Background: Significant progress has been made in the fight against P. falciparum malaria due in part to the widespread adoption of rapid diagnostic tests (RDTs) that detect histidine-rich protein 2 (PfHRP2) and its paralog PfHRP3 encoded by the pfhrp2 and pfhrp3 genes, respectively. Parasites without pfhrp2 and pfhrp3 genes are not detected by these RDTs. Pfhrp3 loss appears to be more common in some geographical regions and has been observed in vitro. We sought to gain insight into geographic patterns of pfhrp3 deletion and define the mechanism of gene loss. Methods: Over 9,830 publicly available, whole-genome sequenced (WGS) P. falciparum field samples were analyzed for genotypes and coverage using PathWeaver for local assembly . DNA from two cultured isolates with pfhrp3 deletion, HB3 and SD01, were sequenced with Oxford Nanopore Technologies (ONT) long-read sequencing, assembled with Canu and Flye, and genes annotated with Companion. Results: Two distinct pfhrp3 deletion patterns were detected: 1) segmental deletion of chromosome 13 just centromeric to pfhrp3 extending to the end of the chromosome, with co-occurring segmental duplication of the chromosome 11 subtelomeric region, and 2) segmental deletion of chromosome 13 starting at various locations centromeric to pfhrp3 without chromosome 11 duplication. Pattern 1 was almost exclusively found in samples from Africa and South America, while pattern 2 was observed predominantly in Southeast Asia. The pattern 1 boundary fell within a 15kb nearly identical duplication on both chromosomes containing ribosomal genes. ONT assembly of HB3 and SD01 parasite lines revealed hybrid chromosomes and long-reads spanning the ribosomal duplication, consistent with recombination between non-homologous chromosomes. Conclusion: Our findings demonstrate duplication-mediated non-homologous recombination creating a hybrid 13-11 chromosome that replaces pfhrp3 and telomeric chromosome 13 with a translocated telomeric chromosome 11 sequence--essentially yielding a deletion of chromosome 13 sequence and interchromosomal duplication of chromosome 11 sequence. Given that existing ribosomal duplications likely predispose to the frequent occurrence of this translocation during meiosis, it suggests that subsequent selective forces are driving its presence or absence in different geographical regions. This mechanism appears to explain pfhrp3 deletion in South America and Africa and may explain why pfhrp3 deletions are more prevalent than pfhrp2 deletions in many localities. However, the forces driving emergence of pfhrp3- parasites may be complex and encompass other genes affected by the recombination event. Further studies of the origins of pfhrp2- and pfhrp3-deleted strains and the selective pressures suppressing their occurrence or driving their expansion are needed.