1: \begin{abstract}

2: In this paper we study a Tikhonov-type method for ill-posed 

3: nonlinear operator equations $\gdag = F\left(\udag\right)$ where 

4: $\gdag$ is an integrable, non-negative function. We assume that

5: data are drawn from a Poisson process with density $t\gdag$ where

6: $t>0$ may be interpreted as an exposure time. Such problems occur

7: in many photonic imaging applications including positron emission tomography,

8: confocal fluorescence microscopy, astronomic observations, and 

9: phase retrieval problems in optics. Our approach uses 

10: a Kullback-Leibler-type data fidelity functional and allows for general 

11: convex penalty terms. We prove convergence rates of the expectation of

12: the reconstruction error under a variational source 

13: condition as $t\to\infty$ both for an a priori and for a 

14: Lepski{\u\i}-type parameter choice rule. 

15: \end{abstract}

16: