacerbate the downstream watershed and stream channel issues that are very well documented in the Impervious Cover Model and other studies, thus undermining key objectives of the Virginia Stormwater Management Program.  

Recommendation: In addition to the helpful modelled scenario runs presented in the VRRM Version 4.1 cover memo, allow a period of provisional testing at real development sites to evaluate outcomes and any unintended consquences (e.g., more high impervious cover sites than Version 3.0).  Adjustments can be made while still maintaining consistency with CAST.

Continue to Conduct a VRRM Update

The VRRM integrates (1) loading rates and reductions necessary to reach site-based load targets, (2) runoff coefficient values, and (3) runoff reduction, pollutant removal, and mass load removals for various BMPs. All three of the components listed above should be based on a thorough anlaysis of the current state of the science.

The VRRM Version 4.1 update addresses #1 and some of #2, as long as CAST represents the state of the science (although all would acknowledge that the CBWM also includes a measure of approximation and stakeholder compromise).  As for Rv values, interesting research on the affects of soil compaction is being conducted at the Center for Urban & Environmental Research at UMBC, Penn State, and other institutions in addition to work in Arlington County concerning on-lot runoff (mentioned below).  This research is leading to the conclusion that compacted soil in urban environments is a source of “hidden impervious cover.”  While the research in its current state may not be ready for regulatory adoption, it certainly represents an updated understanding of the state of the science.  It is interesting to note that the increased Rv/CNs values for Managed Turf in the VRRM 4.1 support this finding (tempered by the fact that the P-ban assumption reduces the 4.1 Managed Turf loading rates from 4X to 1.4X compared to Version 3.0).

As for BMP runoff reduction (RR) and pollutant removal (PR) efficiences, the updated VRRM does not change these.  It must be said that the original rates were based on research conducted up through, at the latest, 2007, and much additional BMP research has been done since that time.  More recent studies have been conducted that ties design parameters for certain BMPs to RR and PR rates.  I led teams for two of these efforts that used updated research through 2017 (see attached reports from Metro Nashville and DOEE). And, indeed, there has been additional research since that time. 

Not including a more robust effort to assemble the state of the science on all components of the VRRM would be a missed opportunity.

Recommendation: Refrain from referring to the current effort as “updating” the VRRM but rather integration with CAST/CBWM with further work needed for a true update.  In the near future, mount an effort to truly update the state of science for each VRRM component.    

New Development & Redevelopment Land Is Different Than General Watershed-Wide Land Covers

The integration with CAST and the CBWM to derive new loading rates (and P targets) is justified.  However, the VRRM was developed for very specific categories of land use: new development and redevelopment subject to the VSMP regulations.  These land uses do not behave like general watershed land covers.  Development sites are often mass-graded, topsoil stripped, and the soil compacted in many areas.  In addition, stabilization techniques like hydroseeding use higher fertilizer rates, which are needed to establish new vegetation.  The upshot is that new development and redevelopment sites are likely to have higher runoff volumes (e.g., due to compaction and mass grading) and nutrient loads. 

One could argue that all urban land cover categories in CAST have undergone similar site grading at some point in time.  However, a recent study in Arlington County found that newly-developed lots are 10X more compacted than older lots.  The County now requires soil profile rebuilding in order to address the excessive runoff generated by new development.

Recommendation: A literature search of runoff volumes and nutrient loading from the post-construction condition at development sites may yield certain adjustments for runoff and loading.

Mixed Open Land Cover

Other commentors (e.g., HRPDC) have better insights on the MO category, and I defer to those recommendations.  It seems methodology-wise, there was rigor applied to to developing MO loading rates and Rv values.  However, my concern rests with how this will be interpreted in the real world of plan review and project construction.  There will certainly be an incentive for designers to use MO in lieu of Managed Turf, so verification during construction and through time will be essential.

In addition, I question I inclusion of MO in the Drainage Area tabs of the spreadsheet.  With the original VRRM, Forest/Open Space was not included in the Drainage Area tab because it seemed misguided to have development sites build BMPs to treat this type of land cover (or, worse, cut down trees to build BMPs to treat the remaining forest).  I question whether the same logic should apply to the new MO category, if it truly is meant to remain undisturbed.  Our emerging understanding of forests, undisturbed open space areas, and the like are that they are BMPs unto themselves, and the VRRM intentionally provides incentives to preserve as many of these features as possible from the pre-development condition.  In this regard, Clearinghouse BMPs should still focus on treating runoff from impervious and managed turf land covers.  

An additional recommendion is, if the MO category remains in the VRRM, use it instead of Forest for the surface area of qualifying BMPs (those that are vegetated and don’t replace an otherwise impervious surface).  Version 3.0 allows these surface areas to be classified at Forest/Open Space, but the MO designation may be more applicable for these structural BMPs.

Recommendation: As suggested by HRPDC, provide clarity on MO conditions, consistent application, and verification through deeds or easements, as currently done for Forest/Open Space.

Overall

The consistent themes of these comments is to base VRRM updates on the thorough and updated analysis of the state of science that underpins the loading rates, Rv values, and BMP mass load removal rates.  In addition, review the modified VRRM as it applies to results at actual development sites to identify and rectify unanticipated real-world outcomes.  

Once again, thank you for this opportunity to submit comments on the VRRM.

Sincerely,

David J. Hirschman, Principal, Hirschman Water & Environment, LLC