Dungeness Spit: Shockingly stable, but also maybe shrinking?

Every year since 2013 I've been able to get out and collect shoreline profile data on Dungeness Spit, along roughly 35 cross-shore oriented transect lines spread out evenly along the length of the spit:

Dungeness Spit is an awesome place to study the dynamics of spit morphology - its long and largely un-modified, with un-modified bluffs to the west that presumably supply sediment for its on-going growth.  And grow it has.  Schwartz et al. (1987) estimated an average annual rate of progradation of the spit of 4.4. m/yr, based on an analysis of 4 data-sets (either survey maps or aerial photos) collected between 1855 and 1985.  I walked into this effort thinking that what I would learn about spit progradation would be how much that average rate of positive growth varied annually.  Indeed, after my first year of surveying (2013) I could see that the end of the spit grew by roughly 5 meters or so based on a comparison to an aerial LiDAR dataset collected in 2012:

Intertidal profile data along a transect bisecting the very tip of Dungeness Spit for 2012, 2013 and 2014.  2012 data are from aerial LiDAR, 2013 and 2014 are from topographic GNSS surveys.

 Exactly as expected.  Since then, though...things have been a little different, with an average annualized erosion rate on the end of the spit of roughly 8 meters/yr.  That is ~25 feet per year of erosion on average:

Same location and data as are shown above, but for 2013-2020, all GNSS surveys.

  Instead of finding myself analyzing the annual variability in spit growth, I'm trying to interpret the annual variability in erosion...and it does vary.  Some years there is more...and some years virtually none.  But the trend here is unmistakable, and fascinating, and vexing.  What is going on? 

Its worth noting that this erosion at the very end of the spit is NOT representative of the spit as a whole.  Largely the rest of the spit, especially the long skinny "strand" connected to land, is astonishingly stable.  I started out this study hypothesizing that I might see a very slow rate of migration of this part of the spit, associated with storm-driven overwash.   But I don't.  Year after year the beach and crest are in an almost identical position despite being regularly battered in the winter:

GNSS survey-based profile data along a transect at roughly mile 2.5 on Dungeness Spit. 

  So what is going on?  I'm not totally sure, and in fact I don't really have too many ideas that I can even use to formulate a proposed mechanism.  And there don't appear to be too many analogous observations that I can find from the limited number of papers out there describing spit dynamics.  There is at least one exception, though - this paper describes a spit on the shoreline of the North Sea that, based on sedimentary evidence, has grown in fits and starts, with periods of growth alternating with periods of erosion.  In a conversation with Maury Schwartz some years ago, he described to me a model of spit growth that required that the spit build out a sub-tidal platform before it could grow sub-aerially...which would manifest in a data set like mine as periods of no- or low-growth punctuated with periods of rapid growth.  That's not really what I see in my data...but Maury thought hard about these things and I'm still trying to fit my observations into that model.

Any other ideas out there?