BEMC in HF Pico DSTs for 2014

In the slides attached labeled RReedHFPico_06252015.pdf, I post my concerns with a quick look at the BEMC in the heavy flavor Pico DSTs.

The main point can be seen in these plots:
Tower Energy HF Pico Dsts 2014Tower Eta Phi HF Pico Dsts 2014
The triggered in this case is HT18.  There are clearly hot hot towers, plus this strange structure around ETower = 12 GeV.  There is also the issue at eta ~ 0.  The towers that cause the spike at 32 GeV are actually coming from very few towers, but they have a lot of counts.  These are not always overlapping with towers that show up red on the right.

The relevant code is:

unsigned int nepid = mPicoDst->numberOfEmcPidTraits();

for (int i = 0;i<nepid;i++){ //with check that btowerId > 0 and that emctraits exists

PicoEmcPidTraits* emctraits = mPicoDst->emcPidTraits(i);

he0->Fill(emctraits->e0());}

hTowerId->Fill(emctraits->btowId());

I got the h-f position from:

StEmcGeom *geomBemc = StEmcGeom::getEmcGeom(1);

geomBemc->getEta(emctraits->btowId(),eta);

geomBemc->getPhi(emctraits->btowId(),phi);

Is there a calibration maker I am missing?